Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

Category articles

General General

ECG-based machine-learning algorithms for heartbeat classification.

In Scientific reports ; h5-index 158.0

Electrocardiogram (ECG) signals represent the electrical activity of the human hearts and consist of several waveforms (P, QRS, and T). The duration and shape of each waveform and the distances between different peaks are used to diagnose heart diseases. In this work, to better analyze ECG signals, a new algorithm that exploits two-event related moving-averages (TERMA) and fractional-Fourier-transform (FrFT) algorithms is proposed. The TERMA algorithm specifies certain areas of interest to locate desired peak, while the FrFT rotates ECG signals in the time-frequency plane to manifest the locations of various peaks. The proposed algorithm's performance outperforms state-of-the-art algorithms. Moreover, to automatically classify heart disease, estimated peaks, durations between different peaks, and other ECG signal features were used to train a machine-learning model. Most of the available studies uses the MIT-BIH database (only 48 patients). However, in this work, the recently reported Shaoxing People's Hospital (SPH) database, which consists of more than 10,000 patients, was used to train the proposed machine-learning model, which is more realistic for classification. The cross-database training and testing with promising results is the uniqueness of our proposed machine-learning model.

Aziz Saira, Ahmed Sajid, Alouini Mohamed-Slim

2021-Sep-21

General General

Universal activation function for machine learning.

In Scientific reports ; h5-index 158.0

This article proposes a universal activation function (UAF) that achieves near optimal performance in quantification, classification, and reinforcement learning (RL) problems. For any given problem, the gradient descent algorithms are able to evolve the UAF to a suitable activation function by tuning the UAF's parameters. For the CIFAR-10 classification using the VGG-8 neural network, the UAF converges to the Mish like activation function, which has near optimal performance [Formula: see text] when compared to other activation functions. In the graph convolutional neural network on the CORA dataset, the UAF evolves to the identity function and obtains [Formula: see text]. For the quantification of simulated 9-gas mixtures in 30 dB signal-to-noise ratio (SNR) environments, the UAF converges to the identity function, which has near optimal root mean square error of [Formula: see text]. In the ZINC molecular solubility quantification using graph neural networks, the UAF morphs to a LeakyReLU/Sigmoid hybrid and achieves RMSE=[Formula: see text]. For the BipedalWalker-v2 RL dataset, the UAF achieves the 250 reward in [Formula: see text] epochs with a brand new activation function, which gives the fastest convergence rate among the activation functions.

Yuen Brosnan, Hoang Minh Tu, Dong Xiaodai, Lu Tao

2021-Sep-21

General General

A novel transcriptomic-based classifier for senescent cancer cells.

In Trends in cancer

The inability to unequivocally identify senescent cancer cells is hindering the development of novel anticancer therapies using senolytic compounds. In a recent study in Cell Reports, Jochems et al. used a machine-learning approach to generate a classifier for senescent cancer cells based on transcriptional signatures.

Varughese Feba Mariam, Demaria Marco

2021-Sep-18

General General

Predicting Survived Events in Nontraumatic Out-of-Hospital Cardiac Arrest: A Comparison Study on Machine Learning and Regression Models.

In The Journal of emergency medicine

BACKGROUND : Prediction of early outcomes of nontraumatic out-of-hospital cardiac arrest (OHCA) by emergency physicians is inaccurate.

OBJECTIVE : Our aim was to develop and validate practical machine learning (ML)-based models to predict early outcomes of nontraumatic OHCA for use in the emergency department (ED). We compared their discrimination and calibration performances with the traditional logistic regression (LR) approach.

METHODS : Between October 1, 2017 and March 31, 2020, prehospital resuscitation was performed on 17,166 OHCA patients. There were 8157 patients 18 years or older with nontraumatic OHCA who received continued resuscitation in the ED included for analysis. Eleven demographic and resuscitation predictor variables were extracted to predict survived events, defined as any sustained return of spontaneous circulation until in-hospital transfer of care. Prediction models based on random forest (RF), multilayer perceptron (MLP), and LR were created with hyperparameter optimization. Model performances on internal and external validation were compared using discrimination and calibration statistics.

RESULTS : The three models showed similar discrimination performances with c-statistics values of 0.712 (95% confidence interval [CI] 0.711-0.713) for LR, 0.714 (95% CI 0.712-0.717) for RF, and 0.712 (95% CI 0.710-0.713) for MLP models on external validation. For calibration, MLP model had a better performance (slope of calibration regression line = 1.10, intercept = -0.09) than LR (slope = 1.17, intercept = -0.11) and RF (slope = 1.16, intercept= -0.10).

CONCLUSIONS : Two practical ML-based and one regression-based clinical prediction models of nontraumatic OHCA for survived events were developed and validated. The ML-based models did not outperform LR in discrimination, but the MLP model showed a better calibration performance.

Lo Yat Hei, Siu Yuet Chung Axel

2021-Sep-18

clinical prediction model, machine learning, out-of-hospital cardiac arrest, prognosis, resuscitation

Public Health Public Health

The State of Mind of Healthcare Professionals in the Light of the COVID-19: Insights from Text Analysis of Twitter's Online Discourses.

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : The COVID-19 pandemic has affected populations worldwide, with extreme health, economic, social, and political implications. Healthcare professionals (HCPs) are at the core of pandemic response and are among the most crucial factors in maintaining coping capacities. Yet, they are also vulnerable to mental health effects, managing a long-lasting emergency under lack of resources and complicated personal concerns. However, there is a lack of longitudinal studies that investigate the HCP population.

OBJECTIVE : To analyse the state of mind of HCPs as expressed in online discussions published on Twitter in light of COVID-19, from the pandemic onset until the end of 2020.

METHODS : The population for this study was selected from followers of a few hundred Twitter accounts of healthcare organizations and common HCP points of interest. We used active learning, a process that iteratively uses machine learning and manual data labeling, to select the large-scale population of Twitter accounts maintained by English-speaking HCPs focusing on individuals rather than official organizations. We analyzed the topics and emotions in their discourse during 2020. The topic distributions were obtained using the Latent Dirichlet Allocation (LDA) algorithm. We defined a measure of topic cohesion and described the most cohesive topics. The emotions expressed in tweets during 2020 were compared to 2019. Finally, the emotion intensities were cross-correlated with the pandemic waves to explore possible associations between the pandemic development and emotional response.

RESULTS : We analyzed timelines of 53,063 Twitter profiles, 90% of which are maintained by individual HCPs. Professional topics account for 44.5% of tweets by HCPs from Jan. 1st to Dec. 6th, 2020. Events such as the pandemic waves, U.S. elections, or the George Floyd case affect the HCPs' discourse. The levels of joy and sadness exceed their minimal and maximal values yesteryear, respectively, 80% of the time, P= .001. Most interestingly, fear precedes the pandemic waves (in terms of the differences in confirmed cases) by two weeks with a Spearman correlations coefficient of ρ(47)= .34, P= .026.

CONCLUSIONS : Analyses of longitudinal data over the 2020 year reveal that a large fraction of HCP discourse is related directly to professional content, including the increase in the volume of discussions following the pandemic waves. The changes in emotional patterns (decrease of joy, an increase of sadness, fear, and disgust) during the year 2020 may indicate the utmost importance in providing emotional support for HCPs to prevent fatigue, burnout, and mental health disorders postpandemic period. The increase of fear two weeks in advance of pandemic waves indicates that HCPs are in a position and with adequate qualification to anticipate the pandemic development, and could serve as a bottom-up pathway for expressing the morbidity and clinical situation to health agencies.

CLINICALTRIAL :

Elyashar Aviad, Plochotnikov Ilia, Cohen Idan-Chaim, Puzis Rami, Cohen Odeya

2021-Jul-23

General General

Performance of deep convolutional neural network for classification and detection of oral potentially malignant disorders in photographic images.

In International journal of oral and maxillofacial surgery ; h5-index 38.0

Oral potentially malignant disorders (OPMDs) are a group of conditions that can transform into oral cancer. The purpose of this study was to evaluate convolutional neural network (CNN) algorithms to classify and detect OPMDs in oral photographs. In this study, 600 oral photograph images were collected retrospectively and grouped into 300 images of OPMDs and 300 images of normal oral mucosa. CNN-based classification models were created using DenseNet-121 and ResNet-50. The detection models were created using Faster R-CNN and YOLOv4. The image data were randomly selected and assigned as training, validating, and testing data. The testing data were evaluated to compare the performance of the CNN models with the diagnosis results produced by oral and maxillofacial surgeons. DenseNet-121 and ResNet-50 were found to produce high efficiency in diagnosis of OPMDs, with an area under the receiver operating characteristic curve (AUC) of 95%. Faster R-CNN yielded the highest detection performance, with an AUC of 74.34%. For the CNN-based classification model, the sensitivity and specificity were 100% and 90%, respectively. For the oral and maxillofacial surgeons, these values were 91.73% and 92.27%, respectively. In conclusion, the DenseNet-121, ResNet-50 and Faster R-CNN models have potential for the classification and detection of OPMDs in oral photographs.

Warin K, Limprasert W, Suebnukarn S, Jinaporntham S, Jantana P

2021-Sep-18

artificial intelligence, deep learning, neural network models, oral neoplasms, precancerous conditions

General General

The Efficiency of U.S. Public Space Utilization During the COVID-19 Pandemic.

In Risk analysis : an official publication of the Society for Risk Analysis

The COVID-19 pandemic has called for and generated massive novel government regulations to increase social distancing for the purpose of reducing disease transmission. A number of studies have attempted to guide and measure the effectiveness of these policies, but there has been less focus on the overall efficiency of these policies. Efficient social distancing requires implementing stricter restrictions during periods of high viral prevalence and rationing social contact to disproportionately preserve gatherings that produce a good ratio of benefits to transmission risk. To evaluate whether U.S. social distancing policy actually produced an efficient social distancing regime, we tracked consumer preferences for, visits to, and crowding in public locations of 26 different types. We show that the United States' rationing of public spaces, postspring 2020, has failed to achieve efficiency along either dimension. In April 2020, the United States did achieve notable decreases in visits to public spaces and focused these reductions at locations that offer poor benefit-to-risk tradeoffs. However, this achievement was marred by an increase, from March to April, in crowding at remaining locations due to fewer locations remaining open. In December 2020, at the height of the pandemic so far, crowding in and total visits to locations were higher than in February, before the U.S. pandemic, and these increases were concentrated in locations with the worst value-to-risk tradeoff.

Benzell Seth G, Collis Avinash, Nicolaides Christos

2021-Sep-22

COVID-19, nonpharmaceutical interventions, social contact, social welfare, transmission risk

Pathology Pathology

Bayesian supervised machine learning classification of neural networks with pathological perturbations.

In Biomedical physics & engineering express

Objective Extraction of temporal features of neuronal activity from electrophysiological data can be used for accurate classification of neural networks in healthy and pathologically perturbed conditions. In this study, we provide an extensive approach for the classification of humanin vitroneural networks with and without an underlying pathology, from electrophysiological recordings obtained using a microelectrode array (MEA) platform. Approach We developed a Dirichlet mixture (DM) Point Process statistical model able to extract temporal features related to neurons. We then applied a machine learning algorithm to discriminate between healthy control and pathologically perturbedin vitroneural networks. Main Results We found a high degree of separability between the classes using DM point process features (p-value <0.001 for all the features, paired t-test), which reaches 93.10 of accuracy (92.37 of ROC AUC) with the Random Forest classifier. In particular, results show a higher latency in firing for pathologically perturbed neurons (43 ± 16 ms vs 67 ± 31 ms,μIGfeature distribution). Significance Our approach has been successful in extracting temporal features related to the neurons' behaviour, as well as distinguishing healthy from pathologically perturbed networks, including classification of responses to a transient induced perturbation.

Levi Riccardo, Valderhaug Vibeke Devold, Castelbuono Salvatore, Sandvig Axel, Sandvig Ioanna, Barbieri Riccardo

2021-Sep-22

in vitro neural networks, machine learning, multi electrode array, neurophysiology, point process

General General

LiDAR-driven spiking neural network for collision avoidance in autonomous driving.

In Bioinspiration & biomimetics

Facilitated by advances in real-time sensing, low and high-level control, and machine learning, autonomous vehicles draw ever-increasing attention from many branches of knowledge. Neuromorphic (brain-inspired) implementation of robotic control has been shown to outperform conventional control paradigms in terms of energy efficiency, robustness to perturbations, and adaptation to varying conditions. Here we propose LiDAR-driven neuromorphic control of both vehicle's speed and steering. We evaluated and compared neuromorphic PID control and online learning for speed control, finally suggesting proportional learning as a preferred control scheme. We employed biologically plausible basal-ganglia and thalamus neural models for steering and collision-avoidance, finally extending them to support a null controller, significantly increasing performance.

Shalumov Albert, Halaly Raz, Ezra Tsur Elishai

2021-Sep-22

PID control, autonomous driving, neural engineering framework, neuromorphic control, neuromorphic engineering, online learning

General General

Improved pathogenicity prediction for rare human missense variants.

In American journal of human genetics

The success of personalized genomic medicine depends on our ability to assess the pathogenicity of rare human variants, including the important class of missense variation. There are many challenges in training accurate computational systems, e.g., in finding the balance between quantity, quality, and bias in the variant sets used as training examples and avoiding predictive features that can accentuate the effects of bias. Here, we describe VARITY, which judiciously exploits a larger reservoir of training examples with uncertain accuracy and representativity. To limit circularity and bias, VARITY excludes features informed by variant annotation and protein identity. To provide a rationale for each prediction, we quantified the contribution of features and feature combinations to the pathogenicity inference of each variant. VARITY outperformed all previous computational methods evaluated, identifying at least 10% more pathogenic variants at thresholds achieving high (90% precision) stringency.

Wu Yingzhou, Li Roujia, Sun Song, Weile Jochen, Roth Frederick P

2021-Sep-16

balanced precision, disease variants, human genetics, machine learning, missense variants, predictive medicine, rare variants, variant pathogenicity

Pathology Pathology

Machine learning methods for automated classification of tumors with papillary thyroid carcinoma-like nuclei: A quantitative analysis.

In PloS one ; h5-index 176.0

When approaching thyroid gland tumor classification, the differentiation between samples with and without "papillary thyroid carcinoma-like" nuclei is a daunting task with high inter-observer variability among pathologists. Thus, there is increasing interest in the use of machine learning approaches to provide pathologists real-time decision support. In this paper, we optimize and quantitatively compare two automated machine learning methods for thyroid gland tumor classification on two datasets to assist pathologists in decision-making regarding these methods and their parameters. The first method is a feature-based classification originating from common image processing and consists of cell nucleus segmentation, feature extraction, and subsequent thyroid gland tumor classification utilizing different classifiers. The second method is a deep learning-based classification which directly classifies the input images with a convolutional neural network without the need for cell nucleus segmentation. On the Tharun and Thompson dataset, the feature-based classification achieves an accuracy of 89.7% (Cohen's Kappa 0.79), compared to the deep learning-based classification of 89.1% (Cohen's Kappa 0.78). On the Nikiforov dataset, the feature-based classification achieves an accuracy of 83.5% (Cohen's Kappa 0.46) compared to the deep learning-based classification 77.4% (Cohen's Kappa 0.35). Thus, both automated thyroid tumor classification methods can reach the classification level of an expert pathologist. To our knowledge, this is the first study comparing feature-based and deep learning-based classification regarding their ability to classify samples with and without papillary thyroid carcinoma-like nuclei on two large-scale datasets.

Böhland Moritz, Tharun Lars, Scherr Tim, Mikut Ralf, Hagenmeyer Veit, Thompson Lester D R, Perner Sven, Reischl Markus

2021

General General

Implementation of artificial intelligence algorithms for melanoma screening in a primary care setting.

In PloS one ; h5-index 176.0

Skin cancer is currently the most common type of cancer among Caucasians. The increase in life expectancy, along with new diagnostic tools and treatments for skin cancer, has resulted in unprecedented changes in patient care and has generated a great burden on healthcare systems. Early detection of skin tumors is expected to reduce this burden. Artificial intelligence (AI) algorithms that support skin cancer diagnoses have been shown to perform at least as well as dermatologists' diagnoses. Recognizing the need for clinically and economically efficient means of diagnosing skin cancers at early stages in the primary care attention, we developed an efficient computer-aided diagnosis (CAD) system to be used by primary care physicians (PCP). Additionally, we developed a smartphone application with a protocol for data acquisition (i.e., photographs, demographic data and short clinical histories) and AI algorithms for clinical and dermoscopic image classification. For each lesion analyzed, a report is generated, showing the image of the suspected lesion and its respective Heat Map; the predicted probability of the suspected lesion being melanoma or malignant; the probable diagnosis based on that probability; and a suggestion on how the lesion should be managed. The accuracy of the dermoscopy model for melanoma was 89.3%, and for the clinical model, 84.7% with 0.91 and 0.89 sensitivity and 0.89 and 0.83 specificity, respectively. Both models achieved an area under the curve (AUC) above 0.9. Our CAD system can screen skin cancers to guide lesion management by PCPs, especially in the contexts where the access to the dermatologist can be difficult or time consuming. Its use can enable risk stratification of lesions and/or patients and dramatically improve timely access to specialist care for those requiring urgent attention.

Giavina-Bianchi Mara, de Sousa Raquel Machado, Paciello Vitor Zago de Almeida, Vitor William Gois, Okita Aline Lissa, Prôa Renata, Severino Gian Lucca Dos Santos, Schinaid Anderson Alves, Espírito Santo Rafael, Machado Birajara Soares

2021

General General

Learning, visualizing and exploring 16S rRNA structure using an attention-based deep neural network.

In PLoS computational biology

Recurrent neural networks with memory and attention mechanisms are widely used in natural language processing because they can capture short and long term sequential information for diverse tasks. We propose an integrated deep learning model for microbial DNA sequence data, which exploits convolutional neural networks, recurrent neural networks, and attention mechanisms to predict taxonomic classifications and sample-associated attributes, such as the relationship between the microbiome and host phenotype, on the read/sequence level. In this paper, we develop this novel deep learning approach and evaluate its application to amplicon sequences. We apply our approach to short DNA reads and full sequences of 16S ribosomal RNA (rRNA) marker genes, which identify the heterogeneity of a microbial community sample. We demonstrate that our implementation of a novel attention-based deep network architecture, Read2Pheno, achieves read-level phenotypic prediction. Training Read2Pheno models will encode sequences (reads) into dense, meaningful representations: learned embedded vectors output from the intermediate layer of the network model, which can provide biological insight when visualized. The attention layer of Read2Pheno models can also automatically identify nucleotide regions in reads/sequences which are particularly informative for classification. As such, this novel approach can avoid pre/post-processing and manual interpretation required with conventional approaches to microbiome sequence classification. We further show, as proof-of-concept, that aggregating read-level information can robustly predict microbial community properties, host phenotype, and taxonomic classification, with performance at least comparable to conventional approaches. An implementation of the attention-based deep learning network is available at https://github.com/EESI/sequence_attention (a python package) and https://github.com/EESI/seq2att (a command line tool).

Zhao Zhengqiao, Woloszynek Stephen, Agbavor Felix, Mell Joshua Chang, Sokhansanj Bahrad A, Rosen Gail L

2021-Sep-22

Public Health Public Health

Incorporating metadata in HIV transmission network reconstruction: A machine learning feasibility assessment.

In PLoS computational biology

HIV molecular epidemiology estimates the transmission patterns from clustering genetically similar viruses. The process involves connecting genetically similar genotyped viral sequences in the network implying epidemiological transmissions. This technique relies on genotype data which is collected only from HIV diagnosed and in-care populations and leaves many persons with HIV (PWH) who have no access to consistent care out of the tracking process. We use machine learning algorithms to learn the non-linear correlation patterns between patient metadata and transmissions between HIV-positive cases. This enables us to expand the transmission network reconstruction beyond the molecular network. We employed multiple commonly used supervised classification algorithms to analyze the San Diego Primary Infection Resource Consortium (PIRC) cohort dataset, consisting of genotypes and nearly 80 additional non-genetic features. First, we trained classification models to determine genetically unrelated individuals from related ones. Our results show that random forest and decision tree achieved over 80% in accuracy, precision, recall, and F1-score by only using a subset of meta-features including age, birth sex, sexual orientation, race, transmission category, estimated date of infection, and first viral load date besides genetic data. Additionally, both algorithms achieved approximately 80% sensitivity and specificity. The Area Under Curve (AUC) is reported 97% and 94% for random forest and decision tree classifiers respectively. Next, we extended the models to identify clusters of similar viral sequences. Support vector machine demonstrated one order of magnitude improvement in accuracy of assigning the sequences to the correct cluster compared to dummy uniform random classifier. These results confirm that metadata carries important information about the dynamics of HIV transmission as embedded in transmission clusters. Hence, novel computational approaches are needed to apply the non-trivial knowledge collected from inter-individual genetic information to metadata from PWH in order to expand the estimated transmissions. We note that feature extraction alone will not be effective in identifying patterns of transmission and will result in random clustering of the data, but its utilization in conjunction with genetic data and the right algorithm can contribute to the expansion of the reconstructed network beyond individuals with genetic data.

Mazrouee Sepideh, Little Susan J, Wertheim Joel O

2021-Sep

Cardiology Cardiology

Overview of featurization techniques used in traditional versus emerging deep learning-based algorithms for automated interpretation of the 12-lead ECG.

In Journal of electrocardiology

Automated interpretation of the 12-lead ECG has remained an underpinning interest in decades of research that has seen a diversity of computing applications in cardiology. The application of computers in cardiology began in the 1960s with early research focusing on the conversion of analogue ECG signals (voltages) to digital samples. Alongside this, software techniques that automated the extraction of wave measurements and provided basic diagnostic statements, began to emerge. In the years since then there have been many significant milestones which include the widespread commercialisation of 12-lead ECG interpretation software, associated clinical utility and the development of the related regulatory frameworks to promote standardised development. In the past few years, the research community has seen a significant rejuvenation in the development of ECG interpretation programs. This is evident in the research literature where a large number of studies have emerged tackling a variety of automated ECG interpretation problems. This is largely due to two factors. Specifically, the technical advances, both software and hardware, that have facilitated the broad adoption of modern artificial intelligence (AI) techniques, and, the increasing availability of large datasets that support modern AI approaches. In this article we provide a very high-level overview of the operation of and approach to the development of early 12-lead ECG interpretation programs and we contrast this to the approaches that are now seen in emerging AI approaches. Our overview is mainly focused on highlighting differences in how input data are handled prior to generation of the diagnostic statement.

Finlay Dewar, Bond Raymond, Jennings Michael, McCausland Christopher, Guldenring Daniel, Kennedy Alan, Biglarbeigi Pardis, Al-Zaiti Salah S, Brisk Rob, McLaughlin James

2021-Aug-17

Artificial intelligence, Automated electrocardiogram interpretation, Deep learning, ECG

General General

COV-VGX: An automated COVID-19 detection system using X-ray images and transfer learning.

In Informatics in medicine unlocked

Coronavirus (COVID-19) has been one of the most dangerous and acute deadly diseases across the world recently. Researchers are trying to develop automated and feasible COVID-19 detection systems with the help of deep neural networks, machine learning techniques, etc. In this paper, a deep learning-based COVID-19 detection system called COV-VGX is proposed that contributes to detecting coronavirus disease automatically using chest X-ray images. The system introduces two types of classifiers, namely, a multiclass classifier that automatically predicts coronavirus, pneumonia, and normal classes and a binary classifier that predicts coronavirus and pneumonia classes. Using transfer learning, a deep CNN model is proposed to extract distinct and high-level features from X-ray images in collaboration with the pretrained model VGG-16. Despite the limitation of the COVID-19 dataset, the model is evaluated with sufficient COVID-19 images. Extensive experiments for multiclass classifier have achieved 98.91% accuracy, 97.31% precision, 99.50% recall, 98.39% F1-score, while 99.37% accuracy, 98.76% precision, 100% recall, 99.38% F1-score for binary classifier. The proposed system can contribute a lot in diagnosing COVID-19 effectively in the medical field.

Saha Prottoy, Sadi Muhammad Sheikh, Aranya O F M Riaz Rahman, Jahan Sadia, Islam Ferdib-Al

2021

COVID-19, Deep learning, Transfer learning, VGG-16, X-ray images

Public Health Public Health

The State of Mind of Healthcare Professionals in the Light of the COVID-19: Insights from Text Analysis of Twitter's Online Discourses.

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : The COVID-19 pandemic has affected populations worldwide, with extreme health, economic, social, and political implications. Healthcare professionals (HCPs) are at the core of pandemic response and are among the most crucial factors in maintaining coping capacities. Yet, they are also vulnerable to mental health effects, managing a long-lasting emergency under lack of resources and complicated personal concerns. However, there is a lack of longitudinal studies that investigate the HCP population.

OBJECTIVE : To analyse the state of mind of HCPs as expressed in online discussions published on Twitter in light of COVID-19, from the pandemic onset until the end of 2020.

METHODS : The population for this study was selected from followers of a few hundred Twitter accounts of healthcare organizations and common HCP points of interest. We used active learning, a process that iteratively uses machine learning and manual data labeling, to select the large-scale population of Twitter accounts maintained by English-speaking HCPs focusing on individuals rather than official organizations. We analyzed the topics and emotions in their discourse during 2020. The topic distributions were obtained using the Latent Dirichlet Allocation (LDA) algorithm. We defined a measure of topic cohesion and described the most cohesive topics. The emotions expressed in tweets during 2020 were compared to 2019. Finally, the emotion intensities were cross-correlated with the pandemic waves to explore possible associations between the pandemic development and emotional response.

RESULTS : We analyzed timelines of 53,063 Twitter profiles, 90% of which are maintained by individual HCPs. Professional topics account for 44.5% of tweets by HCPs from Jan. 1st to Dec. 6th, 2020. Events such as the pandemic waves, U.S. elections, or the George Floyd case affect the HCPs' discourse. The levels of joy and sadness exceed their minimal and maximal values yesteryear, respectively, 80% of the time, P= .001. Most interestingly, fear precedes the pandemic waves (in terms of the differences in confirmed cases) by two weeks with a Spearman correlations coefficient of ρ(47)= .34, P= .026.

CONCLUSIONS : Analyses of longitudinal data over the 2020 year reveal that a large fraction of HCP discourse is related directly to professional content, including the increase in the volume of discussions following the pandemic waves. The changes in emotional patterns (decrease of joy, an increase of sadness, fear, and disgust) during the year 2020 may indicate the utmost importance in providing emotional support for HCPs to prevent fatigue, burnout, and mental health disorders postpandemic period. The increase of fear two weeks in advance of pandemic waves indicates that HCPs are in a position and with adequate qualification to anticipate the pandemic development, and could serve as a bottom-up pathway for expressing the morbidity and clinical situation to health agencies.

CLINICALTRIAL :

Elyashar Aviad, Plochotnikov Ilia, Cohen Idan-Chaim, Puzis Rami, Cohen Odeya

2021-Jul-23

General General

Multikernel Correntropy for Robust Learning.

In IEEE transactions on cybernetics

As a novel similarity measure that is defined as the expectation of a kernel function between two random variables, correntropy has been successfully applied in robust machine learning and signal processing to combat large outliers. The kernel function in correntropy is usually a zero-mean Gaussian kernel. In a recent work, the concept of mixture correntropy (MC) was proposed to improve the learning performance, where the kernel function is a mixture Gaussian kernel, namely, a linear combination of several zero-mean Gaussian kernels with different widths. In both correntropy and MC, the center of the kernel function is, however, always located at zero. In the present work, to further improve the learning performance, we propose the concept of multikernel correntropy (MKC), in which each component of the mixture Gaussian kernel can be centered at a different location. The properties of the MKC are investigated and an efficient approach is proposed to determine the free parameters in MKC. Experimental results show that the learning algorithms under the maximum MKC criterion (MMKCC) can outperform those under the original maximum correntropy criterion (MCC) and the maximum MC criterion (MMCC).

Chen Badong, Xie Yuqing, Wang Xin, Yuan Zejian, Ren Pengju, Qin Jing

2021-Sep-22

General General

Toward Robust Fault Identification of Complex Industrial Processes Using Stacked Sparse-Denoising Autoencoder With Softmax Classifier.

In IEEE transactions on cybernetics

This article proposes a robust end-to-end deep learning-induced fault recognition scheme by stacking multiple sparse-denoising autoencoders with a Softmax classifier, called stacked spare-denoising autoencoder (SSDAE)-Softmax, for the fault identification of complex industrial processes (CIPs). Specifically, sparse denoising autoencoder (SDAE) is established by integrating a sparse AE (SAE) with a denoising AE (DAE) for the low-dimensional but intrinsic feature representation of the CIP monitoring data (CIPMD) with possible noise contamination. SSDAE-Softmax is established by stacking multiple SDAEs with a layerwise pretraining procedure, and a Softmax classifier with a global fine-tuning strategy. Furthermore, SSDAE-Softmax hyperparameters are optimized by a relatively new global optimization algorithm, referred to as the state transition algorithm (STA). Benefiting from the deep learning-based feature representation scheme with the STA-based hyperparameter optimization, the underlying intrinsic characteristics of CIPMD can be learned automatically and adaptively for accurate fault identification. A numeric simulation system, the benchmark Tennessee Eastman process (TEP), and a real industrial process, that is, the continuous casting process (CCP) from a top steel plant of China, are used to validate the performance of the proposed method. Experimental results show that the proposed SSDAE-Softmax model can effectively identify various process faults, and has stronger robustness and adaptability against the noise interference in CIPMD for the process monitoring of CIPs.

Liu Jinping, Xu Longcheng, Xie Yongfang, Ma Tianyu, Wang Jie, Tang Zhaohui, Gui Weihua, Yin Huazhan, Jahanshahi Hadi

2021-Sep-22

General General

Spatiotemporal Sequence Prediction With Point Processes and Self-Organizing Decision Trees.

In IEEE transactions on neural networks and learning systems

We study the spatiotemporal prediction problem and introduce a novel point-process-based prediction algorithm. Spatiotemporal prediction is extensively studied in machine learning literature due to its critical real-life applications, such as crime, earthquake, and social event prediction. Despite these thorough studies, specific problems inherent to the application domain are not yet fully explored. Here, we address the nonstationary spatiotemporal prediction problem on both densely and sparsely distributed sequences. We introduce a probabilistic approach that partitions the spatial domain into subregions and models the event arrivals in each region with interacting point processes. Our algorithm can jointly learn the spatial partitioning and the interaction between these regions through a gradient-based optimization procedure. Finally, we demonstrate the performance of our algorithm on both simulated data and two real-life datasets. We compare our approach with baseline and state-of-the-art deep learning-based approaches, where we achieve significant performance improvements. Moreover, we also show the effect of using different parameters on the overall performance through empirical results and explain the procedure for choosing the parameters.

Karaahmetoglu Oguzhan, Kozat Suleyman Serdar

2021-Sep-22

General General

Tracking Beyond Detection: Learning a Global Response Map for End-to-End Multi-Object Tracking.

In IEEE transactions on image processing : a publication of the IEEE Signal Processing Society

Most of the existing Multi-Object Tracking (MOT) approaches follow the Tracking-by-Detection and Data Association paradigm, in which objects are firstly detected and then associated in the tracking process. In recent years, deep neural network has been utilized to obtain more discriminative appearance features for cross-frame association, and noticeable performance improvement has been reported. On the other hand, the Tracking-by-Detection framework is yet not completely end-to-end, which leads to huge computation and limited performance especially in the inference (tracking) process. To address this problem, we present an effective end-to-end deep learning framework which can directly take image-sequence/video as input and output the located and tracked objects of learned types. Specifically, a novel global response network is learned to project multiple objects in the image-sequence/video into a continuous response map, and the trajectory of each tracked object can then be easily picked out. The overall process is similar to how a detector inputs an image and outputs the bounding boxes of each detected object. Experimental results based on the MOT16 and MOT17 benchmarks show that our proposed on-line tracker achieves state-of-the-art performance on several tracking metrics.

Wan Xingyu, Cao Jiakai, Zhou Sanping, Wang Jinjun, Zheng Nanning

2021-Sep-22

General General

High-throughput single-cell quantification of hundreds of proteins using conventional flow cytometry and machine learning.

In Science advances

[Figure: see text].

Becht Etienne, Tolstrup Daniel, Dutertre Charles-Antoine, Morawski Peter A, Campbell Daniel J, Ginhoux Florent, Newell Evan W, Gottardo Raphael, Headley Mark B

2021-Sep-24

Ophthalmology Ophthalmology

Refractive outcomes of second-eye adjustment methods on intraocular lens power calculation in second eye.

In Clinical & experimental ophthalmology

BACKGROUND : To investigate the refractive outcomes of second-eye adjustment (SEA) methods in different intraocular lens (IOL) power calculation formulas for second eye following bilateral sequential cataract surgery.

METHODS : This retrospective consecutive case-series study included 234 eyes from 234 patients who underwent bilateral sequential phacoemulsification and implantation of enVista MX60 in a hospital setting. Postoperative refraction outcomes calculated by standard formulas (SRK/T and Barrett Universal II, BUII) with SEA method were compared with those calculated by an artificial intelligence-based IOL power calculation formula (PEARL DGS) under second eye enhancement (SEE) method. The median absolute error (MedAE), mean absolute error (MAE) and percentage prediction errors (PE) of eyes within ±0.25 diopters (D), ±0.50 D, ±0.75 D and ± 1.00 D were determined.

RESULTS : Overall, the improvement in MedAE after SEA was significant for PEARL DGS (P < 0.01), SRK/T (P < 0.001) and BUII (P = 0.031), which increased from 74.36%, 71.37%, and 77.78% to 83.33%, 80.34%, and 79.49% of eyes within a PE of ±0.50 D, respectively. For first eyes with a medium axial length (22-26 mm), PEARL DGS with SEE had the lowest MedAE (0.21 D). For a first-eye MAE over 0.50 D, SEA method led to significant improvement in the second eye (P < 0.01). Interocular axis length differences exceeding 0.3 mm were associated with weaker effects using SEA in the studied formulas (P > 0.05).

CONCLUSIONS : Either second-eye adjustment method with SRK/T and BUII formulas or second-eye enhancement method based on the PEARL DGS formula can improve postoperative refractive outcomes in second eye. This article is protected by copyright. All rights reserved.

Mao Yan, Li Jianbing, Xu Yanxin, Qin Yingyan, Liu Liangping, Wu Mingxing

2021-Sep-22

Axial length, Intraocular lens power calculation, Mean absolute error, Second-eye adjustment

oncology Oncology

Immune-related eight-lncRNA signature for improving prognosis prediction of lung adenocarcinoma.

In Journal of clinical laboratory analysis ; h5-index 21.0

BACKGROUND : Lung adenocarcinoma (LUAD) is the leading cause of cancer-related deaths worldwide. Therefore, the identification of a novel prediction signature for predicting the prognosis risk and survival outcomes is urgently demanded.

METHODS : We integrated a machine-learning frame by combing the Cox regression and Least Absolute Shrinkage and Selection Operator (LASSO) regression model to identify the LUAD-related long non-coding RNA (lncRNA) survival biomarkers. Subsequently, the Spearman correlation test was employed to interrogate the relationships between lncRNA signature and tumor immunity and constructed the competing endogenous RNA (ceRNA) network.

RESULTS : Herein, we identified an eight-lncRNA signature (PR-lncRNA signature, NPSR1-AS1, SATB2-AS1, LINC01090, FGF12-AS2, AC005256.1, MAFA-AS1, BFSP2-AS1, and CPC5-AS1), which contributes to predicting LUAD patient's prognosis risk and survival outcomes. The PR-lncRNA signature has also been confirmed as the robust signature in independent datasets. Further parsing of the LUAD tumor immune infiltration showed the PR-lncRNAs were closely associated with the abundance of multiple immune cells infiltration and the expression of MHC molecules. Furthermore, by constructing the PR-lncRNA-related ceRNA network, we interrogated more potential anti-cancer therapy targets.

CONCLUSION : lncRNAs, as emerging cancer biomarkers, play an important role in a variety of cancer processes. Identification of PR-lncRNA signatures allows us to better predict patient's survival outcomes and disease risk. Finally, the PR-lncRNA signatures could help us to develop novel LUAD anti-cancer therapeutic strategies.

Chen Yan, Zhang Xiuxiu, Li Jinze, Zhou Min

2021-Sep-22

long non-coding RNA, lung adenocarcinoma, machine learning, prognosis, tumor immunoactivity

Public Health Public Health

Semi-automated Tools for Systematic Searches.

In Methods in molecular biology (Clifton, N.J.)

Traditionally, literature identification for systematic reviews has relied on a two-step process: first, searching databases to identify potentially relevant citations, and then manually screening those citations. A number of tools have been developed to streamline and semi-automate this process, including tools to generate terms; to visualize and evaluate search queries; to trace citation linkages; to deduplicate, limit, or translate searches across databases; and to prioritize relevant abstracts for screening. Research is ongoing into tools that can unify searching and screening into a single step, and several protype tools have been developed. As this field grows, it is becoming increasingly important to develop and codify methods for evaluating the extent to which these tools fulfill their purpose.

Adam Gaelen P, Wallace Byron C, Trikalinos Thomas A

2022

Information science, Literature identification, Machine learning, Systematic review methods, Text mining

Public Health Public Health

A Machine Learning Approach to Predict the Added Sugar Content of Packaged Foods.

In The Journal of nutrition ; h5-index 61.0

BACKGROUND : Dietary guidelines recommend limiting the intake of added sugars. However, despite the public health importance, most countries have not mandated the labeling of added sugar content on packaged foods and beverages, making it difficult for consumers to avoid products with added sugar, and limiting the ability of policymakers to identify priority products for intervention.

OBJECTIVE : To develop a machine learning approach for the prediction of added sugar content in packaged products using available nutrient, ingredient, and food category information.

DESIGN : The added sugar prediction algorithm was developed using k-Nearest Neighbors (KNN) and packaged food information from the US Label Insight dataset (n = 70,522). A synthetic dataset of Australian packaged products (n = 500) was used to assess validity and generalization. Performance metrics included the coefficient of determination (R2), mean absolute error (MAE), and Spearman rank correlation (ρ). To benchmark the KNN approach, the KNN approach was compared to an existing added sugar prediction approach that relies on a series of manual steps.

RESULTS : Compared to the existing added sugar prediction approach, the KNN approach was similarly apt at explaining variation in added sugar content (R2 = 0.96 vs. 0.97 respectively) and ranking products from highest to lowest in added sugar content (ρ = 0.91 vs. 0.93 respectively), while less apt at minimizing absolute deviations between predicted and true values (MAE = 1.68 g vs. 1.26 g per 100 g or 100 mL respectively).

CONCLUSIONS : KNN can be used to predict added sugar content in packaged products with a high degree of validity. Being automated, KNN can easily be applied to large datasets. Such predicted added sugar levels can be used to monitor the food supply and inform interventions aimed at reducing added sugar intake.

Davies Tazman, Louie Jimmy Chun Yu, Ndanuko Rhoda, Barbieri Sebastiano, Perez-Concha Oscar, Wu Jason H Y

2021-Sep-22

added sugar, automated, estimation, machine learning, packaged foods, prediction

Ophthalmology Ophthalmology

Modular deep neural networks for automatic quality control of retinal optical coherence tomography scans.

In Computers in biology and medicine

Retinal optical coherence tomography (OCT) with intraretinal layer segmentation is increasingly used not only in ophthalmology but also for neurological diseases such as multiple sclerosis (MS). Signal quality influences segmentation results, and high-quality OCT images are needed for accurate segmentation and quantification of subtle intraretinal layer changes. Among others, OCT image quality depends on the ability to focus, patient compliance and operator skills. Current criteria for OCT quality define acceptable image quality, but depend on manual rating by experienced graders and are time consuming and subjective. In this paper, we propose and validate a standardized, grader-independent, real-time feedback system for automatic quality assessment of retinal OCT images. We defined image quality criteria for scan centering, signal quality and image completeness based on published quality criteria and typical artifacts identified by experienced graders when inspecting OCT images. We then trained modular neural networks on OCT data with manual quality grading to analyze image quality features. Quality analysis by a combination of these trained networks generates a comprehensive quality report containing quantitative results. We validated the approach against quality assessment according to the OSCAR-IB criteria by an experienced grader. Here, 100 OCT files with volume, circular and radial scans, centered on optic nerve head and macula, were analyzed and classified. A specificity of 0.96, a sensitivity of 0.97 and an accuracy of 0.97 as well as a Matthews correlation coefficient of 0.93 indicate a high rate of correct classification. Our method shows promising results in comparison to manual OCT grading and may be useful for real-time image quality analysis or analysis of large data sets, supporting standardized application of image quality criteria.

Kauer-Bonin Josef, Yadav Sunil K, Beckers Ingeborg, Gawlik Kay, Motamedi Seyedamirhosein, Zimmermann Hanna G, Kadas Ella M, Haußer Frank, Paul Friedemann, Brandt Alexander U

2021-Sep-18

Automatic quality analysis, Deep learning, OCT quality Analysis, OCT quality Standard, Quality classification

Public Health Public Health

Estimating the Impact of COVID-19 on the PM2.5 Levels in China with a Satellite-Driven Machine Learning Model.

In Remote sensing

China implemented an aggressive nationwide lockdown procedure immediately after the COVID-19 outbreak in January 2020. As China emerges from the impact of COVID-19 on national economic and industrial activities, it has become the site of a large-scale natural experiment to evaluate the impact of COVID-19 on regional air quality. However, ground measurements of fine particulate matters (PM2.5) concentrations do not offer comprehensive spatial coverage, especially in suburban and rural regions. In this study, we developed a machine learning method with satellite aerosol remote sensing data, meteorological fields and land use parameters as major predictor variables to estimate spatiotemporally resolved daily PM2.5 concentrations in China. Our study period consists of a reference semester (1 November 2018-30 April 2019) and a pandemic semester (1 November 2019-30 April 2020), with six modeling months in each semester. Each period was then divided into subperiod 1 (November and December), subperiod 2 (January and February) and subperiod 3 (March and April). The reference semester model obtained a 10-fold cross-validated R2 (RMSE) of 0.79 (17.55 μg/m3) and the pandemic semester model obtained a 10-fold cross-validated R2 (RMSE) of 0.83 (13.48 μg/m3) for daily PM2.5 predictions. Our prediction results showed high PM2.5 concentrations in the North China Plain, Yangtze River Delta, Sichuan Basin and Xinjiang Autonomous Region during the reference semester. PM2.5 levels were lowered by 4.8 μg/m3 during the pandemic semester compared to the reference semester and PM2.5 levels during subperiod 2 decreased most, by 18%. The southeast region was affected most by the COVID-19 outbreak with PM2.5 levels during subperiod 2 decreasing by 31%, followed by the Northern Yangtze River Delta (29%) and Pearl River Delta (24%).

Li Qiulun, Zhu Qingyang, Xu Muwu, Zhao Yu, Narayan K M Venkat, Liu Yang

2021-Apr

COVID-19, China, MAIAC AOD, PM2.5, air pollution, machine learning, random forest, remote sensing

Dermatology Dermatology

Lack of Transparency and Potential Bias in Artificial Intelligence Data Sets and Algorithms: A Scoping Review.

In JAMA dermatology ; h5-index 54.0

Importance : Clinical artificial intelligence (AI) algorithms have the potential to improve clinical care, but fair, generalizable algorithms depend on the clinical data on which they are trained and tested.

Objective : To assess whether data sets used for training diagnostic AI algorithms addressing skin disease are adequately described and to identify potential sources of bias in these data sets.

Data Sources : In this scoping review, PubMed was used to search for peer-reviewed research articles published between January 1, 2015, and November 1, 2020, with the following paired search terms: deep learning and dermatology, artificial intelligence and dermatology, deep learning and dermatologist, and artificial intelligence and dermatologist.

Study Selection : Studies that developed or tested an existing deep learning algorithm for triage, diagnosis, or monitoring using clinical or dermoscopic images of skin disease were selected, and the articles were independently reviewed by 2 investigators to verify that they met selection criteria.

Consensus Process : Data set audit criteria were determined by consensus of all authors after reviewing existing literature to highlight data set transparency and sources of bias.

Results : A total of 70 unique studies were included. Among these studies, 1 065 291 images were used to develop or test AI algorithms, of which only 257 372 (24.2%) were publicly available. Only 14 studies (20.0%) included descriptions of patient ethnicity or race in at least 1 data set used. Only 7 studies (10.0%) included any information about skin tone in at least 1 data set used. Thirty-six of the 56 studies developing new AI algorithms for cutaneous malignant neoplasms (64.3%) met the gold standard criteria for disease labeling. Public data sets were cited more often than private data sets, suggesting that public data sets contribute more to new development and benchmarks.

Conclusions and Relevance : This scoping review identified 3 issues in data sets that are used to develop and test clinical AI algorithms for skin disease that should be addressed before clinical translation: (1) sparsity of data set characterization and lack of transparency, (2) nonstandard and unverified disease labels, and (3) inability to fully assess patient diversity used for algorithm development and testing.

Daneshjou Roxana, Smith Mary P, Sun Mary D, Rotemberg Veronica, Zou James

2021-Sep-22

Radiology Radiology

Sleep and perivascular spaces in the middle-aged and elderly population.

In Journal of sleep research

Sleep has been hypothesised to facilitate waste clearance from the brain. We aimed to determine whether sleep is associated with perivascular spaces on brain magnetic resonance imaging (MRI), a potential marker of impaired brain waste clearance, in a population-based cohort of middle-aged and elderly people. In 559 participants (mean [SD] age 62 [6] years, 52% women) from the population-based Rotterdam Study, we measured total sleep time, sleep onset latency, wake after sleep onset and sleep efficiency with actigraphy and polysomnography. Perivascular space load was determined with brain MRI in four regions (centrum semiovale, basal ganglia, hippocampus, and midbrain) via a validated machine learning algorithm using T2-weighted MR images. Associations between sleep characteristics and perivascular space load were analysed with zero-inflated negative binomial regression models adjusted for various confounders. We found that higher actigraphy-estimated sleep efficiency was associated with a higher perivascular space load in the centrum semiovale (odds ratio 1.10, 95% confidence interval 1.04-1.16, p = 0.0008). No other actigraphic or polysomnographic sleep characteristics were associated with perivascular space load in other brain regions. We conclude that, contrary to our hypothesis, associations of sleep with perivascular space load in this middle-aged and elderly population remained limited to an association of a high actigraphy-estimated sleep efficiency with a higher perivascular space load in the centrum semiovale.

Lysen Thom S, Yilmaz Pinar, Dubost Florian, Ikram M Arfan, de Bruijne Marleen, Vernooij Meike W, Luik Annemarie I

2021-Sep-22

VRS, Virchow-Robin, community-dwelling, epidemiology, glymphatic, paravascular

Surgery Surgery

Preoperative Survival Prediction in Intrahepatic Cholangiocarcinoma Using a Ultrasound-Based Radiographic-Radiomics Signature.

In Journal of ultrasound in medicine : official journal of the American Institute of Ultrasound in Medicine

OBJECTIVES : To construct a preoperative model for survival prediction in intrahepatic cholangiocarcinoma (ICC) patients using ultrasound (US) based radiographic-radiomics signatures.

METHODS : Between April 2010 and September 2015, 170 patients with ICC who underwent curative resection were retrospectively recruited. Overall survival (OS)-related radiographic signatures and radiomics signatures based on preoperative US were built and assessed through a time-dependent receiver operating characteristic curve analysis. A nomogram was developed based on the selected predictors from the radiographic-radiomics signatures and clinical and laboratory results of the training cohort (n = 127), validated in an independent testing cohort (n = 43) by the concordance index (C-index), and compared with the Tumor Node Metastasis (TNM) cancer staging system as well as the radiographic and radiomics nomograms.

RESULTS : The median areas under the curve of the radiomics signature and radiographic signature were higher than that of the TNM staging system in the testing cohort, although the values were not significantly different (0.76-0.82 versus 0.62, P = .485 and .264). The preoperative nomogram with CA 19-9, sex, ascites, radiomics signature, and radiographic signature had C-indexes of 0.72 and 0.75 in the training and testing cohorts, respectively, and it had significantly higher predictive performance than the 8th TNM staging system in the testing cohort (C-index: 0.75 versus 0.67, P = .004) and a higher C-index than the radiomics nomograms (0.75 versus 0.68, P = .044).

CONCLUSIONS : The preoperative nomogram integrated with the radiographic-radiomics signature demonstrated good predictive performance for OS in ICC and was superior to the 8th TNM staging system.

Li Ming-De, Lu Xiao-Zhou, Liu Jun-Feng, Chen Bin, Xu Ming, Xie Xiao-Yan, Lu Ming-De, Kuang Ming, Wang Wei, Shen Shun-Li, Chen Li-Da

2021-Sep-21

intrahepatic cholangiocarcinoma, outcome, ultrasound

General General

Challenges and Limitations in the Studies of Glycoproteins: A Computational Chemist's Perspective.

In Proteins

Experimenters face challenges and limitations while analyzing glycoproteins due to their high flexibility, stereochemistry, anisotropic effects, and hydration phenomena. Computational studies complement experiments and have been used in characterization of the structural properties of glycoproteins. However, recent investigations revealed that computational studies face significant challenges as well. Here, we introduce and discuss some of these challenges and weaknesses in the investigations of glycoproteins. We also present requirements of future developments in computational biochemistry and computational biology areas that could be necessary for providing more accurate structural property analyses of glycoproteins using computational tools. Further theoretical strategies that need to be and can be developed are discussed herein. This article is protected by copyright. All rights reserved.

Balli Oyku Irem, Uversky Vladimir N, Durdagi Serdar, Coskuner-Weber Orkid

2021-Sep-22

Glycoproteins, classical molecular simulations, experimental tools, machine learning, quantum mechanics/molecular mechanics

General General

The Efficiency of U.S. Public Space Utilization During the COVID-19 Pandemic.

In Risk analysis : an official publication of the Society for Risk Analysis

The COVID-19 pandemic has called for and generated massive novel government regulations to increase social distancing for the purpose of reducing disease transmission. A number of studies have attempted to guide and measure the effectiveness of these policies, but there has been less focus on the overall efficiency of these policies. Efficient social distancing requires implementing stricter restrictions during periods of high viral prevalence and rationing social contact to disproportionately preserve gatherings that produce a good ratio of benefits to transmission risk. To evaluate whether U.S. social distancing policy actually produced an efficient social distancing regime, we tracked consumer preferences for, visits to, and crowding in public locations of 26 different types. We show that the United States' rationing of public spaces, postspring 2020, has failed to achieve efficiency along either dimension. In April 2020, the United States did achieve notable decreases in visits to public spaces and focused these reductions at locations that offer poor benefit-to-risk tradeoffs. However, this achievement was marred by an increase, from March to April, in crowding at remaining locations due to fewer locations remaining open. In December 2020, at the height of the pandemic so far, crowding in and total visits to locations were higher than in February, before the U.S. pandemic, and these increases were concentrated in locations with the worst value-to-risk tradeoff.

Benzell Seth G, Collis Avinash, Nicolaides Christos

2021-Sep-22

COVID-19, nonpharmaceutical interventions, social contact, social welfare, transmission risk

Ophthalmology Ophthalmology

Determinants of brain swelling in pediatric and adult cerebral malaria.

In JCI insight

Cerebral malaria (CM) affects children and adults, but brain swelling is more severe in children. To investigate features associated with brain swelling in malaria, we performed blood profiling and brain MRI in a cohort of pediatric and adult patients with CM in Rourkela, India, and compared them with an African pediatric CM cohort in Malawi. We determined that higher plasma Plasmodium falciparum histidine rich protein 2 (PfHRP2) levels and elevated var transcripts that encode for binding to endothelial protein C receptor (EPCR) were linked to CM at both sites. Machine learning models trained on the African pediatric cohort could classify brain swelling in Indian children CM cases but had weaker performance for adult classification, due to overall lower parasite var transcript levels in this age group and more severe thrombocytopenia in Rourkela adults. Subgrouping of patients with CM revealed higher parasite biomass linked to severe thrombocytopenia and higher Group A-EPCR var transcripts in mild thrombocytopenia. Overall, these findings provide evidence that higher parasite biomass and a subset of Group A-EPCR binding variants are common features in children and adult CM cases, despite age differences in brain swelling.

Sahu Praveen K, Duffy Fergal J, Dankwa Selasi, Vishnyakova Maria, Majhi Megharay, Pirpamer Lukas, Vigdorovich Vladimir, Bage Jabamani, Maharana Sameer, Mandala Wilson, Rogerson Stephen J, Seydel Karl B, Taylor Terrie E, Kim Kami, Sather D Noah, Mohanty Akshaya, Mohanty Rashmi R, Mohanty Anita, Pattnaik Rajyabardhan, Aitchison John D, Hoffman Angelika, Mohanty Sanjib, Smith Joseph D, Bernabeu Maria, Wassmer Samuel C

2021-Sep-22

Infectious disease, Malaria, Microbiology, Parasitology, Platelets

General General

[Identification of potential regulatory genes for embryonic stem cell self-renewal and pluripotency by random forest].

In Nan fang yi ke da xue xue bao = Journal of Southern Medical University

OBJECTIVE : To identify novel genes associated with self-renewal and pluripotency of mouse embryonic stem cells(mESCs)by integrating multiomics data based on machine learning methods.

METHODS : We integrated multiomics information of mESCs involving transcriptome, histone modifications, chromatin accessibility, transcription factor binding and architectural protein binding, and compared the signal differences between known stem cell self-renewal and pluripotency genes and other genes.By integrating these multiomics data, we established prediction models based on several machine learning classifiers including random forests and performed 5-fold cross validations.The model was trained using the training dataset containing two thirds of the input samples, and the remaining one third of the input samples were used as the test dataset to assess the performance of the model in independent tests.Finally, the results predicted by the model were validated through gene function annotation and cell function experiments including cell viability assay, colony formation assay and cell cycle analysis.

RESULTS : Compared with the random genes, the genes known to be associated with self-renewal and pluripotency of mESCs in the multiomics data showed significantly different features.Random forest outperformed the other machine learning algorithms tested on these multiomics data, with an area under the curve (AUC) of 0.883±0.018 for cross validation and an AUC of 0.880±0.028 for independent test.Based on this model, we identified 893 potential regulatory genes associated wwith self-renewal and pluripotency of mESCs, which were similar to the known genes in functional annotation.Known-down of the predicted novel regulator gene Cct6a resulted in significant decreases in the cell viability of mESCs (P < 0.0001) and the number of cell clones (P < 0.01), significantly increased the number of cells in G1 phase (P < 0.01) and decreasedthe number of S phase cells (P < 0.05).Knockdown of Cct6a also led to failure of positive alkaline phosphatase staining of the mESCs.

CONCLUSION : Machine learning model based on multiomics data can be used to predict potential self-renewal and pluripotency regulators with high performance.By using this model, we predicted potential self-renewal and pluripotency regulatory genes including Cct6a and applied experimental validation.This model provides new insights into the regulatory mechanism of mESCs and contribute to stem cell research.

Zeng P, Tang X, Wu T, Tian Q, Li M, Ding J

2021-Aug-20

machine learning, mouse embryonic stem cells, pluripotency, random forest, self-renewal

oncology Oncology

Use of classifiers to optimise the identification and characterisation of metastatic breast cancer in a nationwide administrative registry.

In Acta oncologica (Stockholm, Sweden)

BAKGROUND : The prognosis for patients with metastatic breast cancer (MBC) is substantially worse when compared with patients with earlier stage disease. Therefore, understanding the differences in epidemiology between these two patient groups is important. Studies using population-based cancer registries to identify MBC are hampered by the quality of reporting. Patients are registered once (at time of initial diagnosis); hence only data for patients with de novo MBC are identifiable, whereas data for patients with recurrent MBC are not. This makes accurate estimation of the epidemiology and healthcare utilisation of MBC challenging. This study aimed to investigate whether machine-learning could improve identification of MBC in national health registries.

MATERIAL AND METHODS : Data for patients with confirmed MBC from a regional breast cancer registry were used to train machine-learning algorithms (or 'classifiers'). The best performing classifier (accuracy 97.3%, positive predictive value 85.1%) was applied to Swedish national registries for 2008 to 2016.

RESULTS : Mean yearly MBC incidence was estimated at 14 per 100,000 person-years (with 18% diagnosed de novo and 76% of the total with HR-positive MBC).

CONCLUSION : To our knowledge, this is the first study to use machine learning to identify MBC regardless of stage at diagnosis in health registries covering the entire population of Sweden.

Valachis Antonis, Carlqvist Peter, Szilcz Máté, Freilich Jonatan, Vertuani Simona, Holm Barbro, Lindman Henrik

2021-Sep-22

Breast cancer, European cohort, classifier, epidemiology, health registries, metastatic, retrospective study

Ophthalmology Ophthalmology

Evaluation of a New Neural Network Classifier for Diabetic Retinopathy.

In Journal of diabetes science and technology ; h5-index 38.0

BACKGROUND : Medical image segmentation is a well-studied subject within the field of image processing. The goal of this research is to create an AI retinal screening grading system that is both accurate and fast. We introduce a new segmentation network which achieves state-of-the-art results on semantic segmentation of color fundus photographs. By applying the net-work to identify anatomical markers of diabetic retinopathy (DR) and diabetic macular edema (DME), we collect sufficient information to classify patients by grades R0 and R1 or above, M0 and M1.

METHODS : The AI grading system was trained on screening data to evaluate the presence of DR and DME. The core algorithm of the system is a deep learning network that segments relevant anatomical features in a retinal image. Patients were graded according to the standard NHS Diabetic Eye Screening Program feature-based grading protocol.

RESULTS : The algorithm performance was evaluated with a series of 6,981 patient retinal images from routine diabetic eye screenings. It correctly predicted 98.9% of retinopathy events and 95.5% of maculopathy events. Non-disease events prediction rate was 68.6% for retinopathy and 81.2% for maculopathy.

CONCLUSION : This novel deep learning model was trained and tested on patient data from annual diabetic retinopathy screenings can classify with high accuracy the DR and DME status of a person with diabetes. The system can be easily reconfigured according to any grading protocol, without running a long AI training procedure. The incorporation of the AI grading system can increase the graders' productivity and improve the final outcome accuracy of the screening process.

Katz Or, Presil Dan, Cohen Liz, Nachmani Roi, Kirshner Naomi, Hoch Yaacov, Lev Tsvi, Hadad Aviel, Hewitt Richard John, Owens David R

2021-Sep-22

AI, diabetic retinopathy, imaging, screening

Public Health Public Health

Integrating domain knowledge with machine learning to detect obstructive sleep apnea: Snore as a significant bio-feature.

In Journal of sleep research

Our study's main purpose is to emphasise the significance of medical knowledge of pathophysiology before machine learning. We investigated whether combining domain knowledge with machine learning results might increase accuracy and minimise the number of bio-features used to detect obstructive sleep apnea (OSA). The present study analysed data on 36 self-reported symptoms and 24 clinical features obtained from 3,495 patients receiving polysomnography at a regional hospital and a medical centre. The area under the receiver operating characteristic (AUC) curve was used to evaluate patients with and without moderate or severe OSA using three prediction models on the basis of various estimation methods: the multiple logistic regression (MLR), support vector machine (SVM), and neural network (NN) methods. Odds ratios stratified by gender and age were also measured to account for clinicians' common sense. We discovered that adding the self-reported snoring item improved the AUC by 0.01-0.10 and helped us to rapidly achieve the optimum level. The performance of four items (gender, age, body mass index [BMI], and snoring) was comparable with that of adding two or more items (neck and waist circumference) for predicting moderate to severe OSA (Apnea-Hypopnea Index ≥15 events/hr) in all three prediction models, demonstrating the medical knowledge value of pathophysiology. The four-item test sample AUCs were 0.83, 0.84, and 0.83 for MLR, SVM, and NN, respectively. Participants with regular snoring and a BMI of ≥25 kg/m2 had a greater chance of moderate to severe OSA according to the stratified adjusted odds ratios. Combining domain knowledge into machine learning could increase efficiency and enable primary care physicians to refer for an OSA diagnosis earlier.

Hsu Yu-Ching, Wang Jung-Der, Huang Po-Hsien, Chien Yu-Wen, Chiu Ching-Ju, Lin Cheng-Yu

2021-Sep-21

bio-feature, prediction model, prior domain knowledge, self-reported snoring

Surgery Surgery

An Algorithm to Personalize Nerve Sparing in Men with Unilateral High-Risk Prostate Cancer.

In The Journal of urology ; h5-index 80.0

PURPOSE : Current guidelines do not provide strong recommendations on the preservation of the neurovascular bundles during RP in case of HR PCa and/or suspicious EPE. We aimed to evaluate when, in case of unilateral HR disease, contralateral NS should be considered or not.

METHODS : Within a multi-institutional dataset we selected patients with unilateral HR PCa defined as: unilateral EPE and/or SVI on mpMRI or unilateral ISUP 4-5 or PSA ≥20 ng/ml. To evaluate when to perform NS based on the risk of contralateral EPE, we relied on CHAID, a recursive machine learning partitioning algorithm developed to identify risk groups, which was fit to predict the presence of EPE on final pathology, contralaterally to the prostate lobe with HR disease.

RESULTS : 705 patients were identified. Contralateral EPE was documented in 87 (12%) patients. The CHAID identified three groups: i) absence of SVI on mpMRI and index lesion's diameter ≤15 mm; ii) index lesion's diameter ≤15 mm and contralateral ISUP 2-3 or index lesion's diameter >15 mm and negative contralateral biopsy or ISUP 1 iii) SVI on mpMRI or index lesion's diameter >15 mm and contralateral biopsy ISUP 2-3. We named those groups as low- intermediate- and high-risk for contralateral EPE. The rate of EPE and PSMs across the groups were: 4.8%, 14%, 26% and 5.6%, 13%, 18%, respectively.

CONCLUSIONS : Our study challenges current guidelines by proving that wide bilateral excision in men with unilateral HR disease is not justified. Pending external validation, we propose performing NS and incremental NS in case of contralateral low- and intermediate EPE risk, respectively.

Martini Alberto, Soeterik Timo F W, Haverdings Hester, Rahota Razvan George, Checcucci Enrico, De Cillis Sabrina, Hermanns Thomas, Fankhauser Christian Daniel, Afferi Luca, Moschini Marco, Mattei Agostino, Kesch Claudia, Heidegger Isabel, Preisser Felix, Zattoni Fabio, Marquis Alessandro, Marra Giancarlo, Gontero Paolo, Briganti Alberto, Montorsi Francesco, Porpiglia Francesco, Van Basten Jean Paul, Van den Bergh Roderick C N, Van Melick Harm H E, Ploussard Guillaume, Gandaglia Giorgio, Valerio Massimo

2021-Sep-22

MRI., nerve sparing, prostate cancer, robot-assisted surgery

General General

Modeling and forecasting the COVID-19 pandemic time-series data.

In Social science quarterly

Objective : We analyze the number of recorded cases and deaths of COVID-19 in many parts of the world, with the aim to understand the complexities of the data, and produce regular forecasts.

Methods : The SARS-CoV-2 virus that causes COVID-19 has affected societies in all corners of the globe but with vastly differing experiences across countries. Health-care and economic systems vary significantly across countries, as do policy responses, including testing, intermittent lockdowns, quarantine, contact tracing, mask wearing, and social distancing. Despite these challenges, the reported data can be used in many ways to help inform policy. We describe how to decompose the reported time series of confirmed cases and deaths into a trend, seasonal, and irregular component using machine learning methods.

Results : This decomposition enables statistical computation of measures of the mortality ratio and reproduction number for any country, and we conduct a counterfactual exercise assuming that the United States had a summer outcome in 2020 similar to that of the European Union. The decomposition is also used to produce forecasts of cases and deaths, and we undertake a forecast comparison which highlights the importance of seasonality in the data and the difficulties of forecasting too far into the future.

Conclusion : Our adaptive data-based methods and purely statistical forecasts provide a useful complement to the output from epidemiological models.

Doornik Jurgen A, Castle Jennifer L, Hendry David F

2021-Aug-07

Covid‐19, epidemiology, nonstationarity, reproduction number, time‐series forecasting

General General

Comparing classification models-a practical tutorial.

In Journal of computer-aided molecular design

While machine learning models have become a mainstay in Cheminformatics, the field has yet to agree on standards for model evaluation and comparison. In many cases, authors compare methods by performing multiple folds of cross-validation and reporting the mean value for an evaluation metric such as the area under the receiver operating characteristic. These comparisons of mean values often lack statistical rigor and can lead to inaccurate conclusions. In the interest of encouraging best practices, this tutorial provides an example of how multiple methods can be compared in a statistically rigorous fashion.

Patrick Walters W

2021-Sep-22

Classification model, Machine learning, QSAR, Statistical validation, Tutorial

Radiology Radiology

Effect of CT image acquisition parameters on diagnostic performance of radiomics in predicting malignancy of pulmonary nodules of different sizes.

In European radiology ; h5-index 62.0

OBJECTIVES : To investigate the effect of CT image acquisition parameters on the performance of radiomics in classifying benign and malignant pulmonary nodules (PNs) with respect to nodule size.

METHODS : We retrospectively collected CT images of 696 patients with PNs from March 2015 to March 2018. PNs were grouped by nodule diameter: T1a (diameter ≤ 1.0 cm), T1b (1.0 cm < diameter ≤ 2.0 cm), and T1c (2.0 cm < diameter ≤ 3.0 cm). CT images were divided into four settings according to slice-thickness-convolution-kernels: setting 1 (slice thickness/reconstruction type: 1.25 mm sharp), setting 2 (5 mm sharp), setting 3 (5 mm smooth), and random setting. We created twelve groups from two interacting conditions. Each PN was segmented and had 1160 radiomics features extracted. Non-redundant features with high predictive ability in training were selected to build a distinct model under each of the twelve subsets.

RESULTS : The performance (AUCs) on predicting PN malignancy were as follows: T1a group: 0.84, 0.64, 0.68, and 0.68; T1b group: 0.68, 0.74, 0.76, and 0.70; T1c group: 0.66, 0.64, 0.63, and 0.70, for the setting 1, setting 2, setting 3, and random setting, respectively. In the T1a group, the AUC of radiomics model in setting 1 was statistically significantly higher than all others; In the T1b group, AUCs of radiomics models in setting 3 were statistically significantly higher than some; and in the T1c group, there were no statistically significant differences among models.

CONCLUSIONS : For PNs less than 1 cm, CT image acquisition parameters have a significant influence on diagnostic performance of radiomics in predicting malignancy, and a model created using images reconstructed with thin section and a sharp kernel algorithm achieved the best performance. For PNs larger than 1 cm, CT reconstruction parameters did not affect diagnostic performance substantially.

KEY POINTS : • CT image acquisition parameters have a significant influence on the diagnostic performance of radiomics in pulmonary nodules less than 1 cm. • In pulmonary nodules less than 1 cm, a radiomics model created by using images reconstructed with thin section and a sharp kernel algorithm achieved the best diagnostic performance. • For PNs larger than 1 cm, CT image acquisition parameters do not affect diagnostic performance substantially.

Xu Yan, Lu Lin, Sun Shawn H, E Lin-Ning, Lian Wei, Yang Hao, Schwartz Lawrence H, Yang Zheng-Han, Zhao Binsheng

2021-Sep-21

Carcinoma, non-small-cell lung, Diagnostic screening programs, Machine learning, Tomography, x-ray computed

General General

Systematic Mapping Study of AI/Machine Learning in Healthcare and Future Directions.

In SN computer science

This study attempts to categorise research conducted in the area of: use of machine learning in healthcare, using a systematic mapping study methodology. In our attempt, we reviewed literature from top journals, articles, and conference papers by using the keywords use of machine learning in healthcare. We queried Google Scholar, resulted in 1400 papers, and then categorised the results on the basis of the objective of the study, the methodology adopted, type of problem attempted and disease studied. As a result we were able to categorize study in five different categories namely, interpretable ML, evaluation of medical images, processing of EHR, security/privacy framework, and transfer learning. In the study we also found that most of the authors have studied cancer, and one of the least studied disease was epilepsy, evaluation of medical images is the most researched and a new field of research, Interpretable ML/Explainable AI, is gaining momentum. Our basic intent is to provide a fair idea to future researchers about the field and future directions.

Parashar Gaurav, Chaudhary Alka, Rana Ajay

2021

Electronic health records (EHR), Healthcare, Interpretable ML, Machine learning (ML), Privacy framework, Security framework, Transfer learning (TL)

General General

Prediction Model of Clearance by a Novel Quantitative Structure-Activity Relationship Approach, Combination DeepSnap-Deep Learning and Conventional Machine Learning.

In ACS omega

Some targets predicted by machine learning (ML) in drug discovery remain a challenge because of poor prediction. In this study, a new prediction model was developed and rat clearance (CL) was selected as a target because it is difficult to predict. A classification model was constructed using 1545 in-house compounds with rat CL data. The molecular descriptors calculated by Molecular Operating Environment (MOE), alvaDesc, and ADMET Predictor software were used to construct the prediction model. In conventional ML using 100 descriptors and random forest selected by DataRobot, the area under the curve (AUC) and accuracy (ACC) were 0.883 and 0.825, respectively. Conversely, the prediction model using DeepSnap and Deep Learning (DeepSnap-DL) with compound features as images had AUC and ACC of 0.905 and 0.832, respectively. We combined the two models (conventional ML and DeepSnap-DL) to develop a novel prediction model. Using the ensemble model with the mean of the predicted probabilities from each model improved the evaluation metrics (AUC = 0.943 and ACC = 0.874). In addition, a consensus model using the results of the agreement between classifications had an increased ACC (0.959). These combination models with a high level of predictive performance can be applied to rat CL as well as other pharmacokinetic parameters, pharmacological activity, and toxicity prediction. Therefore, these models will aid in the design of more rational compounds for the development of drugs.

Mamada Hideaki, Nomura Yukihiro, Uesawa Yoshihiro

2021-Sep-14

General General

Impact of a Diverse Combination of Metal Oxide Gas Sensors on Machine Learning-Based Gas Recognition in Mixed Gases.

In ACS omega

A challenge for chemiresistive-type gas sensors distinguishing mixture gases is that for highly accurate recognition, massive data processing acquired from various types of sensor configurations must be considered. The impact of data processing is indeed ineffective and time-consuming. Herein, we systemically investigate the effect of the selectivity for a target gas on the prediction accuracy of gas concentration via machine learning based on a support vector machine model. The selectivity factor S(X) of a gas sensor for a target gas "X" is introduced to reveal the correlation between the prediction accuracy and selectivity of gas sensors. The presented work suggests that (i) the strong correlation between the selectivity factor and prediction accuracy has a proportional relationship, (ii) the enhancement of the prediction accuracy of an elemental sensor with a low sensitivity factor can be attained by a complementary combination of the other sensor with a high selectivity factor, and (iii) it can also be boosted by combining the sensor having even a low selectivity factor.

Bae Garam, Kim Minji, Song Wooseok, Myung Sung, Lee Sun Sook, An Ki-Seok

2021-Sep-14

General General

A novel ensemble deep learning model for stock prediction based on stock prices and news.

In International journal of data science and analytics

In recent years, machine learning and deep learning have become popular methods for financial data analysis, including financial textual data, numerical data, and graphical data. One of the most popular and complex deep learning in finance topics is future stock prediction. The difficulty that causes the future stock forecast is that there are too many different factors that affect the amplitude and frequency of the rise and fall of stocks at the same time. Some of the company-specific factors that can affect the share price like news releases on earnings and profits, future estimated earnings, the announcement of dividends, introduction of a new product or a product recall, secure a new large contract, employee layoffs, a major change of management, anticipated takeover or merger, and accounting errors or scandals. Furthermore, these factors are only company factors, and other factors affect the future trend of stocks, such as industry performance, investor sentiment, and economic factors. This paper proposes a novel deep learning approach to predict future stock movement. The model employs a blending ensemble learning method to combine two recurrent neural networks, followed by a fully connected neural network. In our research, we use the S&P 500 Index as our test case. Our experiments show that our blending ensemble deep learning model outperforms the best existing prediction model substantially using the same dataset, reducing the mean-squared error from 438.94 to 186.32, a 57.55% reduction, increasing precision rate by 40%, recall by 50%, F1-score by 44.78%, and movement direction accuracy by 33.34%, respectively. The purpose of this work is to explain our design philosophy and show that ensemble deep learning technologies can truly predict future stock price trends more effectively and can better assist investors in making the right investment decision than other traditional methods.

Li Yang, Pan Yi

2021-Sep-17

Deep learning, Ensemble learning, Machine learning, Statistical finance, Stock prediction

General General

COV-VGX: An automated COVID-19 detection system using X-ray images and transfer learning.

In Informatics in medicine unlocked

Coronavirus (COVID-19) has been one of the most dangerous and acute deadly diseases across the world recently. Researchers are trying to develop automated and feasible COVID-19 detection systems with the help of deep neural networks, machine learning techniques, etc. In this paper, a deep learning-based COVID-19 detection system called COV-VGX is proposed that contributes to detecting coronavirus disease automatically using chest X-ray images. The system introduces two types of classifiers, namely, a multiclass classifier that automatically predicts coronavirus, pneumonia, and normal classes and a binary classifier that predicts coronavirus and pneumonia classes. Using transfer learning, a deep CNN model is proposed to extract distinct and high-level features from X-ray images in collaboration with the pretrained model VGG-16. Despite the limitation of the COVID-19 dataset, the model is evaluated with sufficient COVID-19 images. Extensive experiments for multiclass classifier have achieved 98.91% accuracy, 97.31% precision, 99.50% recall, 98.39% F1-score, while 99.37% accuracy, 98.76% precision, 100% recall, 99.38% F1-score for binary classifier. The proposed system can contribute a lot in diagnosing COVID-19 effectively in the medical field.

Saha Prottoy, Sadi Muhammad Sheikh, Aranya O F M Riaz Rahman, Jahan Sadia, Islam Ferdib-Al

2021

COVID-19, Deep learning, Transfer learning, VGG-16, X-ray images

Public Health Public Health

Estimating the Impact of COVID-19 on the PM2.5 Levels in China with a Satellite-Driven Machine Learning Model.

In Remote sensing

China implemented an aggressive nationwide lockdown procedure immediately after the COVID-19 outbreak in January 2020. As China emerges from the impact of COVID-19 on national economic and industrial activities, it has become the site of a large-scale natural experiment to evaluate the impact of COVID-19 on regional air quality. However, ground measurements of fine particulate matters (PM2.5) concentrations do not offer comprehensive spatial coverage, especially in suburban and rural regions. In this study, we developed a machine learning method with satellite aerosol remote sensing data, meteorological fields and land use parameters as major predictor variables to estimate spatiotemporally resolved daily PM2.5 concentrations in China. Our study period consists of a reference semester (1 November 2018-30 April 2019) and a pandemic semester (1 November 2019-30 April 2020), with six modeling months in each semester. Each period was then divided into subperiod 1 (November and December), subperiod 2 (January and February) and subperiod 3 (March and April). The reference semester model obtained a 10-fold cross-validated R2 (RMSE) of 0.79 (17.55 μg/m3) and the pandemic semester model obtained a 10-fold cross-validated R2 (RMSE) of 0.83 (13.48 μg/m3) for daily PM2.5 predictions. Our prediction results showed high PM2.5 concentrations in the North China Plain, Yangtze River Delta, Sichuan Basin and Xinjiang Autonomous Region during the reference semester. PM2.5 levels were lowered by 4.8 μg/m3 during the pandemic semester compared to the reference semester and PM2.5 levels during subperiod 2 decreased most, by 18%. The southeast region was affected most by the COVID-19 outbreak with PM2.5 levels during subperiod 2 decreasing by 31%, followed by the Northern Yangtze River Delta (29%) and Pearl River Delta (24%).

Li Qiulun, Zhu Qingyang, Xu Muwu, Zhao Yu, Narayan K M Venkat, Liu Yang

2021-Apr

COVID-19, China, MAIAC AOD, PM2.5, air pollution, machine learning, random forest, remote sensing

oncology Oncology

Novel insights into the molecular mechanisms underlying risk of colorectal cancer from smoking and red/processed meat carcinogens by modeling exposure in normal colon organoids.

In Oncotarget ; h5-index 104.0

Tobacco smoke and red/processed meats are well-known risk factors for colorectal cancer (CRC). Most research has focused on studies of normal colon biopsies in epidemiologic studies or treatment of CRC cell lines in vitro. These studies are often constrained by challenges with accuracy of self-report data or, in the case of CRC cell lines, small sample sizes and lack of relationship to normal tissue at risk. In an attempt to address some of these limitations, we performed a 24-hour treatment of a representative carcinogens cocktail in 37 independent organoid lines derived from normal colon biopsies. Machine learning algorithms were applied to bulk RNA-sequencing and revealed cellular composition changes in colon organoids. We identified 738 differentially expressed genes in response to carcinogens exposure. Network analysis identified significantly different modules of co-expression, that included genes related to MSI-H tumor biology, and genes previously implicated in CRC through genome-wide association studies. Our study helps to better define the molecular effects of representative carcinogens from smoking and red/processed meat in normal colon epithelial cells and in the etiology of the MSI-H subtype of CRC, and suggests an overlap between molecular mechanisms involved in inherited and environmental CRC risk.

Devall Matthew, Dampier Christopher H, Eaton Stephen, Ali Mourad W, Díez-Obrero Virginia, Moratalla-Navarro Ferran, Bryant Jennifer, Jennelle Lucas T, Moreno Victor, Powell Steven M, Peters Ulrike, Casey Graham

2021-Sep-14

colon organoids, microsatellite instability, single-cell deconvolution, smoking, weighted gene co-expression network analysis

General General

Transfer learning for image classification using VGG19: Caltech-101 image data set.

In Journal of ambient intelligence and humanized computing

Image classification is getting more attention in the area of computer vision. During the past few years, a lot of research has been done on image classification using classical machine learning and deep learning techniques. Presently, deep learning-based techniques have given stupendous results. The performance of a classification system depends on the quality of features extracted from an image. The better is the quality of extracted features, the more the accuracy will be. Although, numerous deep learning-based methods have shown enormous performance in image classification, still due to various challenges deep learning methods are not able to extract all the important information from the image. This results in a reduction in overall classification accuracy. The goal of the present research is to improve the image classification performance by combining the deep features extracted using popular deep convolutional neural network, VGG19, and various handcrafted feature extraction methods, i.e., SIFT, SURF, ORB, and Shi-Tomasi corner detector algorithm. Further, the extracted features from these methods are classified using various machine learning classification methods, i.e., Gaussian Naïve Bayes, Decision Tree, Random Forest, and eXtreme Gradient Boosting (XGBClassifier) classifier. The experiment is carried out on a benchmark dataset Caltech-101. The experimental results indicate that Random Forest using the combined features give 93.73% accuracy and outperforms other classifiers and methods proposed by other authors. The paper concludes that a single feature extractor whether shallow or deep is not enough to achieve satisfactory results. So, a combined approach using deep learning features and traditional handcrafted features is better for image classification.

Bansal Monika, Kumar Munish, Sachdeva Monika, Mittal Ajay

2021-Sep-17

K-Means, LPP, ORB, PCA, SIFT, SURF

General General

Pharmacoprint: A Combination of a Pharmacophore Fingerprint and Artificial Intelligence as a Tool for Computer-Aided Drug Design.

In Journal of chemical information and modeling

Structural fingerprints and pharmacophore modeling are methodologies that have been used for at least 2 decades in various fields of cheminformatics, from similarity searching to machine learning (ML). Advances in in silico techniques consequently led to combining both these methodologies into a new approach known as the pharmacophore fingerprint. Herein, we propose a high-resolution, pharmacophore fingerprint called Pharmacoprint that encodes the presence, types, and relationships between pharmacophore features of a molecule. Pharmacoprint was evaluated in classification experiments by using ML algorithms (logistic regression, support vector machines, linear support vector machines, and neural networks) and outperformed other popular molecular fingerprints (i.e., ECFP4, Estate, MACCS, PubChem, Substructure, Klekota-Roth, CDK, Extended, and GraphOnly) and the ChemAxon pharmacophoric features fingerprint. Pharmacoprint consisted of 39 973 bits; several methods were applied for dimensionality reduction, and the best algorithm not only reduced the length of the bit string but also improved the efficiency of the ML tests. Further optimization allowed us to define the best parameter settings for using Pharmacoprint in discrimination tests and for maximizing statistical parameters. Finally, Pharmacoprint generated for three-dimensional (3D) structures with defined hydrogens as input data was applied to neural networks with a supervised autoencoder for selecting the most important bits and allowed us to maximize the Matthews correlation coefficient up to 0.962. The results show the potential of Pharmacoprint as a new, perspective tool for computer-aided drug design.

Warszycki Dawid, Struski Łukasz, Śmieja Marek, Kafel Rafał, Kurczab Rafał

2021-Sep-21

General General

Loan default prediction of Chinese P2P market: a machine learning methodology.

In Scientific reports ; h5-index 158.0

Repayment failures of borrowers have greatly affected the sustainable development of the peer-to-peer (P2P) lending industry. The latest literature reveals that existing risk evaluation systems may ignore important signals and risk factors affecting P2P repayment. In our study, we applied four machine learning methods (random forest (RF), extreme gradient boosting tree (XGBT), gradient boosting model (GBM), and neural network (NN)) to predict important factors affecting repayment by utilizing data from Renrendai.com in China from Thursday, January 1, 2015, to Tuesday, June 30, 2015. The results showed that borrowers who have passed video, mobile phone, job, residence or education level verification are more likely to default on loan repayment, whereas those who have passed identity and asset certification are less likely to default on loans. The accuracy and kappa value of the four methods all exceed 90%, and RF is superior to the other classification models. Our findings demonstrate important techniques for borrower screening by P2P companies and risk regulation by regulatory agencies. Our methodology and findings will help regulators, banks and creditors combat current financial disasters caused by the coronavirus disease 2019 (COVID-19) pandemic by addressing various financial risks and translating credit scoring improvements.

Xu Junhui, Lu Zekai, Xie Ying

2021-Sep-21

oncology Oncology

Image Features of Magnetic Resonance Imaging under the Deep Learning Algorithm in the Diagnosis and Nursing of Malignant Tumors.

In Contrast media & molecular imaging

In order to explore the effect of convolutional neural network (CNN) algorithm based on deep learning on magnetic resonance imaging (MRI) images of brain tumor patients and evaluate the practical value of MRI image features based on deep learning algorithm in the clinical diagnosis and nursing of malignant tumors, in this study, a brain tumor MRI image model based on the CNN algorithm was constructed, and 80 patients with brain tumors were selected as the research objects. They were divided into an experimental group (CNN algorithm) and a control group (traditional algorithm). The patients were nursed in the whole process. The macroscopic characteristics and imaging index of the MRI image and anxiety of patients in two groups were compared and analyzed. In addition, the image quality after nursing was checked. The results of the study revealed that the MRI characteristics of brain tumors based on CNN algorithm were clearer and more accurate in the fluid-attenuated inversion recovery (FLAIR), MRI T1, T1c, and T2; in terms of accuracy, sensitivity, and specificity, the mean value was 0.83, 0.84, and 0.83, which had obvious advantages compared with the traditional algorithm (P < 0.05). The patients in the nursing group showed lower depression scores and better MRI images in contrast to the control group (P < 0.05). Therefore, the deep learning algorithm can further accurately analyze the MRI image characteristics of brain tumor patients on the basis of conventional algorithms, showing high sensitivity and specificity, which improved the application value of MRI image characteristics in the diagnosis of malignant tumors. In addition, effective nursing for patients undergoing analysis and diagnosis on brain tumor MRI image characteristics can alleviate the patient's anxiety and ensure that high-quality MRI images were obtained after the examination.

Sun Lifang, Hu Xi, Liu Yutao, Cai Hengyu

2021

General General

Going Beyond Saliency Maps: Training Deep Models to Interpret Deep Models.

In Information processing in medical imaging : proceedings of the ... conference

Interpretability is a critical factor in applying complex deep learning models to advance the understanding of brain disorders in neuroimaging studies. To interpret the decision process of a trained classifier, existing techniques typically rely on saliency maps to quantify the voxel-wise or feature-level importance for classification through partial derivatives. Despite providing some level of localization, these maps are not human-understandable from the neuroscience perspective as they often do not inform the specific type of morphological changes linked to the brain disorder. Inspired by the image-to-image translation scheme, we propose to train simulator networks to inject (or remove) patterns of the disease into a given MRI based on a warping operation, such that the classifier increases (or decreases) its confidence in labeling the simulated MRI as diseased. To increase the robustness of training, we propose to couple the two simulators into a unified model based on conditional convolution. We applied our approach to interpreting classifiers trained on a synthetic dataset and two neuroimaging datasets to visualize the effect of Alzheimer's disease and alcohol dependence. Compared to the saliency maps generated by baseline approaches, our simulations and visualizations based on the Jacobian determinants of the warping field reveal meaningful and understandable patterns related to the diseases.

Liu Zixuan, Adeli Ehsan, Pohl Kilian M, Zhao Qingyu

2021-Jun

oncology Oncology

High-Dimensional Precision Medicine From Patient-Derived Xenografts.

In Journal of the American Statistical Association

The complexity of human cancer often results in significant heterogeneity in response to treatment. Precision medicine offers the potential to improve patient outcomes by leveraging this heterogeneity. Individualized treatment rules (ITRs) formalize precision medicine as maps from the patient covariate space into the space of allowable treatments. The optimal ITR is that which maximizes the mean of a clinical outcome in a population of interest. Patient-derived xenograft (PDX) studies permit the evaluation of multiple treatments within a single tumor, and thus are ideally suited for estimating optimal ITRs. PDX data are characterized by correlated outcomes, a high-dimensional feature space, and a large number of treatments. Here we explore machine learning methods for estimating optimal ITRs from PDX data. We analyze data from a large PDX study to identify biomarkers that are informative for developing personalized treatment recommendations in multiple cancers. We estimate optimal ITRs using regression-based (Q-learning) and direct-search methods (outcome weighted learning). Finally, we implement a superlearner approach to combine multiple estimated ITRs and show that the resulting ITR performs better than any of the input ITRs, mitigating uncertainty regarding user choice. Our results indicate that PDX data are a valuable resource for developing individualized treatment strategies in oncology. Supplementary materials for this article are available online.

Rashid Naim U, Luckett Daniel J, Chen Jingxiang, Lawson Michael T, Wang Longshaokan, Zhang Yunshu, Laber Eric B, Liu Yufeng, Yeh Jen Jen, Zeng Donglin, Kosorok Michael R

2021

Biomarkers, Deep learning autoencoders, Machine learning, Outcome weighted learning, Precision medicine, Q-learning

General General

Modeling and forecasting the COVID-19 pandemic time-series data.

In Social science quarterly

Objective : We analyze the number of recorded cases and deaths of COVID-19 in many parts of the world, with the aim to understand the complexities of the data, and produce regular forecasts.

Methods : The SARS-CoV-2 virus that causes COVID-19 has affected societies in all corners of the globe but with vastly differing experiences across countries. Health-care and economic systems vary significantly across countries, as do policy responses, including testing, intermittent lockdowns, quarantine, contact tracing, mask wearing, and social distancing. Despite these challenges, the reported data can be used in many ways to help inform policy. We describe how to decompose the reported time series of confirmed cases and deaths into a trend, seasonal, and irregular component using machine learning methods.

Results : This decomposition enables statistical computation of measures of the mortality ratio and reproduction number for any country, and we conduct a counterfactual exercise assuming that the United States had a summer outcome in 2020 similar to that of the European Union. The decomposition is also used to produce forecasts of cases and deaths, and we undertake a forecast comparison which highlights the importance of seasonality in the data and the difficulties of forecasting too far into the future.

Conclusion : Our adaptive data-based methods and purely statistical forecasts provide a useful complement to the output from epidemiological models.

Doornik Jurgen A, Castle Jennifer L, Hendry David F

2021-Aug-07

Covid‐19, epidemiology, nonstationarity, reproduction number, time‐series forecasting

General General

Patient apprehensions about the use of artificial intelligence in healthcare.

In NPJ digital medicine

While there is significant enthusiasm in the medical community about the use of artificial intelligence (AI) technologies in healthcare, few research studies have sought to assess patient perspectives on these technologies. We conducted 15 focus groups examining patient views of diverse applications of AI in healthcare. Our results indicate that patients have multiple concerns, including concerns related to the safety of AI, threats to patient choice, potential increases in healthcare costs, data-source bias, and data security. We also found that patient acceptance of AI is contingent on mitigating these possible harms. Our results highlight an array of patient concerns that may limit enthusiasm for applications of AI in healthcare. Proactively addressing these concerns is critical for the flourishing of ethical innovation and ensuring the long-term success of AI applications in healthcare.

Richardson Jordan P, Smith Cambray, Curtis Susan, Watson Sara, Zhu Xuan, Barry Barbara, Sharp Richard R

2021-Sep-21

Ophthalmology Ophthalmology

Development and clinical deployment of a smartphone-based visual field deep learning system for glaucoma detection.

In NPJ digital medicine

By 2040, ~100 million people will have glaucoma. To date, there are a lack of high-efficiency glaucoma diagnostic tools based on visual fields (VFs). Herein, we develop and evaluate the performance of 'iGlaucoma', a smartphone application-based deep learning system (DLS) in detecting glaucomatous VF changes. A total of 1,614,808 data points of 10,784 VFs (5542 patients) from seven centers in China were included in this study, divided over two phases. In Phase I, 1,581,060 data points from 10,135 VFs of 5105 patients were included to train (8424 VFs), validate (598 VFs) and test (3 independent test sets-200, 406, 507 samples) the diagnostic performance of the DLS. In Phase II, using the same DLS, iGlaucoma cloud-based application further tested on 33,748 data points from 649 VFs of 437 patients from three glaucoma clinics. With reference to three experienced expert glaucomatologists, the diagnostic performance (area under curve [AUC], sensitivity and specificity) of the DLS and six ophthalmologists were evaluated in detecting glaucoma. In Phase I, the DLS outperformed all six ophthalmologists in the three test sets (AUC of 0.834-0.877, with a sensitivity of 0.831-0.922 and a specificity of 0.676-0.709). In Phase II, iGlaucoma had 0.99 accuracy in recognizing different patterns in pattern deviation probability plots region, with corresponding AUC, sensitivity and specificity of 0.966 (0.953-0.979), 0.954 (0.930-0.977), and 0.873 (0.838-0.908), respectively. The 'iGlaucoma' is a clinically effective glaucoma diagnostic tool to detect glaucoma from humphrey VFs, although the target population will need to be carefully identified with glaucoma expertise input.

Li Fei, Song Diping, Chen Han, Xiong Jian, Li Xingyi, Zhong Hua, Tang Guangxian, Fan Sujie, Lam Dennis S C, Pan Weihua, Zheng Yajuan, Li Ying, Qu Guoxiang, He Junjun, Wang Zhe, Jin Ling, Zhou Rouxi, Song Yunhe, Sun Yi, Cheng Weijing, Yang Chunman, Fan Yazhi, Li Yingjie, Zhang Hengli, Yuan Ye, Xu Yang, Xiong Yunfan, Jin Lingfei, Lv Aiguo, Niu Lingzhi, Liu Yuhong, Li Shaoli, Zhang Jiani, Zangwill Linda M, Frangi Alejandro F, Aung Tin, Cheng Ching-Yu, Qiao Yu, Zhang Xiulan, Ting Daniel S W

2020-Sep-22

General General

Loan default prediction of Chinese P2P market: a machine learning methodology.

In Scientific reports ; h5-index 158.0

Repayment failures of borrowers have greatly affected the sustainable development of the peer-to-peer (P2P) lending industry. The latest literature reveals that existing risk evaluation systems may ignore important signals and risk factors affecting P2P repayment. In our study, we applied four machine learning methods (random forest (RF), extreme gradient boosting tree (XGBT), gradient boosting model (GBM), and neural network (NN)) to predict important factors affecting repayment by utilizing data from Renrendai.com in China from Thursday, January 1, 2015, to Tuesday, June 30, 2015. The results showed that borrowers who have passed video, mobile phone, job, residence or education level verification are more likely to default on loan repayment, whereas those who have passed identity and asset certification are less likely to default on loans. The accuracy and kappa value of the four methods all exceed 90%, and RF is superior to the other classification models. Our findings demonstrate important techniques for borrower screening by P2P companies and risk regulation by regulatory agencies. Our methodology and findings will help regulators, banks and creditors combat current financial disasters caused by the coronavirus disease 2019 (COVID-19) pandemic by addressing various financial risks and translating credit scoring improvements.

Xu Junhui, Lu Zekai, Xie Ying

2021-Sep-21

Cardiology Cardiology

Deep learning-based prediction of early cerebrovascular events after transcatheter aortic valve replacement.

In Scientific reports ; h5-index 158.0

Cerebrovascular events (CVE) are among the most feared complications of transcatheter aortic valve replacement (TAVR). CVE appear difficult to predict due to their multifactorial origin incompletely explained by clinical predictors. We aimed to build a deep learning-based predictive tool for TAVR-related CVE. Integrated clinical and imaging characteristics from consecutive patients enrolled into a prospective TAVR registry were analysed. CVE comprised any strokes and transient ischemic attacks. Predictive variables were selected by recursive feature reduction to train an autoencoder predictive model. Area under the curve (AUC) represented the model's performance to predict 30-day CVE. Among 2279 patients included between 2007 and 2019, both clinical and imaging data were available in 1492 patients. Median age was 83 years and STS score was 4.6%. Acute (< 24 h) and subacute (day 2-30) CVE occurred in 19 (1.3%) and 36 (2.4%) patients, respectively. The occurrence of CVE was associated with an increased risk of death (HR [95% CI] 2.62 [1.82-3.78]). The constructed predictive model uses less than 107 clinical and imaging variables and has an AUC of 0.79 (0.65-0.93). TAVR-related CVE can be predicted using a deep learning-based predictive algorithm. The model is implemented online for broad usage.

Okuno Taishi, Overtchouk Pavel, Asami Masahiko, Tomii Daijiro, Stortecky Stefan, Praz Fabien, Lanz Jonas, Siontis George C M, Gräni Christoph, Windecker Stephan, Pilgrim Thomas

2021-Sep-21

Pathology Pathology

A machine learning interpretation of the contribution of foliar fungicides to soybean yield in the north-central United States.

In Scientific reports ; h5-index 158.0

Foliar fungicide usage in soybeans in the north-central United States increased steadily over the past two decades. An agronomically-interpretable machine learning framework was used to understand the importance of foliar fungicides relative to other factors associated with realized soybean yields, as reported by growers surveyed from 2014 to 2016. A database of 2738 spatially referenced fields (of which 30% had been sprayed with foliar fungicides) was fit to a random forest model explaining soybean yield. Latitude (a proxy for unmeasured agronomic factors) and sowing date were the two most important factors associated with yield. Foliar fungicides ranked 7th out of 20 factors in terms of relative importance. Pairwise interactions between latitude, sowing date and foliar fungicide use indicated more yield benefit to using foliar fungicides in late-planted fields and in lower latitudes. There was a greater yield response to foliar fungicides in higher-yield environments, but less than a 100 kg/ha yield penalty for not using foliar fungicides in such environments. Except in a few production environments, yield gains due to foliar fungicides sufficiently offset the associated costs of the intervention when soybean prices are near-to-above average but do not negate the importance of disease scouting and fungicide resistance management.

Shah Denis A, Butts Thomas R, Mourtzinis Spyridon, Rattalino Edreira Juan I, Grassini Patricio, Conley Shawn P, Esker Paul D

2021-Sep-21

General General

Prediction of coating thickness for polyelectrolyte multilayers via machine learning.

In Scientific reports ; h5-index 158.0

Layer-by-layer (LbL) deposition method of polyelectrolytes is a versatile way of developing functional nanoscale coatings. Even though the mechanisms of LbL film development are well-established, currently there are no predictive models that can link film components with their final properties. The current health crisis has shown the importance of accelerated development of biomedical solutions such as antiviral coatings, and the implementation of machine learning methodologies for coating development can enable achieving this. In this work, using literature data and newly generated experimental results, we first analyzed the relative impact of 23 coating parameters on the coating thickness. Next, a predictive model has been developed using aforementioned parameters and molecular descriptors of polymers from the DeepChem library. Model performance was limited because of insufficient number of data points in the training set, due to the scarce availability of data in the literature. Despite this limitation, we demonstrate, for the first time, utilization of machine learning for prediction of LbL coating properties. It can decrease the time necessary to obtain functional coating with desired properties, as well as decrease experimental costs and enable the fast first response to crisis situations (such as pandemics) where coatings can positively contribute. Besides coating thickness, which was selected as an output value in this study, machine learning approach can be potentially used to predict functional properties of multilayer coatings, e.g. biocompatibility, cell adhesive, antibacterial, antiviral or anti-inflammatory properties.

Gribova Varvara, Navalikhina Anastasiia, Lysenko Oleksandr, Calligaro Cynthia, Lebaudy Eloïse, Deiber Lucie, Senger Bernard, Lavalle Philippe, Vrana Nihal Engin

2021-Sep-21

General General

Segmentation of Static and Dynamic Atomic-Resolution Microscopy Data Sets with Unsupervised Machine Learning Using Local Symmetry Descriptors.

In Microscopy and microanalysis : the official journal of Microscopy Society of America, Microbeam Analysis Society, Microscopical Society of Canada

We present an unsupervised machine learning approach for segmentation of static and dynamic atomic-resolution microscopy data sets in the form of images and video sequences. In our approach, we first extract local features via symmetry operations. Subsequent dimension reduction and clustering analysis are performed in feature space to assign pattern labels to each pixel. Furthermore, we propose the stride and upsampling scheme as well as separability analysis to speed up the segmentation process of image sequences. We apply our approach to static atomic-resolution scanning transmission electron microscopy images and video sequences. Our code is released as a python module that can be used as a standalone program or as a plugin to other microscopy packages.

Wang Ning, Freysoldt Christoph, Zhang Siyuan, Liebscher Christian H, Neugebauer Jörg

2021-Sep-21

HAADF-STEM, segmentation, symmetry descriptors, unsupervised learning

Public Health Public Health

Using informative features in machine learning based method for COVID-19 drug repurposing.

In Journal of cheminformatics

Coronavirus disease 2019 (COVID-19) is caused by a novel virus named Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2). This virus induced a large number of deaths and millions of confirmed cases worldwide, creating a serious danger to public health. However, there are no specific therapies or drugs available for COVID-19 treatment. While new drug discovery is a long process, repurposing available drugs for COVID-19 can help recognize treatments with known clinical profiles. Computational drug repurposing methods can reduce the cost, time, and risk of drug toxicity. In this work, we build a graph as a COVID-19 related biological network. This network is related to virus targets or their associated biological processes. We select essential proteins in the constructed biological network that lead to a major disruption in the network. Our method from these essential proteins chooses 93 proteins related to COVID-19 pathology. Then, we propose multiple informative features based on drug-target and protein-protein interaction information. Through these informative features, we find five appropriate clusters of drugs that contain some candidates as potential COVID-19 treatments. To evaluate our results, we provide statistical and clinical evidence for our candidate drugs. From our proposed candidate drugs, 80% of them were studied in other studies and clinical trials.

Aghdam Rosa, Habibi Mahnaz, Taheri Golnaz

2021-Sep-20

Clustering method, Coronavirus disease 2019, Protein−protein interaction, SARS-CoV-2

General General

A hybrid framework for improving uncertainty quantification in deep learning-based QSAR regression modeling.

In Journal of cheminformatics

Reliable uncertainty quantification for statistical models is crucial in various downstream applications, especially for drug design and discovery where mistakes may incur a large amount of cost. This topic has therefore absorbed much attention and a plethora of methods have been proposed over the past years. The approaches that have been reported so far can be mainly categorized into two classes: distance-based approaches and Bayesian approaches. Although these methods have been widely used in many scenarios and shown promising performance with their distinct superiorities, being overconfident on out-of-distribution examples still poses challenges for the deployment of these techniques in real-world applications. In this study we investigated a number of consensus strategies in order to combine both distance-based and Bayesian approaches together with post-hoc calibration for improved uncertainty quantification in QSAR (Quantitative Structure-Activity Relationship) regression modeling. We employed a set of criteria to quantitatively assess the ranking and calibration ability of these models. Experiments based on 24 bioactivity datasets were designed to make critical comparison between the model we proposed and other well-studied baseline models. Our findings indicate that the hybrid framework proposed by us can robustly enhance the model ability of ranking absolute errors. Together with post-hoc calibration on the validation set, we show that well-calibrated uncertainty quantification results can be obtained in domain shift settings. The complementarity between different methods is also conceptually analyzed.

Wang Dingyan, Yu Jie, Chen Lifan, Li Xutong, Jiang Hualiang, Chen Kaixian, Zheng Mingyue, Luo Xiaomin

2021-Sep-20

Applicability domain, Artificial intelligence, Bayesian inference, Bayesian neural network, Error prediction, Quantitative structure–activity relationship, Uncertainty quantification

Radiology Radiology

3D MRI in Osteoarthritis.

In Seminars in musculoskeletal radiology

Osteoarthritis (OA) is among the top 10 burdensome diseases, with the knee the most affected joint. Magnetic resonance imaging (MRI) allows whole-knee assessment, making it ideally suited for imaging OA, considered a multitissue disease. Three-dimensional (3D) MRI enables the comprehensive assessment of OA, including quantitative morphometry of various joint tissues. Manual tissue segmentation on 3D MRI is challenging but may be overcome by advanced automated image analysis methods including artificial intelligence (AI). This review presents examples of the utility of 3D MRI for knee OA, focusing on the articular cartilage, bone, meniscus, synovium, and infrapatellar fat pad, and it highlights several applications of AI that facilitate segmentation, lesion detection, and disease classification.

Oei Edwin H G, van Zadelhoff Tijmen A, Eijgenraam Susanne M, Klein Stefan, Hirvasniemi Jukka, van der Heijden Rianne A

2021-Jun

Radiology Radiology

3D MRI Models of the Musculoskeletal System.

In Seminars in musculoskeletal radiology

Computed tomography (CT) is most commonly used to produce three-dimensional (3D) models for evaluating bone and joint morphology in clinical practice. However, 3D models created from magnetic resonance imaging (MRI) data can be equally effective for comprehensive and accurate assessment of osseous and soft tissue structure morphology and pathology. The quality of 3D MRI models has steadily increased over time, with growing potential to replace 3D CT models in various musculoskeletal (MSK) applications. In practice, a single MRI examination for two-dimensional and 3D assessments can increase the value of MRI and simplify the pre- and postoperative imaging work-up. Multiple studies have shown excellent performance of 3D MRI models in shoulder injuries, in the hip in the setting of femoroacetabular impingement, and in the knee for the creation of bone surface models. Therefore, the utility of 3D MRI postprocessed models is expected to continue to rise and broaden in applications. Computer-based and artificial intelligence-assisted postprocessing techniques have tremendous potential to improve the efficiency of 3D model creation, opening many research avenues to validate the applicability of 3D MRI and establish 3D-specific quantitative assessment criteria. We provide a practice-focused overview of 3D MRI acquisition strategies, postprocessing techniques for 3D model creation, MSK applications of 3D MRI models, and an illustration of cases from our daily clinical practice.

Samim Mohammad

2021-Jun

General General

Effect of visual input on syllable parsing in a computational model of a neural microcircuit for speech processing.

In Journal of neural engineering ; h5-index 52.0

Seeing a person talking can help to understand them, in particular in a noisy environment. However, how the brain integrates the visual information with the auditory signal to enhance speech comprehension remains poorly understood. Here we address this question in a computational model of a cortical microcircuit for speech processing. The model consists of an excitatory and an inhibitory neural population that together create oscillations in the theta frequency range. When simulated with speech, the theta rhythm becomes entrained to the onsets of syllables, such that the onsets can be inferred from the network activity. We investigate how well the obtained syllable parsing performs when different types of visual stimuli are added. In particular, we consider currents related to the rate of syllables as well as currents related to the mouth-opening area of the talking faces. We find that currents that target the excitatory neuronal population can influence speech comprehension, both boosting it or impeding it, depending on the temporal delay and on whether the currents are excitatory or inhibitory. In contrast, currents that act on the inhibitory neurons do not impact speech comprehension significantly. Our results suggest neural mechanisms for the integration of visual information with the acoustic information in speech and make experimentally-testable predictions.

Kulkarni Anirudh, Kegler Mikolaj, Reichenbach Tobias

2021-Sep-21

computational, input, model, neural, parsings, syllable

Surgery Surgery

Highly accurate diagnosis of lung adenocarcinoma and squamous cell carcinoma tissues by deep learning.

In Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy

Intraoperative detection of the marginal tissues is the last and most important step to complete the resection of adenocarcinoma and squamous cell carcinoma. However, the current intraoperative diagnosis is time-consuming and requires numerous steps including staining. In this paper, we present the use of Raman spectroscopy with deep learning to achieve accurate diagnosis with stain-free process. To make the spectrum more suitable for deep learning, we utilize an unusual way of thinking which regards Raman spectral signal as a sequence and then converts it into two-dimensional Raman spectrogram by short-time Fourier transform as input. The normal-adenocarcinoma deep learning model and normal-squamous carcinoma deep learning model both achieve more than 96% accuracy, 95% sensitivity and 98% specificity when test, which higher than the conventional principal components analysis-linear discriminant analysis method with normal-adenocarcinoma model (0.896 accuracy, 0.867 sensitivity, 0.926 specificity) and normal-squamous carcinoma model (0.821 accuracy, 0.776 sensitivity, 1.000 specificity). The high performance of deep learning models provides a reliable way for intraoperative detection of marginal tissue, and is expected to reduce the detection time and save human lives.

Qi Yafeng, Yang Lin, Liu Bangxu, Liu Li, Liu Yuhong, Zheng Qingfeng, Liu Dameng, Luo Jianbin

2021-Sep-14

Deep learning, Lung adenocarcinoma and squamous cell carcinoma, Raman spectrogram, Tissue diagnosis

General General

Nonlinear tensor train format for deep neural network compression.

In Neural networks : the official journal of the International Neural Network Society

Deep neural network (DNN) compression has become a hot topic in the research of deep learning since the scale of modern DNNs turns into too huge to implement on practical resource constrained platforms such as embedded devices. Among variant compression methods, tensor decomposition appears to be a relatively simple and efficient strategy owing to its solid mathematical foundations and regular data structure. Generally, tensorizing neural weights into higher-order tensors for better decomposition, and directly mapping efficient tensor structure to neural architecture with nonlinear activation functions, are the two most common ways. However, the considerable accuracy loss is still a fly in the ointment for the tensorizing way especially for convolutional neural networks (CNNs), while the number of studies in the mapping way is comparatively limited and corresponding compression ratio appears to be not considerable. Therefore, in this work, by researching multiple types of tensor decompositions, we realize that tensor train (TT), which has specific and efficient sequenced contractions, is potential to take into account both of tensorizing and mapping ways. Then we propose a novel nonlinear tensor train (NTT) format, which contains extra nonlinear activation functions embedded in sequenced contractions and convolutions on the top of the normal TT decomposition and the proposed TT format connected by convolutions, to compensate the accuracy loss that normal TT cannot give. Further than just shrinking the space complexity of original weight matrices and convolutional kernels, we prove that NTT can afford an efficient inference time as well. Extensive experiments and discussions demonstrate that the compressed DNNs in our NTT format can almost maintain the accuracy at least on MNIST, UCF11 and CIFAR-10 datasets, and the accuracy loss caused by normal TT could be compensated significantly on large-scale datasets such as ImageNet.

Wang Dingheng, Zhao Guangshe, Chen Hengnu, Liu Zhexian, Deng Lei, Li Guoqi

2021-Sep-08

Neural network compression, Nonlinear tensor train, Sequenced contractions, Sequenced convolutions, Tensor train decomposition

Surgery Surgery

The application of machine learning algorithms in predicting the length of stay following femoral neck fracture.

In International journal of medical informatics ; h5-index 49.0

PURPOSE : Femoral neck fracture is a frequent cause of hospitalization, and length of stay is an important marker of hospital cost and quality of care provided. As an extension of traditional statistical methods, machine learning provides the possibility of accurately predicting the length of hospital stay. The aim of this paper is to retrospectively identify predictive factors of the length of hospital stay (LOS) and predict the postoperative LOS by using machine learning algorithms.

METHOD : Based on the admission and perioperative data of the patients, linear regression was used to analyze the predictive factors of the LOS. Multiple machine learning models were developed, and the performance of different models was compared.

RESULT : Stepwise linear regression showed that preoperative calcium level (P = 0.017) and preoperative lymphocyte percentage (P = 0.007), in addition to intraoperative bleeding (p = 0.041), glucose and sodium chloride infusion after surgery (P = 0.019), Charlson Comorbidity Index (p = 0.007) and BMI (P = 0.031), were significant predictors of LOS. The best performing model was the principal component regression (PCR) with an optimal MAE (1.525) and a proportion of prediction error within 3 days of 90.91%.

CONCLUSION : Excessive intravenous glucose and sodium chloride infusion after surgery, preoperative hypocalcemia, preoperative high percentages of lymphocytes, excessive intraoperative bleeding, lower BMI and higher CCI scores were related to prolonged LOS by using linear regression. Machine learning could accurately predict the postoperative LOS. This information allows hospital administrators to plan reasonable resource allocation to fulfill demand, leading to direct care quality improvement and more reasonable use of scarce resources.

Zhong Hao, Wang Bingpu, Wang Dawei, Liu Zirui, Xing Cong, Wu Yu, Gao Qiang, Zhu Shibo, Qu Haodong, Jia Zeyu, Qu Zhigang, Ning Guangzhi, Feng Shiqing

2021-Sep-13

Femoral neck fracture, Intravenous fluid management, Length of stay, Machine learning

General General

Machine learning based early mortality prediction in the emergency department.

In International journal of medical informatics ; h5-index 49.0

BACKGROUND : It is a great challenge for emergency physicians to early detect the patient's deterioration and prevent unexpected death through a large amount of clinical data, which requires sufficient experience and keen insight.

OBJECTIVE : To evaluate the performance of machine learning models in quantifying the severity of emergency department (ED) patients and identifying high-risk patients.

METHODS : Using routinely-available demographics, vital signs and laboratory tests extracted from electronic health records (EHRs), a framework based on machine learning and feature engineering was proposed for mortality prediction. Patients who had one complete record of vital signs and laboratory tests in ED were included. The following patients were excluded: pediatric patients aged < 18 years, pregnant woman, and patients died or were discharged or hospitalized within 12 h after admission. Based on 76 original features extracted, 9 machine learning models were adopted to validate our proposed framework. Their optimal hyper-parameters were fine-tuned using the grid search method. The prediction results were evaluated on performance metrics (i.e., accuracy, area under the curve (AUC), recall and precision) with repeated 5-fold cross-validation (CV). The time window from patient admission to the prediction was analyzed at 12 h, 24 h, 48 h, and entire stay.

RESULTS : We studied a total of 1114 ED patients with 71.54% (797/1114) survival and 28.46% (317/1114) death in the hospital. The results revealed a more complete time window leads to better prediction performance. Using the entire stay records, the LightGBM model with refined feature engineering demonstrated high discrimination and achieved 93.6% (±0.008) accuracy, 97.6% (±0.003) AUC, 97.1% (±0.008) recall, and 94.2% (±0.006) precision, even if no diagnostic information was utilized.

CONCLUSIONS : This study quantifies the criticality of ED patients and appears to have significant potential as a clinical decision support tool in assisting physicians in their clinical routine. While the model requires validation before use elsewhere, the same methodology could be used to create a strong model for the new hospital.

Li Cong, Zhang Zhuo, Ren Yazhou, Nie Hu, Lei Yuqing, Qiu Hang, Xu Zenglin, Pu Xiaorong

2021-Sep-08

Electronic health records, Emergency department, Feature engineering, Machine learning, Mortality prediction

General General

LightGBM: accelerated genomically designed crop breeding through ensemble learning.

In Genome biology ; h5-index 114.0

LightGBM is an ensemble model of decision trees for classification and regression prediction. We demonstrate its utility in genomic selection-assisted breeding with a large dataset of inbred and hybrid maize lines. LightGBM exhibits superior performance in terms of prediction precision, model stability, and computing efficiency through a series of benchmark tests. We also assess the factors that are essential to ensure the best performance of genomic selection prediction by taking complex scenarios in crop hybrid breeding into account. LightGBM has been implemented as a toolbox, CropGBM, encompassing multiple novel functions and analytical modules to facilitate genomically designed breeding in crops.

Yan Jun, Xu Yuetong, Cheng Qian, Jiang Shuqin, Wang Qian, Xiao Yingjie, Ma Chuang, Yan Jianbing, Wang Xiangfeng

2021-Sep-20

Crop breeding, Ensemble learning, Genomic prediction, Genomic selection, LightGBM, Machine learning, Maize, rrBLUP

General General

Structure-based protein design with deep learning.

In Current opinion in chemical biology

Since the first revelation of proteins functioning as macromolecular machines through their three dimensional structures, researchers have been intrigued by the marvelous ways the biochemical processes are carried out by proteins. The aspiration to understand protein structures has fueled extensive efforts across different scientific disciplines. In recent years, it has been demonstrated that proteins with new functionality or shapes can be designed via structure-based modeling methods, and the design strategies have combined all available information - but largely piece-by-piece - from sequence derived statistics to the detailed atomic-level modeling of chemical interactions. Despite the significant progress, incorporating data-derived approaches through the use of deep learning methods can be a game changer. In this review, we summarize current progress, compare the arc of developing the deep learning approaches with the conventional methods, and describe the motivation and concepts behind current strategies that may lead to potential future opportunities.

Ovchinnikov Sergey, Huang Po-Ssu

2021-Sep-18

Deep learning, Neural networks, Protein design, Protein sequence design, Protein structure, Protein structure design

Public Health Public Health

Artificial intelligence approach towards assessment of condition of COVID-19 patients - Identification of predictive biomarkers associated with severity of clinical condition and disease progression.

In Computers in biology and medicine

BACKGROUND AND OBJECTIVES : Although ML has been studied for different epidemiological and clinical issues as well as for survival prediction of COVID-19, there is a noticeable shortage of literature dealing with ML usage in prediction of disease severity changes through the course of the disease. In that way, predicting disease progression from mild towards moderate, severe and critical condition, would help not only to respond in a timely manner to prevent lethal results, but also to minimize the number of patients in hospitals where this is not necessary.

METHODS : We present a methodology for the classification of patients into 4 distinct categories of the clinical condition of COVID-19 disease. Classification of patients is based on the values of blood biomarkers that were assessed by Gradient boosting regressor and which were selected as biomarkers that have the greatest influence in the classification of patients with COVID-19.

RESULTS : The results show that among several tested algorithms, XGBoost classifier achieved best results with an average accuracy of 94% and an average F1-score of 94.3%. We have also extracted 10 best features from blood analysis that are strongly associated with patient condition and based on those features we can predict the severity of the clinical condition.

CONCLUSIONS : The main advantage of our system is that it is a decision tree-based algorithm which is easier to interpret, instead of the use of black box models, which are not appealing in medical practice.

Blagojević Anđela, Šušteršič Tijana, Lorencin Ivan, Šegota Sandi Baressi, Anđelić Nikola, Milovanović Dragan, Baskić Danijela, Baskić Dejan, Petrović Nataša Zdravković, Sazdanović Predrag, Car Zlatan, Filipović Nenad

2021-Sep-14

COVID-19, Clinical condition assessment, Personalized model, Predictive blood biomarkers, Rule-based machine learning

General General

The role of machine learning in the primary prevention of work-related musculoskeletal disorders: A scoping review.

In Applied ergonomics

To determine the applications of machine learning (ML) techniques used for the primary prevention of work-related musculoskeletal disorders (WMSDs), a scoping review was conducted using seven literature databases. Of the 4,639 initial results, 130 primary research studies were deemed relevant for inclusion. Studies were reviewed and classified as a contribution to one of six steps within the primary WMSD prevention research framework by van der Beek et al. (2017). ML techniques provided the greatest contributions to the development of interventions (48 studies), followed by risk factor identification (33 studies), underlying mechanisms (29 studies), incidence of WMSDs (14 studies), evaluation of interventions (6 studies), and implementation of effective interventions (0 studies). Nearly a quarter (23.8%) of all included studies were published in 2020. These findings provide insight into the breadth of ML techniques used for primary WMSD prevention and can help identify areas for future research and development.

Chan Victor C H, Ross Gwyneth B, Clouthier Allison L, Fischer Steven L, Graham Ryan B

2021-Sep-18

Artificial intelligence, Classification, Cluster analysis, Occupational injury, Prediction

Dermatology Dermatology

Deep neural network for early image diagnosis of Stevens-Johnson syndrome/toxic epidermal necrolysis.

In The journal of allergy and clinical immunology. In practice

BACKGROUND : Stevens-Johnson syndrome (SJS)/toxic epidermal necrolysis (TEN) is a life-threatening cutaneous adverse drug reaction (cADR). However, distinguishing SJS/TEN from non-severe cADRS is difficult, especially in the early stages of the disease.

OBJECTIVE : To overcome this limitation, we developed a computer-aided diagnosis system for the early diagnosis of SJS/TEN, powered by a deep convolutional neural network (DCNN).

METHODS : We trained a DCNN using a dataset of 26,661 individual lesion images obtained from 123 patients diagnosed with SJS/TEN or non-severe cADRs. The DCNN's classification accuracy was compared to that of 10 board-certified dermatologists and 24 trainee dermatologists.

RESULTS : The DCNN achieved 84.6% (95% confidence interval [CI], 80.6－88.6) sensitivity, whereas the sensitivities of the board-certified dermatologists and the trainee dermatologists were 31.3 % (95% CI, 20.9－41.8; P<0.0001) and 27.8% (95% CI, 22.6-32.5; P<0.0001), respectively. The negative predictive value was 94.6% (95% CI, 93.2－96.0) for the DCNN, 68.1 % (95% CI, 66.1－70.0; P<0.0001) for the board-certified dermatologists, and 67.4% (95% CI, 66.1－68.7; P<0.0001) for the trainee dermatologists. The area under the receiver operating characteristic curve of the DCNN for SJS/TEN diagnosis was 0.873, which was significantly higher than that of all the board-certified dermatologists and trainee dermatologists.results CONCLUSIONS: We developed a DCNN to classify SJS/TEN and non-severe cADRs based on individual lesion images of erythema. The DCNN performed significantly better than dermatologists in classifying SJS/TEN from skin images.

Fujimoto Atsushi, Iwai Yuki, Ishikawa Takashi, Shinkuma Satoru, Shido Kosuke, Yamasaki Kenshi, Fujisawa Yasuhiro, Fujimoto Manabu, Muramatsu Shogo, Abe Riichiro

2021-Sep-18

Stevens–Johnson syndrome/toxic epidermal necrolysis, artificial intelligence, cutaneous adverse drug reaction, deep convolutional neural network, early diagnosis, image diagnosis

General General

Imaging in Focus: An Introduction to Denoising Bioimages in the Era of Deep Learning.

In The international journal of biochemistry & cell biology

Fluorescence microscopy enables the direct observation of previously hidden dynamic processes of life, allowing profound insights into mechanisms of health and disease. However, imaging of live samples is fundamentally limited by the toxicity of the illuminating light and images are often acquired using low light conditions. As a consequence, images can become very noisy which severely complicates their interpretation. In recent years, deep learning (DL) has emerged as a very successful approach to remove this noise while retaining the useful signal. Unlike classical algorithms which use well-defined mathematical functions to remove noise, DL methods learn to denoise from example data, providing a powerful content-aware approach. In this review, we first describe the different types of noise that typically corrupt fluorescence microscopy images and introduce the denoising task. We then present the main DL-based denoising methods and their relative advantages and disadvantages. We aim to provide insights into how DL-based denoising methods operate and help users choose the most appropriate tools for their applications.

Laine Romain F, Jacquemet Guillaume, Krull Alexander

2021-Sep-18

Deep learning, denoising, live-cell imaging, microscopy, noise

General General

The maternal blood lipidome is indicative of the pathogenesis of severe preeclampsia.

In Journal of lipid research

Preeclampsia is a pregnancy-specific syndrome characterized by hypertension and proteinuria after 20 weeks of gestation. However, it is not well understood what lipids are involved in the development of this condition, and even less is known how these lipids mediate its formation. To reveal the relationship between lipids and preeclampsia, we conducted lipidomic profiling of maternal sera of 44 severe preeclamptic and 20 healthy pregnant women from a multi-ethnic cohort in Hawaii. Correlation network analysis showed that oxidized phospholipids (OxPLs) have increased inter-correlations and connections in preeclampsia, while other lipids, including triacylglycerols (TAGs), have reduced network correlations and connections. A total of 10 lipid species demonstrate significant changes uniquely associated with preeclampsia but not any other clinical confounders. These species are from the lipid classes of lysophosphatidylcholines (LPC), phosphatidylcholines (PC), cholesterol esters (CE), phosphatidylethanolamines (PE), lysophosphatidylethanolamines (LPE) and ceramides (Cer). A random forest (RF) classifier built on these lipids shows highly accurate and specific prediction (F1 statistic 0.94, balanced accuracy 0.88) of severe preeclampsia, demonstrating their potential as biomarkers for this condition. These lipid species are enriched in dysregulated biological pathways including insulin signaling, immune response, and phospholipid metabolism. Moreover, causality inference shows that various PCs and LPCs mediate severe preeclampsia through PC 35:1e. Our results suggest that the lipidome may play a role in and serve as biomarkers and the pathogenesis of severe preeclampsia.

He Bing, Liu Yu, Maurya Mano R, Benny Paula, Lassiter Cameron, Li Hui, Subramaniam Shankar, Garmire Lana X

2021-Sep-18

biomarker, classification, hypertension, lipidomics, machine learning, maternal blood, metabolomics, pathway, preeclampsia, pregnancy

General General

Real-time artificial intelligence for detecting focal lesions and diagnosing neoplasms of the stomach by white-light endoscopy (with videos).

In Gastrointestinal endoscopy ; h5-index 72.0

BACKGROUND AND AIMS : White-light endoscopy (WLE) is the most pivotal tool to detect gastric cancer in the early stage. However, the skill among endoscopists varies greatly. Here, we aim to develop a deep learning-based system named ENDOANGEL-LD (lesion detection) to assist in detecting all focal gastric lesions and predicting neoplasms by WLE.

METHODS : Endoscopic images were retrospectively obtained from Renmin Hospital of Wuhan University (RHWU) for the development, validation and internal test of the system. Additional external tests were conducted in 5 other hospitals to evaluate the robustness. Stored videos from RHWU were used for assessing and comparing the performance of ENDOANGEL-LD with that of experts. Prospective consecutive patients undergoing upper endoscopy were enrolled from May 6, 2021 to Aug 2 2021 in RHWU to assess clinical practice applicability.

RESULTS : Over 10 thousand patients undergoing upper endoscopy were enrolled in this study. The sensitivities were 96.9% and 95.6% for detecting gastric lesions, 92.9% and 91.7% for diagnosing neoplasm in internal and external patients, respectively. In 100 videos, ENDOANGEL-LD achieved superior sensitivity and negative predictive value for detecting gastric neoplasm than that of experts (100% vs 85.5%±3.4%, p=0.003; 100% vs 86.4%±2.8%, p=0.002). In 2010 prospective consecutive patients, ENDOANGEL-LD achieved a sensitivity of 92.8% for detecting gastric lesions with 3.04±3.04 false positives per gastroscopy, and a sensitivity of 91.8% and specificity of 92.4% for diagnosing neoplasm.

CONCLUSIONS : The results show that ENDOANGEL-LD has great potential for assisting endoscopists in screening gastric lesions and suspicious neoplasms in clinical work.

Wu Lianlian, Xu Ming, Jiang Xiaoda, He Xinqi, Zhang Heng, Ai Yaowei, Tong Qiaoyun, Lv Peihua, Lu Bin, Guo Mingwen, Huang Manling, Ye Liping, Shen Lei, Yu Honggang

2021-Sep-18

Deep learning, Early gastric cancer, Esophagogastroduodenoscopy, Gastric neoplasm

General General

Comparing the performance of machine learning algorithms for remote and in situ estimations of chlorophyll-a content: a case study in the Tri An Reservoir, Vietnam.

In Water environment research : a research publication of the Water Environment Federation

Chlorophyll-a (Chl-a) is one of the most important indicators of the trophic status of inland waters, and its continued monitoring is essential. Recently, the operated Sentinel-2 MSI satellite offers high spatial resolution images for remote water quality monitoring. In this study, we tested the performance of the three well-known machine learning (ML) [random forest (RF), support vector machine (SVM), Gaussian process (GP)] and the two novel ML [extreme gradient boost (XGB), CatBoost (CB)] models for estimation a wide-range of Chl-a concentration (10.1 to 798.7 μg/L) using the Sentinel-2 MSI data, and in situ water quality measurement in the Tri An Reservoir (TAR), Vietnam. GP indicated the most reliable model for predicting Chl-a from water quality parameters [R2 = 0.85, RMSE = 56.65 μg/L, Akaike's information criterion (AIC) = 575.10, and Bayesian information criterion (BIC) = 595.24]. Regarding input model as water surface reflectance, CB was the superior model for Chl-a retrieval (R2 = 0.84, RMSE = 46.28 μg/L, AIC = 229.18, and BIC = 238.50). Our results indicated that GP and CB are the two best models for the prediction of Chl-a in TAR. Overall, the Sentinel-2 MSI coupled with ML algorithms is a reliable, inexpensive, and accurate instrument for monitoring Chl-a in inland waters.

Nguyen Hao Quang, Ha Nam Thang, Nguyen Ngoc Lam, Pham Thanh Luu

2021-Sep-21

CatBoost, Gaussian process, machine learning, remote sensing, sentinel-2 MSI

General General

Environmental DNA gives comparable results to morphology-based indices of macroinvertebrates in a large-scale ecological assessment.

In PloS one ; h5-index 176.0

Anthropogenic activities are changing the state of ecosystems worldwide, affecting community composition and often resulting in loss of biodiversity. Rivers are among the most impacted ecosystems. Recording their current state with regular biomonitoring is important to assess the future trajectory of biodiversity. Traditional monitoring methods for ecological assessments are costly and time-intensive. Here, we compared monitoring of macroinvertebrates based on environmental DNA (eDNA) sampling with monitoring based on traditional kick-net sampling to assess biodiversity patterns at 92 river sites covering all major Swiss river catchments. From the kick-net community data, a biotic index (IBCH) based on 145 indicator taxa had been established. The index was matched by the taxonomically annotated eDNA data by using a machine learning approach. Our comparison of diversity patterns only uses the zero-radius Operational Taxonomic Units assigned to the indicator taxa. Overall, we found a strong congruence between both methods for the assessment of the total indicator community composition (gamma diversity). However, when assessing biodiversity at the site level (alpha diversity), the methods were less consistent and gave complementary data on composition. Specifically, environmental DNA retrieved significantly fewer indicator taxa per site than the kick-net approach. Importantly, however, the subsequent ecological classification of rivers based on the detected indicators resulted in similar biotic index scores for the kick-net and the eDNA data that was classified using a random forest approach. The majority of the predictions (72%) from the random forest classification resulted in the same river status categories as the kick-net approach. Thus, environmental DNA validly detected indicator communities and, combined with machine learning, provided reliable classifications of the ecological state of rivers. Overall, while environmental DNA gives complementary data on the macroinvertebrate community composition compared to the kick-net approach, the subsequently calculated indices for the ecological classification of river sites are nevertheless directly comparable and consistent.

Brantschen Jeanine, Blackman Rosetta C, Walser Jean-Claude, Altermatt Florian

2021

Radiology Radiology

Multiparametric MRI for the improved diagnostic accuracy of Alzheimer's disease and mild cognitive impairment: Research protocol of a case-control study design.

In PloS one ; h5-index 176.0

BACKGROUND : Alzheimer's disease (AD) is a major neurocognitive disorder identified by memory loss and a significant cognitive decline based on previous level of performance in one or more cognitive domains that interferes in the independence of everyday activities. The accuracy of imaging helps to identify the neuropathological features that differentiate AD from its common precursor, mild cognitive impairment (MCI). Identification of early signs will aid in risk stratification of disease and ensures proper management is instituted to reduce the morbidity and mortality associated with AD. Magnetic resonance imaging (MRI) using structural MRI (sMRI), functional MRI (fMRI), diffusion tensor imaging (DTI), and magnetic resonance spectroscopy (1H-MRS) performed alone is inadequate. Thus, the combination of multiparametric MRI is proposed to increase the accuracy of diagnosing MCI and AD when compared to elderly healthy controls.

METHODS : This protocol describes a non-interventional case control study. The AD and MCI patients and the healthy elderly controls will undergo multi-parametric MRI. The protocol consists of sMRI, fMRI, DTI, and single-voxel proton MRS sequences. An eco-planar imaging (EPI) will be used to perform resting-state fMRI sequence. The structural images will be analysed using Computational Anatomy Toolbox-12, functional images will be analysed using Statistical Parametric Mapping-12, DPABI (Data Processing & Analysis for Brain Imaging), and Conn software, while DTI and 1H-MRS will be analysed using the FSL (FMRIB's Software Library) and Tarquin respectively. Correlation of the MRI results and the data acquired from the APOE genotyping, neuropsychological evaluations (i.e. Montreal Cognitive Assessment [MoCA], and Mini-Mental State Examination [MMSE] scores) will be performed. The imaging results will also be correlated with the sociodemographic factors. The diagnosis of AD and MCI will be standardized and based on the DSM-5 criteria and the neuropsychological scores.

DISCUSSION : The combination of sMRI, fMRI, DTI, and MRS sequences can provide information on the anatomical and functional changes in the brain such as regional grey matter volume atrophy, impaired functional connectivity among brain regions, and decreased metabolite levels specifically at the posterior cingulate cortex/precuneus. The combination of multiparametric MRI sequences can be used to stratify the management of MCI and AD patients. Accurate imaging can decide on the frequency of follow-up at memory clinics and select classifiers for machine learning that may aid in the disease identification and prognostication. Reliable and consistent quantification, using standardised protocols, are crucial to establish an optimal diagnostic capability in the early detection of Alzheimer's disease.

Piersson Albert Dayor, Ibrahim Buhari, Suppiah Subapriya, Mohamad Mazlyfarina, Hassan Hasyma Abu, Omar Nur Farhayu, Ibrahim Mohd Izuan, Yusoff Ahmad Nazlim, Ibrahim Normala, Saripan M Iqbal, Razali Rizah Mazzuin

2021

General General

Minimizing bias in massive multi-arm observational studies with BCAUS: balancing covariates automatically using supervision.

In BMC medical research methodology

BACKGROUND : Observational studies are increasingly being used to provide supplementary evidence in addition to Randomized Control Trials (RCTs) because they provide a scale and diversity of participants and outcomes that would be infeasible in an RCT. Additionally, they more closely reflect the settings in which the studied interventions will be applied in the future. Well-established propensity-score-based methods exist to overcome the challenges of working with observational data to estimate causal effects. These methods also provide quality assurance diagnostics to evaluate the degree to which bias has been removed and the estimates can be trusted. In large medical datasets it is common to find the same underlying health condition being treated with a variety of distinct drugs or drug combinations. Conventional methods require a manual iterative workflow, making them scale poorly to studies with many intervention arms. In such situations, automated causal inference methods that are compatible with traditional propensity-score-based workflows are highly desirable.

METHODS : We introduce an automated causal inference method BCAUS, that features a deep-neural-network-based propensity model that is trained with a loss which penalizes both the incorrect prediction of the assigned treatment as well as the degree of imbalance between the inverse probability weighted covariates. The network is trained end-to-end by dynamically adjusting the loss term for each training batch such that the relative contributions from the two loss components are held fixed. Trained BCAUS models can be used in conjunction with traditional propensity-score-based methods to estimate causal treatment effects.

RESULTS : We tested BCAUS on the semi-synthetic Infant Health & Development Program dataset with a single intervention arm, and a real-world observational study of diabetes interventions with over 100,000 individuals spread across more than a hundred intervention arms. When compared against other recently proposed automated causal inference methods, BCAUS had competitive accuracy for estimating synthetic treatment effects and provided highly concordant estimates on the real-world dataset but was an order-of-magnitude faster.

CONCLUSIONS : BCAUS is directly compatible with trusted protocols to estimate treatment effects and diagnose the quality of those estimates, while making the established approaches automatically scalable to an arbitrary number of simultaneous intervention arms without any need for manual iteration.

Belthangady Chinmay, Stedden Will, Norgeot Beau

2021-Sep-20

Causal inference, Deep learning, Neural networks, Observational studies

Public Health Public Health

Artificial intelligence approach towards assessment of condition of COVID-19 patients - Identification of predictive biomarkers associated with severity of clinical condition and disease progression.

In Computers in biology and medicine

BACKGROUND AND OBJECTIVES : Although ML has been studied for different epidemiological and clinical issues as well as for survival prediction of COVID-19, there is a noticeable shortage of literature dealing with ML usage in prediction of disease severity changes through the course of the disease. In that way, predicting disease progression from mild towards moderate, severe and critical condition, would help not only to respond in a timely manner to prevent lethal results, but also to minimize the number of patients in hospitals where this is not necessary.

METHODS : We present a methodology for the classification of patients into 4 distinct categories of the clinical condition of COVID-19 disease. Classification of patients is based on the values of blood biomarkers that were assessed by Gradient boosting regressor and which were selected as biomarkers that have the greatest influence in the classification of patients with COVID-19.

RESULTS : The results show that among several tested algorithms, XGBoost classifier achieved best results with an average accuracy of 94% and an average F1-score of 94.3%. We have also extracted 10 best features from blood analysis that are strongly associated with patient condition and based on those features we can predict the severity of the clinical condition.

CONCLUSIONS : The main advantage of our system is that it is a decision tree-based algorithm which is easier to interpret, instead of the use of black box models, which are not appealing in medical practice.

Blagojević Anđela, Šušteršič Tijana, Lorencin Ivan, Šegota Sandi Baressi, Anđelić Nikola, Milovanović Dragan, Baskić Danijela, Baskić Dejan, Petrović Nataša Zdravković, Sazdanović Predrag, Car Zlatan, Filipović Nenad

2021-Sep-14

COVID-19, Clinical condition assessment, Personalized model, Predictive blood biomarkers, Rule-based machine learning

General General

Voice Assistants and Cancer Screening: A Comparison of Alexa, Siri, Google Assistant, and Cortana.

In Annals of family medicine

Despite increasing interest in how voice assistants like Siri or Alexa might improve health care delivery and information dissemination, there is limited research assessing the quality of health information provided by these technologies. Voice assistants present both opportunities and risks when facilitating searches for or answering health-related questions, especially now as fewer patients are seeing their physicians for preventive care due to the ongoing pandemic. In our study, we compared the 4 most widely used voice assistants (Amazon Alexa, Apple Siri, Google Assistant, and Microsoft Cortana) and their ability to understand and respond accurately to questions about cancer screening. We show that there are clear differences among the 4 voice assistants and that there is room for improvement across all assistants, particularly in their ability to provide accurate information verbally. In order to ensure that voice assistants provide accurate information about cancer screening and support, rather than undermine efforts to improve preventive care delivery and population health, we suggest that technology providers prioritize partnership with health professionals and organizations.

Hong Grace, Folcarelli Albino, Less Jacob, Wang Claire, Erbasi Neslihan, Lin Steven

artificial intelligence, early detection of cancer, preventive medicine

General General

Technology-Enabled and Artificial Intelligence Support for Pre-Visit Planning in Ambulatory Care: Findings From an Environmental Scan.

In Annals of family medicine

PURPOSE : Pre-visit planning (PVP) is believed to improve effectiveness, efficiency, and experience of care, yet numerous implementation barriers exist. There are opportunities for technology-enabled and artificial intelligence (AI) support to augment existing human-driven PVP processes-from appointment reminders and pre-visit questionnaires to pre-visit order sets and care gap closures. This study aimed to explore the current state of PVP, barriers to implementation, evidence of impact, and potential use of non-AI and AI tools to support PVP.

METHODS : We used an environmental scan approach involving: (1) literature review; (2) key informant interviews with PVP experts in ambulatory care; and (3) a search of the public domain for technology-enabled and AI solutions that support PVP. We then synthesized the findings using a qualitative matrix analysis.

RESULTS : We found 26 unique PVP implementations in the literature and conducted 16 key informant interviews. Demonstration of impact is typically limited to process outcomes, with improved patient outcomes remaining elusive. Our key informants reported that many PVP barriers are human effort-related and see potential for non-AI and AI technologies to support certain aspects of PVP. We identified 8 examples of commercially available technology-enabled tools that support PVP, some with AI capabilities; however, few of these have been independently evaluated.

CONCLUSIONS : As health systems transition toward value-based payment models in a world where the coronavirus disease 2019 pandemic has shifted patient care into the virtual space, PVP activities-driven by humans and supported by technology-may become more important and powerful and should be rigorously evaluated.

Holdsworth Laura M, Park Chance, Asch Steven M, Lin Steven

ambulatory care, artificial intelligence, qualitative methods

General General

Correction: Predicting Health Material Accessibility: Development of Machine Learning Algorithms.

In JMIR medical informatics ; h5-index 23.0

[This corrects the article DOI: 10.2196/29175.].

Ji Meng, Liu Yanmeng, Hao Tianyong

2021-Sep-21

General General

A Novel Multiple-View Adversarial Learning Network for Unsupervised Domain Adaptation Action Recognition.

In IEEE transactions on cybernetics

Abstract-domain adaptation action recognition is a hot research topic in machine learning and some effective approaches have been proposed. However, samples in the target domain with label information are often required by these approaches. Moreover, domain-invariant discriminative feature learning, feature fusion, and classifier module learning have not been explored in an end-to-end framework. Thus, in this study, we propose a novel end-to-end multiple-view adversarial learning network (MAN) for unsupervised domain adaptation action recognition in which the fusion of RGB and optical-flow features, domain-invariant discrimination feature learning, and action recognition is conducted in a unified framework. Specifically, a robust spatiotemporal feature extraction network, including a spatial transform network and an adaptive intrachannel weight network, is proposed to improve the scale invariance and robustness of the method. Then, a self-attention mechanism fusion module is designed to adaptively fuse the RGB and optical-flow features. Moreover, a multiview adversarial learning loss is developed to obtain domain-invariant discriminative features. In addition, three benchmark datasets are constructed for unsupervised domain adaptation action recognition, for which all actions and samples are carefully collected from public action datasets, and their action categories are hierarchically augmented, which can guide how to extend existing action datasets. We conduct extensive experiments on four benchmark datasets, and the experimental results demonstrate that our proposed MAN can outperform several state-of-the-art unsupervised domain adaptation action recognition approaches. When the SDAI Action II-6 and SDAI Action II-11 datasets are used, MAN can achieve 3.7% (H→ U) and 6.1% (H→ U) improvements over the temporal attentive adversarial adaptation network (published in ICCV 2019) module, respectively. As an added contribution, the SDAI Action II-6, SDAI Action II-11, and SDAI Action II-16 datasets will be released to facilitate future research on domain adaptation action recognition.

Gao Zan, Zhao Yibo, Zhang Hua, Chen Da, Liu An-An, Chen Shengyong

2021-Sep-21

General General

Interpretability-Based Multimodal Convolutional Neural Networks for Skin Lesion Diagnosis.

In IEEE transactions on cybernetics

Skin lesion diagnosis is a key step for skin cancer screening, which requires high accuracy and interpretability. Though many computer-aided methods, especially deep learning methods, have made remarkable achievements in skin lesion diagnosis, their generalization and interpretability are still a challenge. To solve this issue, we propose an interpretability-based multimodal convolutional neural network (IM-CNN), which is a multiclass classification model with skin lesion images and metadata of patients as input for skin lesion diagnosis. The structure of IM-CNN consists of three main paths to deal with metadata, features extracted from segmented skin lesion with domain knowledge, and skin lesion images, respectively. We add interpretable visual modules to provide explanations for both images and metadata. In addition to area under the ROC curve (AUC), sensitivity, and specificity, we introduce a new indicator, an AUC curve with a sensitivity larger than 80% (AUC_SEN_80) for performance evaluation. Extensive experimental studies are conducted on the popular HAM10000 dataset, and the results indicate that the proposed model has overwhelming advantages compared with popular deep learning models, such as DenseNet, ResNet, and other state-of-the-art models for melanoma diagnosis. The proposed multimodal model also achieves on average 72% and 21% improvement in terms of sensitivity and AUC_SEN_80, respectively, compared with the single-modal model. The visual explanations can also help gain trust from dermatologists and realize man-machine collaborations, effectively reducing the limitation of black-box models in supporting medical decision making.\enlargethispage-8pt.

Wang Sutong, Yin Yunqiang, Wang Dujuan, Wang Yanzhang, Jin Yaochu

2021-Sep-21

General General

New Insights Into Drug Repurposing for COVID-19 Using Deep Learning.

In IEEE transactions on neural networks and learning systems

The coronavirus disease 2019 (COVID-19) has continued to spread worldwide since late 2019. To expedite the process of providing treatment to those who have contracted the disease and to ensure the accessibility of effective drugs, numerous strategies have been implemented to find potential anti-COVID-19 drugs in a short span of time. Motivated by this critical global challenge, in this review, we detail approaches that have been used for drug repurposing for COVID-19 and suggest improvements to the existing deep learning (DL) approach to identify and repurpose drugs to treat this complex disease. By optimizing hyperparameter settings, deploying suitable activation functions, and designing optimization algorithms, the improved DL approach will be able to perform feature extraction from quality big data, turning the traditional DL approach, referred to as a ``black box,'' which generalizes and learns the transmitted data, into a ``glass box'' that will have the interpretability of its rationale while maintaining a high level of prediction accuracy. When adopted for drug repurposing for COVID-19, this improved approach will create a new generation of DL approaches that can establish a cause and effect relationship as to why the repurposed drugs are suitable for treating COVID-19. Its ability can also be extended to repurpose drugs for other complex diseases, develop appropriate treatment strategies for new diseases, and provide precision medical treatment to patients, thus paving the way to discover new drugs that can potentially be effective for treating COVID-19.

Lee Chun Yen, Chen Yi-Ping Phoebe

2021-Sep-21

General General

Globally Localized Multisource Domain Adaptation for Cross-Domain Fault Diagnosis With Category Shift.

In IEEE transactions on neural networks and learning systems

Deep learning has demonstrated splendid performance in mechanical fault diagnosis on condition that source and target data are identically distributed. In engineering practice, however, the domain shift between source and target domains significantly limits the further application of intelligent algorithms. Despite various transfer techniques proposed, either they focus on single-source domain adaptation (SDA) or they utilize multisource domain globally or locally, which both cannot address the cross-domain diagnosis effectively, especially with category shift. To this end, we propose globally localized multisource DA for cross-domain fault diagnosis with category shift. Specifically, we construct a GlocalNet to fuse multisource information comprehensively, which consists of a feature generator and three classifiers. By optimizing the Wasserstein discrepancy of classifiers locally and accumulative higher order multisource moment globally, multisource DA is achieved from domain and class levels thus to reduce the shift on domain and category. To refine the classifier at sample level, a distilling strategy is presented. Finally, an adaptive weighting policy is employed for reliable result. To evaluate the effectiveness, the proposed method is compared with multiple methods on four bearing vibration datasets. Experimental results indicate the superiority and practicability of the proposed method for cross-domain fault diagnosis.

Feng Yong, Chen Jinglong, He Shuilong, Pan Tongyang, Zhou Zitong

2021-Sep-21

General General

Learning Versatile Convolution Filters for Efficient Visual Recognition.

In IEEE transactions on pattern analysis and machine intelligence ; h5-index 127.0

This paper introduces versatile filters to construct efficient convolutional neural networks that are widely used in various visual recognition tasks. Considering the demands of efficient deep learning techniques running on cost-effective hardware, a number of methods have been developed to learn compact neural networks. Most of these works aim to slim down filters in different ways, e.g. investigating small, sparse or quantized filters. In contrast, we treat filters from an additive perspective. A series of secondary filters can be derived from a primary filter with the help of binary masks. These secondary filters all inherit in the primary filter without occupying more storage, but once been unfolded in computation they could significantly enhance the capability of the filter by integrating information extracted from different receptive fields. Besides spatial versatile filters, we additionally investigate versatile filters from the channel perspective. Binary masks can be further customized for different primary filters under orthogonal constraints. We conduct theoretical analysis on network complexity and an efficient convolution scheme is introduced. Experimental results on benchmark datasets and neural networks demonstrate that our versatile filters are able to achieve comparable accuracy as that of original filters, but require less memory and computation cost.

Han Kai, Wang Yunhe, Xu Chang, Xu Chunjing, Wu Enhua, Tao Dacheng

2021-Sep-21

Public Health Public Health

Novel Application of Machine Learning Algorithms and Model-Agnostic Methods to Identify Factors Influencing Childhood Blood Lead Levels.

In Environmental science & technology ; h5-index 132.0

Blood lead (Pb) poisoning remains a global concern, particularly for children in their early developmental years. Broken Hill is Australia's oldest operating silver-zinc-lead mine. In this study, we utilized recent advances in machine learning to assess multiple algorithms and identify the most optimal model for predicting childhood blood Pb levels (BLL) using Broken Hill children's (<5 years of age) data (n = 23,749) from 1991 to 2015, combined with demographic, socio-economic, and environmental influencing factors. We applied model-agnostic methods to interpret the most optimal model, investigating different environmental and human factors influencing childhood BLL. Algorithm assessment showed that stacked ensemble, a method for automatically and optimally combining multiple prediction algorithms, enhanced predictive performance by 1.1% with respect to mean absolute error (p < 0.01) and 2.6% for root-mean-squared error (p < 0.01) compared to the best performing constituent algorithm (random forest). By interpreting the model, the following information was acquired: children had higher BLL if they resided within 1.0 km to the central mine area or 1.37 km to the railroad; year of testing had the greatest interactive strength with all other factors; BLL increased faster in Aboriginal than in non-Aboriginal children at 9-10 and 12-18 months of age. This "stacked ensemble + model-agnostic interpretation" framework achieved both prediction accuracy and model interpretability, identifying previously unconnected variables associated with elevated childhood BLL, offering a marked advantage over previous works. Thus, this approach has a clear value and potential for application to other environmental health issues.

Liu Xiaochi, Taylor Mark P, Aelion C Marjorie, Dong Chenyin

2021-Sep-21

Pb exposure, black-box model interpretation, childhood exposures, ethnic disparities, mining contamination, modeling methods, stacked ensemble

General General

Virtual Sensor Array Based on Butterworth-Van Dyke Equivalent Model of QCM for Selective Detection of Volatile Organic Compounds.

In ACS applied materials & interfaces ; h5-index 147.0

Recently virtual sensor arrays (VSAs) have been developed to improve the selectivity of volatile organic compound (VOC) sensors. However, most reported VSAs rely on detecting single property change of the sensing material after their exposure to VOCs, thus resulting in a loss of much valuable information. In this work, we propose a VSA with the high dimensionality of outputs based on a quartz crystal microbalance (QCM) and a sensing layer of MXene. Changes in both mechanical and electrical properties of the MXene film are utilized in the detection of the VOCs. We take the changes of parameters of the Butterworth-van Dyke model for the QCM-based sensor operated at multiple harmonics as the responses of the VSA to various VOCs. The dimensionality of the VSA's responses has been expanded to four independent outputs, and the responses to the VOCs have shown good linearity in multidimensional space. The response and recovery times are 16 and 54 s, respectively. Based on machine learning algorithms, the proposed VSA accurately identifies different VOCs and mixtures, as well as quantifies the targeted VOC in complex backgrounds (with an accuracy of 90.6%). Moreover, we demonstrate the capacity of the VSA to identify "patients with diabetic ketosis" from volunteers with an accuracy of 95%, based on the detection of their exhaled breath. The QCM-based VSA shows great potential for detecting VOC biomarkers in human breath for disease diagnosis.

Li Dongsheng, Xie Zihao, Qu Mengjiao, Zhang Qian, Fu Yongqing, Xie Jin

2021-Sep-21

MXene, QCM, VOCs sensor, breath analysis, selectivity, virtual sensor array

General General

Identifying homogeneous subgroups of patients and important features: a topological machine learning approach.

In BMC bioinformatics

BACKGROUND : This paper exploits recent developments in topological data analysis to present a pipeline for clustering based on Mapper, an algorithm that reduces complex data into a one-dimensional graph.

RESULTS : We present a pipeline to identify and summarise clusters based on statistically significant topological features from a point cloud using Mapper.

CONCLUSIONS : Key strengths of this pipeline include the integration of prior knowledge to inform the clustering process and the selection of optimal clusters; the use of the bootstrap to restrict the search to robust topological features; the use of machine learning to inspect clusters; and the ability to incorporate mixed data types. Our pipeline can be downloaded under the GNU GPLv3 license at https://github.com/kcl-bhi/mapper-pipeline .

Carr Ewan, Carrière Mathieu, Michel Bertrand, Chazal Frédéric, Iniesta Raquel

2021-Sep-20

Clustering, Machine learning, Topological data analysis

General General

New Insights Into Drug Repurposing for COVID-19 Using Deep Learning.

In IEEE transactions on neural networks and learning systems

The coronavirus disease 2019 (COVID-19) has continued to spread worldwide since late 2019. To expedite the process of providing treatment to those who have contracted the disease and to ensure the accessibility of effective drugs, numerous strategies have been implemented to find potential anti-COVID-19 drugs in a short span of time. Motivated by this critical global challenge, in this review, we detail approaches that have been used for drug repurposing for COVID-19 and suggest improvements to the existing deep learning (DL) approach to identify and repurpose drugs to treat this complex disease. By optimizing hyperparameter settings, deploying suitable activation functions, and designing optimization algorithms, the improved DL approach will be able to perform feature extraction from quality big data, turning the traditional DL approach, referred to as a ``black box,'' which generalizes and learns the transmitted data, into a ``glass box'' that will have the interpretability of its rationale while maintaining a high level of prediction accuracy. When adopted for drug repurposing for COVID-19, this improved approach will create a new generation of DL approaches that can establish a cause and effect relationship as to why the repurposed drugs are suitable for treating COVID-19. Its ability can also be extended to repurpose drugs for other complex diseases, develop appropriate treatment strategies for new diseases, and provide precision medical treatment to patients, thus paving the way to discover new drugs that can potentially be effective for treating COVID-19.

Lee Chun Yen, Chen Yi-Ping Phoebe

2021-Sep-21

Radiology Radiology

A Quantitative Comparison of Epistemic Uncertainty Maps Applied to Multi-Class Segmentation

ArXiv Preprint

Uncertainty assessment has gained rapid interest in medical image analysis. A popular technique to compute epistemic uncertainty is the Monte-Carlo (MC) dropout technique. From a network with MC dropout and a single input, multiple outputs can be sampled. Various methods can be used to obtain epistemic uncertainty maps from those multiple outputs. In the case of multi-class segmentation, the number of methods is even larger as epistemic uncertainty can be computed voxelwise per class or voxelwise per image. This paper highlights a systematic approach to define and quantitatively compare those methods in two different contexts: class-specific epistemic uncertainty maps (one value per image, voxel and class) and combined epistemic uncertainty maps (one value per image and voxel). We applied this quantitative analysis to a multi-class segmentation of the carotid artery lumen and vessel wall, on a multi-center, multi-scanner, multi-sequence dataset of (MR) images. We validated our analysis over 144 sets of hyperparameters of a model. Our main analysis considers the relationship between the order of the voxels sorted according to their epistemic uncertainty values and the misclassification of the prediction. Under this consideration, the comparison of combined uncertainty maps reveals that the multi-class entropy and the multi-class mutual information statistically out-perform the other combined uncertainty maps under study. In a class-specific scenario, the one-versus-all entropy statistically out-performs the class-wise entropy, the class-wise variance and the one versus all mutual information. The class-wise entropy statistically out-performs the other class-specific uncertainty maps in terms of calibration. We made a python package available to reproduce our analysis on different data and tasks.

Robin Camarasa, Daniel Bos, Jeroen Hendrikse, Paul Nederkoorn, M. Eline Kooi, Aad van der Lugt, Marleen de Bruijne

2021-09-22

General General

Enhanced detection of abnormalities in heart rate variability and dynamics by 7-day continuous ECG monitoring.

In Annals of noninvasive electrocardiology : the official journal of the International Society for Holter and Noninvasive Electrocardiology, Inc

BACKGROUND : The analysis of heart rate variability (HRV) and heart rate (HR) dynamics by Holter ECG has been standardized to 24 hs, but longer-term continuous ECG monitoring has become available in clinical practice. We investigated the effects of long-term ECG on the assessment of HRV and HR dynamics.

METHODS : Intraweek variations in HRV and HR dynamics were analyzed in 107 outpatients with sinus rhythm. ECG was recorded continuously for 7 days with a flexible, codeless, waterproof sensor attached on the upper chest wall. Data were divided into seven 24-h segments, and standard time- and frequency-domain HRV and nonlinear HR dynamics indices were computed for each segment.

RESULTS : The intraweek coefficients of variance of HRV and HR dynamics indices ranged from 2.9% to 26.0% and were smaller for frequency-domain than for time-domain indices, and for indices reflecting slower HR fluctuations than faster fluctuations. The indices with large variance often showed transient abnormalities from day to day over 7 days, reducing the positive predictive accuracy of the 24-h ECG for detecting persistent abnormalities over 7 days. Conversely, 7-day ECG provided 2.3- to 6.5-fold increase in sensitivity to detect persistent plus transient abnormalities compared with 24-h ECG. It detected an average of 1.74 to 2.91 times as many abnormal indices as 24-h ECG.

CONCLUSIONS : Long-term ECG monitoring increases the accuracy and sensitivity of detecting persistent and transient abnormalities in HRV and HR dynamics and allows discrimination between the two types of abnormalities. Whether this discrimination improves risk stratification deserves further studies.

Hayano Junichiro, Yuda Emi

2021-Sep-21

Holter ECG, day-to-day variation, heart rate dynamics, heart rate variability, intraweek variation, long-term ECG monitoring

Pathology Pathology

Evaluation of Scopio Labs X100 Full Field PBS: The first high-resolution full field viewing of peripheral blood specimens combined with artificial intelligence-based morphological analysis.

In International journal of laboratory hematology ; h5-index 29.0

BACKGROUND : Current digital cell imaging systems perform peripheral blood smear (PBS) analysis in limited regions of the PBS and require the support of manual microscopy without achieving full digital microscopy. We report a multicenter study that validated the Scopio Labs X100 Full Field PBS, a novel digital imaging system that utilizes a full field view approach for cell recognition and classification, in a decision support system mode.

METHODS : We analyzed 335 normal and 310 abnormal PBS from patients with various clinical conditions and compared the performance of Scopio's Full Field PBS as the test method, with manual PBS analysis as the reference method. Deming regression analysis was utilized for comparisons of WBC and platelet estimates. Measurements of WBC and platelet estimation accuracy along with the agreement on RBC morphology evaluation were performed. Reproducibility and repeatability (R&R) of the system were also evaluated.

RESULTS : Scopio's Full Field PBS WBC accuracy was evaluated with an efficiency of 96.29%, sensitivity of 87.86%, and specificity of 97.62%. The agreement between the test and reference method for RBC morphology reached 99.77%, and the accuracy for platelet estimation resulted in an efficiency of 94.89%, sensitivity of 90.00%, and specificity of 96.28%, with successful R&R tests. The system enabled a comprehensive review of full field PBS as shown in representative samples.

CONCLUSIONS : Scopio's Full Field PBS showed a high degree of correlation of all tested parameters with manual microscopy. The novel full field view of specimens facilitates the long-expected disengagement between the digital application and the manual microscope.

Katz Ben-Zion, Feldman Michael D, Tessema Minychel, Benisty Dan, Toles Grace Stewart, Andre Alicia, Shtreker Bronka, Paz Fatima Maria, Edwards Joshua, Jengehino Darrin, Bagg Adam, Avivi Irit, Pozdnyakova Olga

2021-Sep-21

Artificial Inteligence, blood smear, laboratory automation, morphology

General General

Incorporating label correlations into deep neural networks to classify protein subcellular location patterns in immunohistochemistry images.

In Proteins

Analysis of protein subcellular localization is a critical part of proteomics. In recent years, as both the number and quality of microscopic images are increasing rapidly, many automated methods, especially convolutional neural networks, have been developed to predict protein subcellular location(s) based on bioimages, but their performance always suffers from some inherent properties of the problem. First, many microscopic images have non-informative or noisy sections, like unstained stroma and unspecific background, which affect the extraction of protein expression information. Second, the patterns of protein subcellular localization are very complex, as a lot of proteins locate in more than one compartment. In this study, we propose a new label-correlation enhanced deep neural network, laceDNN, to classify the subcellular locations of multi-label proteins from immunohistochemistry images. The model uses small representative patches as input to alleviate the image noise issue, and its backbone is a hybrid architecture of convolutional neural network and recurrent neural network, where the former network extracts representative image features and the latter learns the organelle dependency relationships. Our experimental results indicate that the proposed model can improve the performance of multi-label protein subcellular classification. This article is protected by copyright. All rights reserved.

Hu Jing-Xian, Yang Yang, Xu Ying-Ying, Shen Hong-Bin

2021-Sep-21

data screening, deep learning, immunohistochemistry images, multi-label classification, protein subcellular localization

General General

Computational Medicine: Past, Present and Future.

In Chinese journal of integrative medicine

Computational medicine is an emerging discipline that uses computer models and complex software to simulate the development and treatment of diseases. Advances in computer hardware and software technology, especially the development of algorithms and graphics processing units (GPUs), have led to the broader application of computers in the medical field. Computer vision based on mathematical biological modelling will revolutionize clinical research and diagnosis, and promote the innovative development of Chinese medicine, some biological models have begun to play a practical role in various types of research. This paper introduces the concepts and characteristics of computational medicine and then reviews the developmental history of the field, including Digital Human in Chinese medicine. Additionally, this study introduces research progress in computational medicine around the world, lists some specific clinical applications of computational medicine, discusses the key problems and limitations of the research and the development and application of computational medicine, and ultimately looks forward to the developmental prospects, especially in the field of computational Chinese medicine.

Lyu Lan-Qing, Cui Hong-Yan, Shao Ming-Yi, Fu Yu, Zhao Rui-Xia, Chen Qiu-Ping

2021-Sep-21

artificial intelligence, big data, computational Chinese medicine, computational medicine, model

General General

Non-melanoma skin cancer diagnosis: a comparison between dermoscopic and smartphone images by unified visual and sonification deep learning algorithms.

In Journal of cancer research and clinical oncology

PURPOSE : Non-melanoma skin cancer (NMSC) is the most frequent keratinocyte-origin skin tumor. It is confirmed that dermoscopy of NMSC confers a diagnostic advantage as compared to visual face-to-face assessment. COVID-19 restrictions diagnostics by telemedicine photos, which are analogous to visual inspection, displaced part of in-person visits. This study evaluated by a dual convolutional neural network (CNN) performance metrics in dermoscopic (DI) versus smartphone-captured images (SI) and tested if artificial intelligence narrows the proclaimed gap in diagnostic accuracy.

METHODS : A CNN that receives a raw image and predicts malignancy, overlaid by a second independent CNN which processes a sonification (image-to-sound mapping) of the original image, were combined into a unified malignancy classifier. All images were histopathology-verified in a comparison between NMSC and benign skin lesions excised as suspected NMSCs. Study criteria outcomes were sensitivity and specificity for the unified output.

RESULTS : Images acquired by DI (n = 132 NMSC, n = 33 benign) were compared to SI (n = 170 NMSC, n = 28 benign). DI and SI analysis metrics resulted in an area under the curve (AUC) of the receiver operator characteristic curve of 0.911 and 0.821, respectively. Accuracy was increased by DI (0.88; CI 81.9-92.4) as compared to SI (0.75; CI 68.1-80.6, p < 0.005). Sensitivity of DI was higher than SI (95.3%, CI 90.4-98.3 vs 75.3%, CI 68.1-81.6, p < 0.001), but not specificity (p = NS).

CONCLUSION : Telemedicine use of smartphone images might result in a substantial decrease in diagnostic performance as compared to dermoscopy, which needs to be considered by both healthcare providers and patients.

Dascalu A, Walker B N, Oron Y, David E O

2021-Sep-21

Deep learning, Dermoscopy, Non-melanoma skin cancer, Preventive medicine, Sonification, Telemedicine

Public Health Public Health

Application of Deep Learning to Community-Science-Based Mosquito Monitoring and Detection of Novel Species.

In Journal of medical entomology ; h5-index 29.0

Mosquito-borne diseases account for human morbidity and mortality worldwide, caused by the parasites (e.g., malaria) or viruses (e.g., dengue, Zika) transmitted through bites of infected female mosquitoes. Globally, billions of people are at risk of infection, imposing significant economic and public health burdens. As such, efficient methods to monitor mosquito populations and prevent the spread of these diseases are at a premium. One proposed technique is to apply acoustic monitoring to the challenge of identifying wingbeats of individual mosquitoes. Although researchers have successfully used wingbeats to survey mosquito populations, implementation of these techniques in areas most affected by mosquito-borne diseases remains challenging. Here, methods utilizing easily accessible equipment and encouraging community-scientist participation are more likely to provide sufficient monitoring. We present a practical, community-science-based method of monitoring mosquito populations using smartphones. We applied deep-learning algorithms (TensorFlow Inception v3) to spectrogram images generated from smartphone recordings associated with six mosquito species to develop a multiclass mosquito identification system, and flag potential invasive vectors not present in our sound reference library. Though TensorFlow did not flag potential invasive species with high accuracy, it was able to identify species present in the reference library at an 85% correct identification rate, an identification rate markedly higher than similar studies employing expensive recording devices. Given that we used smartphone recordings with limited sample sizes, these results are promising. With further optimization, we propose this novel technique as a way to accurately and efficiently monitor mosquito populations in areas where doing so is most critical.

Khalighifar Ali, Jiménez-García Daniel, Campbell Lindsay P, Ahadji-Dabla Koffi Mensah, Aboagye-Antwi Fred, Ibarra-Juárez Luis Arturo, Peterson A Townsend

2021-Sep-21

bioacoustics, convolutional neural networks, smartphones, transfer learning, vector-borne diseases

General General

DLAB-Deep learning methods for structure-based virtual screening of antibodies.

In Bioinformatics (Oxford, England)

MOTIVATION : Antibodies are one of the most important classes of pharmaceuticals, with over 80 approved molecules currently in use against a wide variety of diseases. The drug discovery process for antibody therapeutic candidates however is time- and cost-intensive and heavily reliant on in-vivo and in-vitro high throughput screens. Here, we introduce a framework for structure-based deep learning for antibodies (DLAB) which can virtually screen putative binding antibodies against antigen targets of interest. DLAB is built to be able to predict antibody-antigen binding for antigens with no known antibody binders.

RESULTS : We demonstrate that DLAB can be used both to improve antibody-antigen docking and structure-based virtual screening of antibody drug candidates. DLAB enables improved pose ranking for antibody docking experiments as well as selection of antibody-antigen pairings for which accurate poses are generated and correctly ranked. We also show that DLAB can identify binding antibodies against specific antigens in a case study. Our results demonstrate the promise of deep learning methods for structure-based virtual screening of antibodies.

AVAILABILITY : The DLAB source code and pre-trained models are available at https://github.com/oxpig/dlab-public.

SUPPLEMENTARY INFORMATION : Supplementary data are available at Bioinformatics online.

Schneider Constantin, Buchanan Andrew, Taddese Bruck, Deane Charlotte M

2021-Sep-21

General General

Feasibility and Accuracy of Artificial Intelligence-Assisted Sponge Cytology for Community-Based Esophageal Squamous Cell Carcinoma Screening in China.

In The American journal of gastroenterology

INTRODUCTION : Screening is the pivotal strategy to relieve the burden of esophageal squamous cell carcinoma (ESCC) in high-risk areas. The cost, invasiveness, and accessibility of esophagogastroduodenoscopy (EGD) necessitate the development of preliminary screening methods.

METHODS : Residents aged 40-85 years were recruited in a high-risk area of ESCC. Esophageal cells were collected using an approved novel capsule sponge, and cytology slides were scanned by a trained artificial intelligence (AI) system before cytologists provided confirmation. Atypical squamous cell or more severe diagnosis was defined as positive cytology. AI-based abnormal cell counts were also reported. EGD was performed subsequently with biopsy as needed. Diagnostic accuracy, adverse events, and acceptability of cytology testing were assessed. Esophageal high-grade lesions (ESCC and high-grade intraepithelial neoplasia) were the primary target lesions.

RESULTS : In total, 1,844 participants were enrolled, and 20 (1.1%) high-grade lesions were confirmed by endoscopic biopsy. The AI-assisted cytologist-confirmed cytology showed good diagnostic accuracy, with a sensitivity of 90.0% (95% confidence interval [CI], 76.9%-100.0%), specificity of 93.7% (95% CI, 92.6%-94.8%), and positive predictive value of 13.5% (95% CI, 7.70%-19.3%) for detecting high-grade lesions. The area under the receiver operation characteristics curve was 0.926 (95% CI, 0.850-1.000) and 0.949 (95% CI, 0.890-1.000) for AI-assisted cytologist-confirmed cytology and AI-based abnormal cell count, respectively. The numbers of EGD could be reduced by 92.5% (from 99.2 to 7.4 to detect 1 high-grade lesion) if only cytology-positive participants were referred to endoscopy. No serious adverse events were documented during the cell collection process, and 96.1% participants reported this process as acceptable.

DISCUSSION : The AI-assisted sponge cytology is feasible, safe, and acceptable for ESCC screening in community, with high accuracy for detecting esophageal squamous high-grade lesions.

Gao Ye, Xin Lei, Feng Ya-Dong, Yao Bin, Lin Han, Sun Chang, An Wei, Li Zhao-Shen, Shi Rui-Hua, Wang Luo-Wei

2021-Sep-21

General General

Automatic Classification of Thyroid Findings Using Static and Contextualized Ensemble Natural Language Processing Systems: Development Study.

In JMIR medical informatics ; h5-index 23.0

BACKGROUND : In the case of Korean institutions and enterprises that collect nonstandardized and nonunified formats of electronic medical examination results from multiple medical institutions, a group of experienced nurses who can understand the results and related contexts initially classified the reports manually. The classification guidelines were established by years of workers' clinical experiences and there were attempts to automate the classification work. However, there have been problems in which rule-based algorithms or human labor-intensive efforts can be time-consuming or limited owing to high potential errors. We investigated natural language processing (NLP) architectures and proposed ensemble models to create automated classifiers.

OBJECTIVE : This study aimed to develop practical deep learning models with electronic medical records from 284 health care institutions and open-source corpus data sets for automatically classifying 3 thyroid conditions: healthy, caution required, and critical. The primary goal is to increase the overall accuracy of the classification, yet there are practical and industrial needs to correctly predict healthy (negative) thyroid condition data, which are mostly medical examination results, and minimize false-negative rates under the prediction of healthy thyroid conditions.

METHODS : The data sets included thyroid and comprehensive medical examination reports. The textual data are not only documented in fully complete sentences but also written in lists of words or phrases. Therefore, we propose static and contextualized ensemble NLP network (SCENT) systems to successfully reflect static and contextual information and handle incomplete sentences. We prepared each convolution neural network (CNN)-, long short-term memory (LSTM)-, and efficiently learning an encoder that classifies token replacements accurately (ELECTRA)-based ensemble model by training or fine-tuning them multiple times. Through comprehensive experiments, we propose 2 versions of ensemble models, SCENT-v1 and SCENT-v2, with the single-architecture-based CNN, LSTM, and ELECTRA ensemble models for the best classification performance and practical use, respectively. SCENT-v1 is an ensemble of CNN and ELECTRA ensemble models, and SCENT-v2 is a hierarchical ensemble of CNN, LSTM, and ELECTRA ensemble models. SCENT-v2 first classifies the 3 labels using an ELECTRA ensemble model and then reclassifies them using an ensemble model of CNN and LSTM if the ELECTRA ensemble model predicted them as "healthy" labels.

RESULTS : SCENT-v1 outperformed all the suggested models, with the highest F1 score (92.56%). SCENT-v2 had the second-highest recall value (94.44%) and the fewest misclassifications for caution-required thyroid condition while maintaining 0 classification error for the critical thyroid condition under the prediction of the healthy thyroid condition.

CONCLUSIONS : The proposed SCENT demonstrates good classification performance despite the unique characteristics of the Korean language and problems of data lack and imbalance, especially for the extremely low amount of critical condition data. The result of SCENT-v1 indicates that different perspectives of static and contextual input token representations can enhance classification performance. SCENT-v2 has a strong impact on the prediction of healthy thyroid conditions.

Shin Dongyup, Kam Hye Jin, Jeon Min-Seok, Kim Ha Young

2021-Sep-21

convolution neural network, deep learning, electronic medical records, ensemble, long short-term memory, natural language processing, thyroid, transformer, word embedding

Radiology Radiology

A Fully Automated Analytic System for Measuring Endolymphatic Hydrops Ratios in Patients With Ménière Disease via Magnetic Resonance Imaging: Deep Learning Model Development Study.

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : Recently, the analysis of endolymphatic hydropses (EHs) via inner ear magnetic resonance imaging (MRI) for patients with Ménière disease has been attempted in various studies. In addition, artificial intelligence has rapidly been incorporated into the medical field. In our previous studies, an automated algorithm for EH analysis was developed by using a convolutional neural network. However, several limitations existed, and further studies were conducted to compensate for these limitations.

OBJECTIVE : The aim of this study is to develop a fully automated analytic system for measuring EH ratios that enhances EH analysis accuracy and clinical usability when studying Ménière disease via MRI.

METHODS : We proposed the 3into3Inception and 3intoUNet networks. Their network architectures were based on those of the Inception-v3 and U-Net networks, respectively. The developed networks were trained for inner ear segmentation by using the magnetic resonance images of 124 people and were embedded in a new, automated EH analysis system-inner-ear hydrops estimation via artificial intelligence (INHEARIT)-version 2 (INHEARIT-v2). After fivefold cross-validation, an additional test was performed by using 60 new, unseen magnetic resonance images to evaluate the performance of our system. The INHEARIT-v2 system has a new function that automatically selects representative images from a full MRI stack.

RESULTS : The average segmentation performance of the fivefold cross-validation was measured via the intersection of union method, resulting in performance values of 0.743 (SD 0.030) for the 3into3Inception network and 0.811 (SD 0.032) for the 3intoUNet network. The representative magnetic resonance slices (ie, from a data set of unseen magnetic resonance images) that were automatically selected by the INHEARIT-v2 system only differed from a maximum of 2 expert-selected slices. After comparing the ratios calculated by experienced physicians and those calculated by the INHEARIT-v2 system, we found that the average intraclass correlation coefficient for all cases was 0.941; the average intraclass correlation coefficient of the vestibules was 0.968, and that of the cochleae was 0.914. The time required for the fully automated system to accurately analyze EH ratios based on a patient's MRI stack was approximately 3.5 seconds.

CONCLUSIONS : In this study, a fully automated full-stack magnetic resonance analysis system for measuring EH ratios was developed (named INHEARIT-v2), and the results showed that there was a high correlation between the expert-calculated EH ratio values and those calculated by the INHEARIT-v2 system. The system is an upgraded version of the INHEARIT system; it has higher segmentation performance and automatically selects representative images from an MRI stack. The new model can help clinicians by providing objective analysis results and reducing the workload for interpreting magnetic resonance images.

Park Chae Jung, Cho Young Sang, Chung Myung Jin, Kim Yi-Kyung, Kim Hyung-Jin, Kim Kyunga, Ko Jae-Wook, Chung Won-Ho, Cho Baek Hwan

2021-Sep-21

Ménière disease, artificial intelligence, automation, clinical decision support system, clinical usability, clinician support, convolutional neural network, deep learning, end-to-end system, endolymphatic hydrops, image selection, inner ear, machine learning, magnetic resonance imaging, medical image segmentation, multi-class segmentation

General General

Better Performance with Transformer: CPPFormer in precise prediction of cell-Penetrating Peptides.

In Current medicinal chemistry ; h5-index 49.0

With its superior performance, the Transformer model, which is based on the 'Encoder-Decoder' paradigm, has become the mainstream in natural language processing. On the other hand, bioinformatics has embraced machine learning and made great progress in drug design and protein property prediction. Cell-penetrating peptides (CPPs) are one kind of permeable protein that is convenient as a kind of 'postman' in drug penetration tasks. However, a small number of CPPs have been discovered by research, let alone practical applications in drug permeability. Therefore, correctly identifying the CPPs has opened up a new way to take macromolecules into cells without other potentially harmful materials in the drug. Most of the previous work only uses trivial machine learning techniques and hand-crafted features to construct a simple classifier. In CPPFormer, we learn from the idea of implementing the attention structure of Transformer, rebuilding the network based on the characteristics of CPPs according to its short length, and using an automatic feature extractor with a few manual engineered features to co-direct the predicted results. Compared to all previous methods and other classic text classification models, the empirical result has shown that our proposed deep model-based method has achieved the best performance of 92.16% accuracy in the CPP924 dataset and has passed various index tests.

Xue Yuyang, Ye Xiucai, Wei Lesong, Zhang Xin, Sakurai Tetsuya, Wei Leyi

2021-Sep-19

Cell-penetrating peptides, Transformer, classification, deep learning, drug penetration, feature extractor

General General

Non-melanoma skin cancer diagnosis: a comparison between dermoscopic and smartphone images by unified visual and sonification deep learning algorithms.

In Journal of cancer research and clinical oncology

PURPOSE : Non-melanoma skin cancer (NMSC) is the most frequent keratinocyte-origin skin tumor. It is confirmed that dermoscopy of NMSC confers a diagnostic advantage as compared to visual face-to-face assessment. COVID-19 restrictions diagnostics by telemedicine photos, which are analogous to visual inspection, displaced part of in-person visits. This study evaluated by a dual convolutional neural network (CNN) performance metrics in dermoscopic (DI) versus smartphone-captured images (SI) and tested if artificial intelligence narrows the proclaimed gap in diagnostic accuracy.

METHODS : A CNN that receives a raw image and predicts malignancy, overlaid by a second independent CNN which processes a sonification (image-to-sound mapping) of the original image, were combined into a unified malignancy classifier. All images were histopathology-verified in a comparison between NMSC and benign skin lesions excised as suspected NMSCs. Study criteria outcomes were sensitivity and specificity for the unified output.

RESULTS : Images acquired by DI (n = 132 NMSC, n = 33 benign) were compared to SI (n = 170 NMSC, n = 28 benign). DI and SI analysis metrics resulted in an area under the curve (AUC) of the receiver operator characteristic curve of 0.911 and 0.821, respectively. Accuracy was increased by DI (0.88; CI 81.9-92.4) as compared to SI (0.75; CI 68.1-80.6, p < 0.005). Sensitivity of DI was higher than SI (95.3%, CI 90.4-98.3 vs 75.3%, CI 68.1-81.6, p < 0.001), but not specificity (p = NS).

CONCLUSION : Telemedicine use of smartphone images might result in a substantial decrease in diagnostic performance as compared to dermoscopy, which needs to be considered by both healthcare providers and patients.

Dascalu A, Walker B N, Oron Y, David E O

2021-Sep-21

Deep learning, Dermoscopy, Non-melanoma skin cancer, Preventive medicine, Sonification, Telemedicine

Pathology Pathology

Label Cleaning Multiple Instance Learning: Refining Coarse Annotations on Single Whole-Slide Images

ArXiv Preprint

Annotating cancerous regions in whole-slide images (WSIs) of pathology samples plays a critical role in clinical diagnosis, biomedical research, and machine learning algorithms development. However, generating exhaustive and accurate annotations is labor-intensive, challenging, and costly. Drawing only coarse and approximate annotations is a much easier task, less costly, and it alleviates pathologists' workload. In this paper, we study the problem of refining these approximate annotations in digital pathology to obtain more accurate ones. Some previous works have explored obtaining machine learning models from these inaccurate annotations, but few of them tackle the refinement problem where the mislabeled regions should be explicitly identified and corrected, and all of them require a - often very large - number of training samples. We present a method, named Label Cleaning Multiple Instance Learning (LC-MIL), to refine coarse annotations on a single WSI without the need of external training data. Patches cropped from a WSI with inaccurate labels are processed jointly with a MIL framework, and a deep-attention mechanism is leveraged to discriminate mislabeled instances, mitigating their impact on the predictive model and refining the segmentation. Our experiments on a heterogeneous WSI set with breast cancer lymph node metastasis, liver cancer, and colorectal cancer samples show that LC-MIL significantly refines the coarse annotations, outperforming the state-of-the-art alternatives, even while learning from a single slide. These results demonstrate the LC-MIL is a promising, lightweight tool to provide fine-grained annotations from coarsely annotated pathology sets.

Zhenzhen Wang, Aleksander S. Popel, Jeremias Sulam

2021-09-22

Ophthalmology Ophthalmology

The Classification of Six Common Skin Diseases Based on Xiangya-Derm: Development of a Chinese Database for Artificial Intelligence.

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : Skin and subcutaneous disease is the fourth-leading cause of the nonfatal disease burden worldwide and constitutes one of the most common burdens in primary care. However, there is a severe lack of dermatologists, particularly in rural Chinese areas. Furthermore, although artificial intelligence (AI) tools can assist in diagnosing skin disorders from images, the database for the Chinese population is limited.

OBJECTIVE : This study aims to establish a database for AI based on the Chinese population and presents an initial study on six common skin diseases.

METHODS : Each image was captured with either a digital camera or a smartphone, verified by at least three experienced dermatologists and corresponding pathology information, and finally added to the Xiangya-Derm database. Based on this database, we conducted AI-assisted classification research on six common skin diseases and then proposed a network called Xy-SkinNet. Xy-SkinNet applies a two-step strategy to identify skin diseases. First, given an input image, we segmented the regions of the skin lesion. Second, we introduced an information fusion block to combine the output of all segmented regions. We compared the performance with 31 dermatologists of varied experiences.

RESULTS : Xiangya-Derm, as a new database that consists of over 150,000 clinical images of 571 different skin diseases in the Chinese population, is the largest and most diverse dermatological data set of the Chinese population. The AI-based six-category classification achieved a top 3 accuracy of 84.77%, which exceeded the average accuracy of dermatologists (78.15%).

CONCLUSIONS : Xiangya-Derm, the largest database for the Chinese population, was created. The classification of six common skin conditions was conducted based on Xiangya-Derm to lay a foundation for product research.

Huang Kai, Jiang Zixi, Li Yixin, Wu Zhe, Wu Xian, Zhu Wu, Chen Mingliang, Zhang Yu, Zuo Ke, Li Yi, Yu Nianzhou, Liu Siliang, Huang Xing, Su Juan, Yin Mingzhu, Qian Buyue, Wang Xianggui, Chen Xiang, Zhao Shuang

2021-Sep-21

China, artificial intelligence, automatic auxiliary diagnoses, classification, convolutional neural network, dermatology, medical image processing, skin, skin disease

Radiology Radiology

Deep Learning for Adjacent Segment Disease at Preoperative MRI for Cervical Radiculopathy.

In Radiology ; h5-index 91.0

Background Patients who undergo surgery for cervical radiculopathy are at risk for developing adjacent segment disease (ASD). Identifying patients who will develop ASD remains challenging for clinicians. Purpose To develop and validate a deep learning algorithm capable of predicting ASD by using only preoperative cervical MRI in patients undergoing single-level anterior cervical diskectomy and fusion (ACDF). Materials and Methods In this Health Insurance Portability and Accountability Act-compliant study, retrospective chart review was performed for 1244 patients undergoing single-level ACDF in two tertiary care centers. After application of inclusion and exclusion criteria, 344 patients were included, of whom 60% (n = 208) were used for training and 40% for validation (n = 43) and testing (n = 93). A deep learning-based prediction model with 48 convolutional layers was designed and trained by using preoperative T2-sagittal cervical MRI. To validate model performance, a neuroradiologist and neurosurgeon independently provided ASD predictions for the test set. Validation metrics included accuracy, areas under the curve, and F1 scores. The difference in proportion of wrongful predictions between the model and clinician was statistically tested by using the McNemar test. Results A total of 344 patients (median age, 48 years; interquartile range, 41-58 years; 182 women) were evaluated. The model predicted ASD on the 93 test images with an accuracy of 88 of 93 (95%; 95% CI: 90, 99), sensitivity of 12 of 15 (80%; 95% CI: 60, 100), and specificity of 76 of 78 (97%; 95% CI: 94, 100). The neuroradiologist and neurosurgeon provided predictions with lower accuracy (54 of 93; 58%; 95% CI: 48, 68), sensitivity (nine of 15; 60%; 95% CI: 35, 85), and specificity (45 of 78; 58%; 95% CI: 56, 77) compared with the algorithm. The McNemar test on the contingency table demonstrated that the proportion of wrongful predictions was significantly lower by the model (test statistic, 2.000; P < .001). Conclusion A deep learning algorithm that used only preoperative cervical T2-weighted MRI outperformed clinical experts at predicting adjacent segment disease in patients undergoing surgery for cervical radiculopathy. © RSNA, 2021.

Goedmakers Caroline M W, Lak Asad M, Duey Akiro H, Senko Alexander W, Arnaout Omar, Groff Michael W, Smith Timothy R, Vleggeert-Lankamp Carmen L A, Zaidi Hasan A, Rana Aakanksha, Boaro Alessandro

2021-Sep-21

General General

Predicting the Presence and Abundance of Bacterial Taxa in Environmental Communities through Flow Cytometric Fingerprinting.

In mSystems

Microbiome management research and applications rely on temporally resolved measurements of community composition. Current technologies to assess community composition make use of either cultivation or sequencing of genomic material, which can become time-consuming and/or laborious in case high-throughput measurements are required. Here, using data from a shrimp hatchery as an economically relevant case study, we combined 16S rRNA gene amplicon sequencing and flow cytometry data to develop a computational workflow that allows the prediction of taxon abundances based on flow cytometry measurements. The first stage of our pipeline consists of a classifier to predict the presence or absence of the taxon of interest, with yielded an average accuracy of 88.13% ± 4.78% across the top 50 operational taxonomic units (OTUs) of our data set. In the second stage, this classifier was combined with a regression model to predict the relative abundances of the taxon of interest, which yielded an average R2 of 0.35 ± 0.24 across the top 50 OTUs of our data set. Application of the models to flow cytometry time series data showed that the generated models can predict the temporal dynamics of a large fraction of the investigated taxa. Using cell sorting, we validated that the model correctly associates taxa to regions in the cytometric fingerprint, where they are detected using 16S rRNA gene amplicon sequencing. Finally, we applied the approach of our pipeline to two other data sets of microbial ecosystems. This pipeline represents an addition to the expanding toolbox for flow cytometry-based monitoring of bacterial communities and complements the current plating- and marker gene-based methods. IMPORTANCE Monitoring of microbial community composition is crucial for both microbiome management research and applications. Existing technologies, such as plating and amplicon sequencing, can become laborious and expensive when high-throughput measurements are required. In recent years, flow cytometry-based measurements of community diversity have been shown to correlate well with those derived from 16S rRNA gene amplicon sequencing in several aquatic ecosystems, suggesting that there is a link between the taxonomic community composition and phenotypic properties as derived through flow cytometry. Here, we further integrated 16S rRNA gene amplicon sequencing and flow cytometry survey data in order to construct models that enable the prediction of both the presence and the abundances of individual bacterial taxa in mixed communities using flow cytometric fingerprinting. The developed pipeline holds great potential to be integrated into routine monitoring schemes and early warning systems for biotechnological applications.

Heyse Jasmine, Schattenberg Florian, Rubbens Peter, Müller Susann, Waegeman Willem, Boon Nico, Props Ruben

2021-Sep-21

16S rRNA gene amplicon sequencing, aquaculture, cell sorting, flow cytometry, machine learning, microbial community dynamics, monitoring

General General

SCONES: Self-Consistent Neural Network for Protein Stability Prediction Upon Mutation.

In The journal of physical chemistry. B

Engineering proteins to have desired properties by mutating amino acids at specific sites is commonplace. Such engineered proteins must be stable to function. Experimental methods used to determine stability at throughputs required to scan the protein sequence space thoroughly are laborious. To this end, many machine learning based methods have been developed to predict thermodynamic stability changes upon mutation. These methods have been evaluated for symmetric consistency by testing with hypothetical reverse mutations. In this work, we propose transitive data augmentation, evaluating transitive consistency with our new Stransitive data set, and a new machine learning based method, the first of its kind, that incorporates both symmetric and transitive properties into the architecture. Our method, called SCONES, is an interpretable neural network that predicts small relative protein stability changes for missense mutations that do not significantly alter the structure. It estimates a residue's contributions toward protein stability (ΔG) in its local structural environment, and the difference between independently predicted contributions of the reference and mutant residues is reported as ΔΔG. We show that this self-consistent machine learning architecture is immune to many common biases in data sets, relies less on data than existing methods, is robust to overfitting, and can explain a substantial portion of the variance in experimental data.

Samaga Yashas B L, Raghunathan Shampa, Priyakumar U Deva

2021-Sep-21

General General

Fiber-Based Electret Nanogenerator with a Semisupported Structure for Wearable Electronics.

In ACS applied materials & interfaces ; h5-index 147.0

Fiber-based nanogenerators have great potential applications in wearable electronics such as portable nanodevices, e-skin, and artificial intelligence system. Here, we report a kind of fiber-based electret nanogenerator (FENG) with a semisupported core-shell structure. Owing to its unique structure, the open-circuit voltage and short-circuit current of the FENG reach 40 V and 0.6 μA, respectively, under a short working distance (∼25 μm). No obvious degradation of the output performance under a long-time continuous work (>16 h) and different humidity environments (20-95%) is observed, which demonstrates the FENG's good reliability and stability. Many universal materials, such as cotton rope, conductive sewing thread, and polyvinyl chloride tube, have been successfully used to fabricate FENG. Meanwhile, the FENG-based wearable fabric has been successfully developed to effectively harvest mechanical energy of human motion. The FENG is highly effective, reliable, and stable, promoting the development of fiber-based nanogenerators and their applications in self-powered wearable electronics.

Zhang Li’ang, Chen Qianqian, Huang Xiaoyu, Jia Xiaofeng, Cheng Bolang, Wang Longfei, Qin Yong

2021-Sep-21

electret, energy harvesting, fiber-based nanogenerator, nanogenerator, self-powered wearable electronics

General General

Training data distribution significantly impacts the estimation of tissue microstructure with machine learning.

In Magnetic resonance in medicine ; h5-index 66.0

PURPOSE : Supervised machine learning (ML) provides a compelling alternative to traditional model fitting for parameter mapping in quantitative MRI. The aim of this work is to demonstrate and quantify the effect of different training data distributions on the accuracy and precision of parameter estimates when supervised ML is used for fitting.

METHODS : We fit a two- and three-compartment biophysical model to diffusion measurements from in-vivo human brain, as well as simulated diffusion data, using both traditional model fitting and supervised ML. For supervised ML, we train several artificial neural networks, as well as random forest regressors, on different distributions of ground truth parameters. We compare the accuracy and precision of parameter estimates obtained from the different estimation approaches using synthetic test data.

RESULTS : When the distribution of parameter combinations in the training set matches those observed in healthy human data sets, we observe high precision, but inaccurate estimates for atypical parameter combinations. In contrast, when training data is sampled uniformly from the entire plausible parameter space, estimates tend to be more accurate for atypical parameter combinations but may have lower precision for typical parameter combinations.

CONCLUSION : This work highlights that estimation of model parameters using supervised ML depends strongly on the training-set distribution. We show that high precision obtained using ML may mask strong bias, and visual assessment of the parameter maps is not sufficient for evaluating the quality of the estimates.

Gyori Noemi G, Palombo Marco, Clark Christopher A, Zhang Hui, Alexander Daniel C

2021-Sep-21

machine learning, microstructure imaging, model fitting, quantitative MRI, training data distribution

Public Health Public Health

The Security State of the German Health Web: An Exploratory Study.

In Studies in health technology and informatics ; h5-index 23.0

** : The internet has become an important resource for health information and for interactions with healthcare providers. However, information of all types can go through many servers and networks before reaching its intended destination and any of these has the potential to intercept or even manipulate the exchanged information if data's transfer is not adequately protected. As trust is a fundamental concept in healthcare relationships, it is crucial to offer a secure medical website to maintain the same level of trust as provided in a face-to-face meeting. This study provides a first analysis of the SSL/TLS security of and the security headers used within the health-related web limited to web pages in German, the German health web (GHW).

METHODS : testssl.sh and TLS-Scanner were used to analyze the URLs of the 1,000 top-ranked health-related web sites (according to PageRank) for each of the country- code top level domains: ".de", ".at" and ".ch".

RESULTS : Our study revealed that most websites in the GHW are potentially vulnerable to common SSL/TLS security vulnerabilities, offer deprecated SSL/TLS protocol versions and mostly do not implement HTTP security headers at all.

CONCLUSIONS : These findings question the concept of trust within the GHW. Website owners should reconsider the use of outdated SSL/TLS protocol versions for compatibility reasons. Additionally, HTTP security headers should be implemented more consequently to provide additional security aspects. In future work, the authors intend to repeat this study and to incorporate a website's category, i.e. governmental or public health, to get a more detailed view of the GHW's security.

Henn Frederic, Zowalla Richard, Mayer Andreas

2021-Sep-21

consumer health information, cyber security, data security, health information seeking, internet, trust

General General

openMNGlab: Data Analysis Framework for Microneurography - A Technical Report.

In Studies in health technology and informatics ; h5-index 23.0

openMNGlab is an open-source software framework for data analysis, tailored for the specific needs of microneurography - a type of electrophysiological technique particularly important for research on peripheral neural fibers coding. Currently, openMNGlab loads data from Spike2 and Dapsys, which are two major data acquisition solutions. By building on top of the Neo software, openMNGlab can be easily extended to handle the most common electrophysiological data formats. Furthermore, it provides methods for data visualization, fiber tracking, and a modular feature database to extract features for data analysis and machine learning.

Schlebusch Fabian, Kehrein Frederic, Röhrig Rainer, Namer Barbara, Kutafina Ekaterina

2021-Sep-21

Microneurography, data science, membrane potential, neurophysiology

General General

Towards the Representation of Genomic Data in HL7 FHIR and OMOP CDM.

In Studies in health technology and informatics ; h5-index 23.0

High throughput sequencing technologies have facilitated an outburst in biological knowledge over the past decades and thus enables improvements in personalized medicine. In order to support (international) medical research with the combination of genomic and clinical patient data, a standardization and harmonization of these data sources is highly desirable. To support this increasing importance of genomic data, we have created semantic mapping from raw genomic data to both FHIR (Fast Healthcare Interoperability Resources) and OMOP (Observational Medical Outcomes Partnership) CDM (Common Data Model) and analyzed the data coverage of both models. For this, we calculated the mapping score for different data categories and the relative data coverage in both FHIR and OMOP CDM. Our results show, that the patients genomic data can be mapped to OMOP CDM directly from VCF (Variant Call Format) file with a coverage of slightly over 50%. However, using FHIR as intermediate representation does not lead to further information loss as the already stored data in FHIR can be further transformed into OMOP CDM format with almost 100% success. Our findings are in favor of extending OMOP CDM with patient genomic data using ETL to enable the researchers to apply different analysis methods including machine learning algorithms on genomic data.

Peng Yuan, Nassirian Azadeh, Ahmadi Najia, Sedlmayr Martin, Bathelt Franziska

2021-Sep-21

FHIR, Genomic data, OMOP CDM, VCF

General General

Automated Creation of Expert Systems with the InteKRator Toolbox.

In Studies in health technology and informatics ; h5-index 23.0

Expert systems have a long tradition in both medical informatics and artificial intelligence research. Traditionally, such systems are created by implementing knowledge provided by experts in a system that can be queried for answers. To automatically generate such knowledge directly from data, the lightweight InteKRator toolbox will be introduced here, which combines knowledge representation and machine learning approaches. The learned knowledge is represented in the form of rules with exceptions that can be inspected and that are easily comprehensible. An inference module allows for the efficient answering of queries, while at the same time offering the possibility of providing explanations for the inference results. The learned knowledge can be revised manually or automatically with new evidence after learning.

Apeldoorn Daan, Panholzer Torsten

2021-Sep-21

Expert system, knowledge representation, machine learning

General General

Guided structure-based ligand identification and design via artificial intelligence modelling.

In Expert opinion on drug discovery ; h5-index 34.0

The implementation of Artificial Intelligence (AI) methodologies to drug discovery (DD) are on the rise. Several applications have been developed for structure-based DD, where AI methods provide an alternative framework for the identification of ligands for validated therapeutic targets, as well as the de novo design of ligands through generative models.Areas covered: Herein, the authors review the contributions between the 2019 to present period regarding the application of AI methods to structure-based virtual screening (SBVS) which encompasses mainly molecular docking applications - binding pose prediction and binary classification for ligand or hit identification-, as well as de novo drug design driven by machine learning (ML) generative models, and the validation of AI models in structure-based screening. Studies are reviewed in terms of their main objective, used databases, implemented methodology, input and output, before their key results are summarized.Expert opinion: It is evident that the emerging methods driven by AI will become standard in the DD pipeline. With the rise of new applications, more profound analyses regarding the validity and applicability of these methods have begun to appear. In the near future, we expect to see more structure-based generative models- which are scarce in comparison to ligand-based generative models-, the implementation of standard guidelines for validating the generated structures, and more analyses regarding the validation of AI methods in structure-based DD.

Di Filippo Juan I, Cavasotto Claudio N

2021-Sep-21

Artificial intelligence, drug discovery, machine learning, molecular docking, structure-based virtual screening

Radiology Radiology

Immune and cellular damage biomarkers to predict COVID-19 mortality in hospitalized patients.

In Current research in immunology

Early prediction of COVID-19 in-hospital mortality relies usually on patients' preexisting comorbidities and is rarely reproducible in independent cohorts. We wanted to compare the role of routinely measured biomarkers of immunity, inflammation, and cellular damage with preexisting comorbidities in eight different machine-learning models to predict mortality, and evaluate their performance in an independent population. We recruited and followed-up consecutive adult patients with SARS-Cov-2 infection in two different Italian hospitals. We predicted 60-day mortality in one cohort (development dataset, n = 299 patients, of which 80% was allocated to the development dataset and 20% to the training set) and retested the models in the second cohort (external validation dataset, n = 402). Demographic, clinical, and laboratory features at admission, treatments and disease outcomes were significantly different between the two cohorts. Notably, significant differences were observed for %lymphocytes (p < 0.05), international-normalized-ratio (p < 0.01), platelets, alanine-aminotransferase, creatinine (all p < 0.001). The primary outcome (60-day mortality) was 29.10% (n = 87) in the development dataset, and 39.55% (n = 159) in the external validation dataset. The performance of the 8 tested models on the external validation dataset were similar to that of the holdout test dataset, indicating that the models capture the key predictors of mortality. The shap analysis in both datasets showed that age, immune features (%lymphocytes, platelets) and LDH substantially impacted on all models' predictions, while creatinine and CRP varied among the different models. The model with the better performance was model 8 (60-day mortality AUROC 0.83 ± 0.06 in holdout test set, 0.79 ± 0.02 in external validation dataset). The features that had the greatest impact on this model's prediction were age, LDH, platelets, and %lymphocytes, more than comorbidities or inflammation markers, and these findings were highly consistent in both datasets, likely reflecting the virus effect at the very beginning of the disease.

Lombardi Carlo, Roca Elena, Bigni Barbara, Bertozzi Bruno, Ferrandina Camillo, Franzin Alberto, Vivaldi Oscar, Cottini Marcello, D’Alessio Andrea, Del Poggio Paolo, Conte Gian Marco, Berti Alvise

2021-Sep-16

COVID-19, CRP, Coronavirus, In-hospital death, LDH, Lymphocytes, Platelets, SARS-CoV-2

General General

Towards The Automatic Coding of Medical Transcripts to Improve Patient-Centered Communication

ArXiv Preprint

This paper aims to provide an approach for automatic coding of physician-patient communication transcripts to improve patient-centered communication (PCC). PCC is a central part of high-quality health care. To improve PCC, dialogues between physicians and patients have been recorded and tagged with predefined codes. Trained human coders have manually coded the transcripts. Since it entails huge labor costs and poses possible human errors, automatic coding methods should be considered for efficiency and effectiveness. We adopted three machine learning algorithms (Na\"ive Bayes, Random Forest, and Support Vector Machine) to categorize lines in transcripts into corresponding codes. The result showed that there is evidence to distinguish the codes, and this is considered to be sufficient for training of human annotators.

Gilchan Park, Julia Taylor Rayz, Cleveland G. Shields

2021-09-22

Cardiology Cardiology

Application of Pre-Trained Deep Learning Models for Clinical ECGs.

In Studies in health technology and informatics ; h5-index 23.0

Automatic electrocardiogram (ECG) analysis has been one of the very early use cases for computer assisted diagnosis (CAD). Most ECG devices provide some level of automatic ECG analysis. In the recent years, Deep Learning (DL) is increasingly used for this task, with the first models that claim to perform better than human physicians. In this manuscript, a pilot study is conducted to evaluate the added value of such a DL model to existing built-in analysis with respect to clinical relevance. 29 12-lead ECGs have been analyzed with a published DL model and results are compared to build-in analysis and clinical diagnosis. We could not reproduce the results of the test data exactly, presumably due to a different runtime environment. However, the errors were in the order of rounding errors and did not affect the final classification. The excellent performance in detection of left bundle branch block and atrial fibrillation that was reported in the publication could be reproduced. The DL method and the built-in method performed similarly good for the chosen cases regarding clinical relevance. While benefit of the DL method for research can be attested and usage in training can be envisioned, evaluation of added value in clinical practice would require a more comprehensive study with further and more complex cases.

Bender Theresa, Seidler Tim, Bengel Philipp, Sax Ulrich, Krefting Dagmar

2021-Sep-21

Atrial Fibrillation, Classification, Deep Learning, Deep Neural Network, ECG, Left Bundle Branch Block, Reproducibility of Results

General General

Towards Interpretable Machine Learning in EEG Analysis.

In Studies in health technology and informatics ; h5-index 23.0

In this paper a machine learning model for automatic detection of abnormalities in electroencephalography (EEG) is dissected into parts, so that the influence of each part on the classification accuracy score can be examined. The most successful setup of several shallow artificial neural networks aggregated via voting results in accuracy of 81%. Stepwise simplification of the model shows the expected decrease in accuracy, but a naive model with thresholding of a single extracted feature (relative wavelet energy) is still able to achieve 75%, which remains strongly above the random guess baseline of 54%. These results suggest the feasibility of building a simple classification model ensuring accuracy scores close to the state-of-the-art research but remaining fully interpretable.

Mortaga Maged, Brenner Alexander, Kutafina Ekaterina

2021-Sep-21

EEG, decision support techniques, epilepsy, supervised machine learning

General General

Multi-Disease Detection in Retinal Imaging Based on Ensembling Heterogeneous Deep Learning Models.

In Studies in health technology and informatics ; h5-index 23.0

Preventable or undiagnosed visual impairment and blindness affect billion of people worldwide. Automated multi-disease detection models offer great potential to address this problem via clinical decision support in diagnosis. In this work, we proposed an innovative multi-disease detection pipeline for retinal imaging which utilizes ensemble learning to combine the predictive capabilities of several heterogeneous deep convolutional neural network models. Our pipeline includes state-of-the-art strategies like transfer learning, class weighting, real-time image augmentation and Focal loss utilization. Furthermore, we integrated ensemble learning techniques like heterogeneous deep learning models, bagging via 5-fold cross-validation and stacked logistic regression models. Through internal and external evaluation, we were able to validate and demonstrate high accuracy and reliability of our pipeline, as well as the comparability with other state-of-the-art pipelines for retinal disease prediction.

Müller Dominik, Soto-Rey Iñaki, Kramer Frank

2021-Sep-21

Class Imbalance, Deep Learning, Ensemble Learning, Multi-label Image Classification, Retinal Disease Detection

oncology Oncology

Artificial intelligence-enhanced intraoperative neurosurgical workflow: state of the art and future perspectives.

In Journal of neurosurgical sciences ; h5-index 16.0

BACKGROUND : Artificial Intelligence (AI) and Machine Learning (ML) augment decision-making processes and productivity by supporting surgeons over a range of clinical activities: from diagnosis and preoperative planning to intraoperative surgical assistance. We reviewed the literature to identify current AI platforms applied to neurosurgical perioperative and intraoperative settings and describe their role in multiple subspecialties.

METHODS : A systematic review of the literature was conducted following the PRISMA guidelines. PubMed, EMBASE, and Scopus databases were searched from inception to December 31, 2020. Original articles were included if they: presented AI platforms implemented in perioperative, intraoperative settings and reported ML models' performance metrics. Due to the heterogeneity in neurosurgical applications, a qualitative synthesis was deemed appropriate. The risk of bias and applicability of predicted outcomes were assessed using the PROBAST tool.

RESULTS : 41 articles were included. All studies evaluated a supervised learning algorithm. A total of 10 ML models were described; the most frequent were neural networks (n = 15) and tree-based models (n = 13). Overall, the risk of bias was medium-high, but applicability was considered positive for all studies. Articles were grouped into 4 categories according to the subspecialty of interest: neuro-oncology, spine, functional and other. For each category, different prediction tasks were identified.

CONCLUSIONS : In this review, we summarize the state-of-art applications of AI for the intraoperative augmentation of neurosurgical workflows across multiple subspecialties. ML models may boost surgical team performances by reducing human errors and providing patient-tailored surgical plans, but further and higher-quality studies need to be conducted.

Tariciotti Leonardo, Palmisciano Paolo, Giordano Martina, Remoli Giulia, Lacorte Eleonora, Bertani Giulio, Locatelli Marco, Dimeco Francesco, Caccavella Valerio M, Prada Francesco

2021-Sep-21

General General

Deep learning methods for automatic segmentation of lower leg muscles and bones from MRI scans of children with and without cerebral palsy.

In NMR in biomedicine ; h5-index 41.0

Cerebral palsy is a neurological condition that is known to affect muscle growth. Detailed investigations of muscle growth require segmentation of muscles from MRI scans, which is typically done manually. In this study, we evaluated the performance of 2D, 3D, and hybrid deep learning models for automatic segmentation of 11 lower leg muscles and two bones from MRI scans of children with and without cerebral palsy. All six models were trained and evaluated on manually segmented T1 -weighted MRI scans of the lower legs of 20 children, six of whom had cerebral palsy. The segmentation results were assessed using the median Dice similarity coefficient (DSC), average symmetric surface distance (ASSD), and volume error (VError) of all 13 labels of every scan. The best performance was achieved by H-DenseUNet, a hybrid model (DSC 0.90, ASSD 0.5 mm, and VError 2.6 cm3 ). The performance was equivalent to the inter-rater performance of manual segmentation (DSC 0.89, ASSD 0.6 mm, and VError 3.3 cm3 ). Models trained with the Dice loss function outperformed models trained with the cross-entropy loss function. Near-optimal performance could be attained using only 11 scans for training. Segmentation performance was similar for scans of typically developing children (DSC 0.90, ASSD 0.5 mm, and VError 2.8 cm3 ) and children with cerebral palsy (DSC 0.85, ASSD 0.6 mm, and VError 2.4 cm3 ). These findings demonstrate the feasibility of fully automatic segmentation of individual muscles and bones from MRI scans of children with and without cerebral palsy.

Zhu Jiayi, Bolsterlee Bart, Chow Brian V Y, Cai Chengxue, Herbert Robert D, Song Yang, Meijering Erik

2021-Sep-21

Deep learning, MRI, cerebral palsy, lower leg, muscle segmentation

oncology Oncology

Robustness of deep learning segmentation of cardiac substructures in non-contrast computed tomography for breast cancer radiotherapy.

In Medical physics ; h5-index 59.0

PURPOSE : To develop and evaluate deep learning based auto-segmentation of cardiac substructures from non-contrast planning computed tomography (CT) images in patients undergoing breast cancer radiotherapy and to investigate the algorithm sensitivity to out-of-distribution data such as CT image artifacts.

METHODS : Nine substructures including Aortic Valve (AV), Left Anterior Descending (LAD), Tricuspid Valve (TV), Mitral Valve (MV), Pulmonic Valve (PV), Right Atrium (RA), Right Ventricle (RV), Left Atrium (LA) and Left Ventricle (LV) were manually delineated by a radiation oncologist on non-contrast CT images of 129 patients with breast cancer; among them 90 were considered in-distribution data, also named 'clean' data. The image/label pairs of 60 subjects were used to train a 3D deep neural network while the remaining 30 were used for testing. The rest of the 39 patients were considered out-of-distribution ('outlier') data which were used to test robustness. Random rigid transformations were used to augment the dataset during training. We investigated multiple loss functions, including Dice Similarity Coefficient (DSC), Cross-entropy (CE), Euclidean loss as well as the variation and combinations of these, data augmentation, and network size on overall performance and sensitivity to image artifacts due to infrequent events such as the presence of implanted devices. The predicted label maps were compared to the ground truth labels via DSC and mean and 90th percentile symmetric surface distance (90th -SSD).

RESULTS : When using modified Dice combined with cross-entropy (MD-CE) as the loss function, the algorithm achieved a mean DSC = 0.79±0.07 for chambers and a mean DSC = 0.39±0.10 for smaller substructures (valves and LAD). The mean and 90th -SSD for chambers was 2.7±1.4 mm and 6.5±2.8 mm and was 4.1±1.7 mm and 8.6±3.2 mm for smaller substructures. Models with MD-CE, Dice-CE, MD, and weighted CE loss had highest performance, and were statistically similar. Data augmentation did not affect model performances on both clean and outlier data and model robustness was susceptible to network size. For a certain type of outlier data, robustness can be improved via incorporating them into the training process. The execution time for segmenting each patient was on average 2.1s.

CONCLUSIONS : A deep neural network provides a fast and accurate segmentation of large cardiac substructures in non-contrast CT images. Model robustness to two types of clinically common outlier data were investigated and potential approaches to improve them were explored. Evaluation of clinical acceptability and integration into clinical workflow are pending. This article is protected by copyright. All rights reserved.

Jin Xiyao, Thomas Maria A, Dise Joseph, Kavanaugh James, Hilliard Jessica, Zoberi Imran, Robinson Clifford G, Hugo Geoffrey D

2021-Sep-20

breast radiotherapy, cardiac substructures, deep learning, image segmentation, model robustness

General General

Predicting ecological footprint based on global macro indicators in G-20 countries using machine learning approaches.

In Environmental science and pollution research international

Paying attention to human activities in terms of land grazing infrastructure, crops, forest products, and carbon impact, the so-called ecological impact (EF) is one of the most important economic issues in the world. For the present study, global database data were used. The ability of the penalized regression (RR) approaches (including Ridge, Lasso and Elastic Net) and artificial neural network (ANN) to predict EF indices in the G-20 countries over the past two decades (1999-2018) was illustrated and compared. For this purpose, 10-fold cross-validation was used to evaluate the predictive performance and determine the penalty parameter for PR models. According to the results, the predictive performance compared to linear regression improved somewhat using the penalized methods. Using the elastic net model, more global macro indices were selected than Lasso. Although Lasso selected only a few indicators, it had better predictive performance among PR ns models. In addition to relative improvement in the predictive performance of PR methods, their interest in selecting a subset of indicators by shrinking coefficients and creating a parsimonious model was evident and significant. As a result, PR methods would be preferred, using variable selection and interpretive considerations to predictive performance alone. On the other hand, ANN models with higher determination coefficients (R2) and lower RMSE values performed significantly better than PR and OLS and showed that they were more accurate in predicting EF. Therefore, ANN could provide considerable and appropriate predictions for EF indicators in the G-20 countries.

Roumiani Ahmad, Mofidi Abbas

2021-Sep-21

Ecological footprint, G-20 countries, Machine learning, Prediction, Variable selection

Radiology Radiology

Machine Learning in the Differentiation of Soft Tissue Neoplasms: Comparison of Fat-Suppressed T2WI and Apparent Diffusion Coefficient (ADC) Features-Based Models.

In Journal of digital imaging

Machine learning has been widely used in the characterization of tumors recently. This article aims to explore the feasibility of the whole tumor fat-suppressed (FS) T2WI and ADC features-based least absolute shrinkage and selection operator (LASSO)-logistic predictive models in the differentiation of soft tissue neoplasms (STN). The clinical and MR findings of 160 cases with 161 histologically proven STN were reviewed, retrospectively, 75 with diffusion-weighted imaging (DWI with b values of 50, 400, and 800 s/mm2). They were divided into benign and malignant groups and further divided into training (70%) and validation (30%) cohorts. The MR FS T2WI and ADC features-based LASSO-logistic models were built and compared. The AUC of the FS T2WI features-based LASSO-logistic regression model for benign and malignant prediction was 0.65 and 0.75 for the training and validation cohorts. The model's sensitivity, specificity, and accuracy of the validation cohort were 55%, 96%, and 76.6%. While the AUC of the ADC features-based model was 0.932 and 0.955 for the training and validation cohorts. The model's sensitivity, specificity, and accuracy were 83.3%, 100%, and 91.7%. The performances of these models were also validated by decision curve analysis (DCA). The AUC of the whole tumor ADC features-based LASSO-logistic regression predictive model was larger than that of FS T2WI features (p = 0.017). The whole tumor fat-suppressed T2WI and ADC features-based LASSO-logistic predictive models both can serve as useful tools in the differentiation of STN. ADC features-based LASSO-logistic regression predictive model did better than that of FS T2WI features.

Hu Peian, Chen Lei, Zhou Zhengrong

2021-Sep-20

Apparent diffusion coefficient (ADC), Diffusion MR weighted imaging, Least absolute shrinkage and selection operator (LASSO), Soft tissue neoplasms (STN), Texture analysis (TA)

General General

Evaluation of ultrasonic fibrosis diagnostic system using convolutional network for ordinal regression.

In International journal of computer assisted radiology and surgery

PURPOSE : Diagnosis of liver fibrosis is important for establishing treatment and assessing the risk of carcinogenesis. Ultrasound imaging is an excellent diagnostic method as a screening test in terms of non-invasiveness and simplicity. The purpose of this study was to automatically diagnose liver fibrosis using ultrasound images to reduce the burden on physicians.

METHODS : We proposed and implemented a system for extracting regions of liver parenchyma utilizing U-Net. Using regions of interest, the stage of fibrosis was classified as F0, F1, F2, F3, or F4 utilizing CORALNet, an ordinal regression model based on ResNet18. The effectiveness of the proposed system was verified.

RESULTS : The system implemented using U-Net had a maximum mean Dice coefficient of 0.929. The results of classification of liver fibrosis utilizing CORALNet had a mean absolute error (MAE) of 1.22 and root mean square error (RMSE) of 1.60. The per-case results had a MAE of 1.55 and RMSE of 1.34.

CONCLUSION : U-Net extracted regions of liver parenchyma from the images with high accuracy, and CORALNet showed effectiveness using ordinal information to classify fibrosis in the images. As a future task, we will study a model that is less dependent on teaching data.

Saito Ryosuke, Koizumi Norihiro, Nishiyama Yu, Imaizumi Tsubasa, Kusahara Kenta, Yagasaki Shiho, Matsumoto Naoki, Masuzaki Ryota, Takahashi Toshimi, Ogawa Masahiro

2021-Sep-21

Deep learning, Liver fibrosis, Ultrasound image

Radiology Radiology

Stakeholders' perspectives on the future of artificial intelligence in radiology: a scoping review.

In European radiology ; h5-index 62.0

OBJECTIVES : Artificial intelligence (AI) has the potential to impact clinical practice and healthcare delivery. AI is of particular significance in radiology due to its use in automatic analysis of image characteristics. This scoping review examines stakeholder perspectives on AI use in radiology, the benefits, risks, and challenges to its integration.

METHODS : A search was conducted from 1960 to November 2019 in EMBASE, PubMed/MEDLINE, Web of Science, Cochrane Library, CINAHL, and grey literature. Publications reflecting stakeholder attitudes toward AI were included with no restrictions.

RESULTS : Commentaries (n = 32), surveys (n = 13), presentation abstracts (n = 8), narrative reviews (n = 8), and a social media study (n = 1) were included from 62 eligible publications. These represent the views of radiologists, surgeons, medical students, patients, computer scientists, and the general public. Seven themes were identified (predicted impact, potential replacement, trust in AI, knowledge of AI, education, economic considerations, and medicolegal implications). Stakeholders anticipate a significant impact on radiology, though replacement of radiologists is unlikely in the near future. Knowledge of AI is limited for non-computer scientists and further education is desired. Many expressed the need for collaboration between radiologists and AI specialists to successfully improve patient care.

CONCLUSIONS : Stakeholder views generally suggest that AI can improve the practice of radiology and consider the replacement of radiologists unlikely. Most stakeholders identified the need for education and training on AI, as well as collaborative efforts to improve AI implementation. Further research is needed to gain perspectives from non-Western countries, non-radiologist stakeholders, on economic considerations, and medicolegal implications.

KEY POINTS : Stakeholders generally expressed that AI alone cannot be used to replace radiologists. The scope of practice is expected to shift with AI use affecting areas from image interpretation to patient care. Patients and the general public do not know how to address potential errors made by AI systems while radiologists believe that they should be "in-the-loop" in terms of responsibility. Ethical accountability strategies must be developed across governance levels. Students, residents, and radiologists believe that there is a lack in AI education during medical school and residency. The radiology community should work with IT specialists to ensure that AI technology benefits their work and centres patients.

Yang Ling, Ene Ioana Cezara, Arabi Belaghi Reza, Koff David, Stein Nina, Santaguida Pasqualina Lina

2021-Sep-21

Artificial intelligence, Attitude, Education, Ethics, Radiology

General General

Image-Based Artificial Intelligence in Wound Assessment: A Systematic Review.

In Advances in wound care

SIGNIFICANCE : Accurately predicting wound healing trajectories is difficult for wound care clinicians due to the complex and dynamic processes involved in wound healing. Wound care teams capture images of wounds during clinical visits generating big data sets over time. Developing novel artificial intelligence (AI) systems can help clinicians diagnose, assess the effectiveness of therapy, and predict healing outcomes. Recent Advances: Rapid developments in computer processing have enabled the development of AI-based systems that can improve the diagnosis and effectiveness of therapy in various clinical specializations. In the past decade, we have witnessed AI revolutionizing all types of medical imaging like X-ray, ultrasound, CT, MRI, etc., but AI-based systems remain to be developed clinically and computationally for high-quality wound care that can result in better patient outcomes.

CRITICAL ISSUES : In the current standard of care, collecting wound images on every clinical visit, interpreting and archiving the data are cumbersome and time-consuming. Commercial platforms are developed to capture images, perform wound measurements, and provide clinicians with a workflow for diagnosis, but AI-based systems are still in their infancy. This systematic review summarizes the breadth and depth of the most recent and relevant work in intelligent image-based data analysis and system developments for wound assessment.

FUTURE DIRECTIONS : With increasing availabilities of massive data (wound images, wound-specific electronic health records, etc.) as well as powerful computing resources, AI-based digital platforms will play a significant role in delivering data-driven care to people suffering from debilitating chronic wounds.

Anisuzzaman D M, Wang Chuanbo, Rostami Behrouz, Gopalakrishnan Sandeep, Niezgoda Jeffrey, Yu Zeyun

2021-Sep-21

Public Health Public Health

Using informative features in machine learning based method for COVID-19 drug repurposing.

In Journal of cheminformatics

Coronavirus disease 2019 (COVID-19) is caused by a novel virus named Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2). This virus induced a large number of deaths and millions of confirmed cases worldwide, creating a serious danger to public health. However, there are no specific therapies or drugs available for COVID-19 treatment. While new drug discovery is a long process, repurposing available drugs for COVID-19 can help recognize treatments with known clinical profiles. Computational drug repurposing methods can reduce the cost, time, and risk of drug toxicity. In this work, we build a graph as a COVID-19 related biological network. This network is related to virus targets or their associated biological processes. We select essential proteins in the constructed biological network that lead to a major disruption in the network. Our method from these essential proteins chooses 93 proteins related to COVID-19 pathology. Then, we propose multiple informative features based on drug-target and protein-protein interaction information. Through these informative features, we find five appropriate clusters of drugs that contain some candidates as potential COVID-19 treatments. To evaluate our results, we provide statistical and clinical evidence for our candidate drugs. From our proposed candidate drugs, 80% of them were studied in other studies and clinical trials.

Aghdam Rosa, Habibi Mahnaz, Taheri Golnaz

2021-Sep-20

Clustering method, Coronavirus disease 2019, Protein−protein interaction, SARS-CoV-2

General General

The First Vision For Vitals (V4V) Challenge for Non-Contact Video-Based Physiological Estimation

ArXiv Preprint

Telehealth has the potential to offset the high demand for help during public health emergencies, such as the COVID-19 pandemic. Remote Photoplethysmography (rPPG) - the problem of non-invasively estimating blood volume variations in the microvascular tissue from video - would be well suited for these situations. Over the past few years a number of research groups have made rapid advances in remote PPG methods for estimating heart rate from digital video and obtained impressive results. How these various methods compare in naturalistic conditions, where spontaneous behavior, facial expressions, and illumination changes are present, is relatively unknown. To enable comparisons among alternative methods, the 1st Vision for Vitals Challenge (V4V) presented a novel dataset containing high-resolution videos time-locked with varied physiological signals from a diverse population. In this paper, we outline the evaluation protocol, the data used, and the results. V4V is to be held in conjunction with the 2021 International Conference on Computer Vision.

Ambareesh Revanur, Zhihua Li, Umur A. Ciftci, Lijun Yin, Laszlo A. Jeni

2021-09-22

General General

The First Vision For Vitals (V4V) Challenge for Non-Contact Video-Based Physiological Estimation

ArXiv Preprint

Telehealth has the potential to offset the high demand for help during public health emergencies, such as the COVID-19 pandemic. Remote Photoplethysmography (rPPG) - the problem of non-invasively estimating blood volume variations in the microvascular tissue from video - would be well suited for these situations. Over the past few years a number of research groups have made rapid advances in remote PPG methods for estimating heart rate from digital video and obtained impressive results. How these various methods compare in naturalistic conditions, where spontaneous behavior, facial expressions, and illumination changes are present, is relatively unknown. To enable comparisons among alternative methods, the 1st Vision for Vitals Challenge (V4V) presented a novel dataset containing high-resolution videos time-locked with varied physiological signals from a diverse population. In this paper, we outline the evaluation protocol, the data used, and the results. V4V is to be held in conjunction with the 2021 International Conference on Computer Vision.

Ambareesh Revanur, Zhihua Li, Umur A. Ciftci, Lijun Yin, Laszlo A. Jeni

2021-09-22

Radiology Radiology

Immune and cellular damage biomarkers to predict COVID-19 mortality in hospitalized patients.

In Current research in immunology

Early prediction of COVID-19 in-hospital mortality relies usually on patients' preexisting comorbidities and is rarely reproducible in independent cohorts. We wanted to compare the role of routinely measured biomarkers of immunity, inflammation, and cellular damage with preexisting comorbidities in eight different machine-learning models to predict mortality, and evaluate their performance in an independent population. We recruited and followed-up consecutive adult patients with SARS-Cov-2 infection in two different Italian hospitals. We predicted 60-day mortality in one cohort (development dataset, n = 299 patients, of which 80% was allocated to the development dataset and 20% to the training set) and retested the models in the second cohort (external validation dataset, n = 402). Demographic, clinical, and laboratory features at admission, treatments and disease outcomes were significantly different between the two cohorts. Notably, significant differences were observed for %lymphocytes (p < 0.05), international-normalized-ratio (p < 0.01), platelets, alanine-aminotransferase, creatinine (all p < 0.001). The primary outcome (60-day mortality) was 29.10% (n = 87) in the development dataset, and 39.55% (n = 159) in the external validation dataset. The performance of the 8 tested models on the external validation dataset were similar to that of the holdout test dataset, indicating that the models capture the key predictors of mortality. The shap analysis in both datasets showed that age, immune features (%lymphocytes, platelets) and LDH substantially impacted on all models' predictions, while creatinine and CRP varied among the different models. The model with the better performance was model 8 (60-day mortality AUROC 0.83 ± 0.06 in holdout test set, 0.79 ± 0.02 in external validation dataset). The features that had the greatest impact on this model's prediction were age, LDH, platelets, and %lymphocytes, more than comorbidities or inflammation markers, and these findings were highly consistent in both datasets, likely reflecting the virus effect at the very beginning of the disease.

Lombardi Carlo, Roca Elena, Bigni Barbara, Bertozzi Bruno, Ferrandina Camillo, Franzin Alberto, Vivaldi Oscar, Cottini Marcello, D’Alessio Andrea, Del Poggio Paolo, Conte Gian Marco, Berti Alvise

2021-Sep-16

COVID-19, CRP, Coronavirus, In-hospital death, LDH, Lymphocytes, Platelets, SARS-CoV-2

General General

Gastrointestinal Tract Disease Classification from Wireless Endoscopy Images Using Pretrained Deep Learning Model.

In Computational and mathematical methods in medicine

Wireless capsule endoscopy is a noninvasive wireless imaging technology that becomes increasingly popular in recent years. One of the major drawbacks of this technology is that it generates a large number of photos that must be analyzed by medical personnel, which takes time. Various research groups have proposed different image processing and machine learning techniques to classify gastrointestinal tract diseases in recent years. Traditional image processing algorithms and a data augmentation technique are combined with an adjusted pretrained deep convolutional neural network to classify diseases in the gastrointestinal tract from wireless endoscopy images in this research. We take advantage of pretrained models VGG16, ResNet-18, and GoogLeNet, a convolutional neural network (CNN) model with adjusted fully connected and output layers. The proposed models are validated with a dataset consisting of 6702 images of 8 classes. The VGG16 model achieved the highest results with 96.33% accuracy, 96.37% recall, 96.5% precision, and 96.5% F1-measure. Compared to other state-of-the-art models, the VGG16 model has the highest Matthews Correlation Coefficient value of 0.95 and Cohen's kappa score of 0.96.

Yogapriya J, Chandran Venkatesan, Sumithra M G, Anitha P, Jenopaul P, Suresh Gnana Dhas C

2021

General General

Application of Various Machine Learning Techniques in Predicting Total Organic Carbon from Well Logs.

In Computational intelligence and neuroscience

Unconventional resources have recently gained a lot of attention, and as a consequence, there has been an increase in research interest in predicting total organic carbon (TOC) as a crucial quality indicator. TOC is commonly measured experimentally; however, due to sampling restrictions, obtaining continuous data on TOC is difficult. Therefore, different empirical correlations for TOC have been presented. However, there are concerns about the generalization and accuracy of these correlations. In this paper, different machine learning (ML) techniques were utilized to develop models that predict TOC from well logs, including formation resistivity (FR), spontaneous potential (SP), sonic transit time (Δt), bulk density (RHOB), neutron porosity (CNP), gamma ray (GR), and spectrum logs of thorium (Th), uranium (Ur), and potassium (K). Over 1250 data points from the Devonian Duvernay shale were utilized to create and validate the model. These datasets were obtained from three wells; the first was used to train the models, while the data sets from the other two wells were utilized to test and validate them. Support vector machine (SVM), random forest (RF), and decision tree (DT) were the ML approaches tested, and their predictions were contrasted with three empirical correlations. Various AI methods' parameters were tested to assure the best possible accuracy in terms of correlation coefficient (R) and average absolute percentage error (AAPE) between the actual and predicted TOC. The three ML methods yielded good matches; however, the RF-based model has the best performance. The RF model was able to predict the TOC for the different datasets with R values range between 0.93 and 0.99 and AAPE values less than 14%. In terms of average error, the ML-based models outperformed the other three empirical correlations. This study shows the capability and robustness of ML models to predict the total organic carbon from readily available logging data without the need for core analysis or additional well interventions.

Siddig Osama, Ibrahim Ahmed Farid, Elkatatny Salaheldin

2021

General General

Effect of Basicity on the Microstructure of Sinter and Its Application Based on Deep Learning.

In Computational intelligence and neuroscience

The influence of the evolution rule of basicity (0.6∼2.4) on the mineral composition and microstructure of sinter is studied by using a polarizing microscope, and the comprehensive application analysis of the drum index, vertical sintering speed, and yield of sinter shows that, over the course of an increase in basicity (0.6∼1.0), the mineral structure changed from the original porphyritic-granular structure to a porphyritic structure. At the same time, there was no calcium ferrite phase in the bonding phase at a basicity of less than 1.0; therefore, the downward trend of the three indicators is obvious. When the basicity was further increased to approximately 1.6, the main structure of the mineral phase changed from a corrosion structure to an interweaving corrosion structure. Because of the existence of a porphyritic-granular structure, the structure of the mineral phase was extremely inhomogeneous and most complex near the basicity of 1.6; although a small amount of calcium ferrite displayed an acicular structure, the drum index appeared to show a very low value. With an increase in basicity to 2.0, the mineral phase structure was dominated by an interweaving corrosion structure with a uniform framework, and the content of calcium ferrite reached the highest value. Moreover, a clear acicular structure developed, and the drum index also increased to the highest value. At a basicity of more than 2.0, a mineral structure began to appear and the corrosion, porphyritic-granular structure, and the drum index also showed a slightly declining trend. Therefore, in the actual production process, basicity should be avoided as far as possible at around 1.0 and 1.6 and it should be controlled at around 2.0. At the same time, based on the mineral facies data set of this paper, the convolutional neural network is used to carry out a simple prediction model experiment on the basicity corresponding to the mineral facies photos, and the effect is quite good, which provides a new idea and method for the follow-up study of mineral facies.

Zhi Jian-Ming, Li Jie, Wang Jia-Hao, Jiang Tian-Yu, Hua Ze-Yi

2021

General General

Enhanced sentiment extraction architecture for social media content analysis using capsule networks.

In Multimedia tools and applications

Recent research has produced efficient algorithms based on deep learning for text-based analytics. Such architectures could be readily applied to text-based social media content analysis. The deep learning techniques, which require comparatively fewer resources for language modeling, can be effectively used to process social media content data that change regularly. Convolutional Neural networks and recurrent neural networks based approaches have reported prominent performance in this domain, yet their limitations make them sub-optimal. Capsule networks sufficiently warrant their applicability in language modelling tasks as a promising technique beyond their initial usage of image classification. This study proposes an approach based on capsule networks for social media content analysis, especially for Twitter. We empirically show that our approach is optimal even without the use of any linguistic resources. The proposed architectures produced an accuracy of 86.87% for the Twitter Sentiment Gold dataset and an accuracy of 82.04% for the CrowdFlower US Airline dataset, indicating state-of-the-art performance. Hence, the research findings indicate noteworthy accuracy enhancement for text processing within social media content analysis.

Demotte P, Wijegunarathna K, Meedeniya D, Perera I

2021-Sep-16

Capsule networks, Deep learning, Sentiment analysis, Social media content analysis, Twitter

Dermatology Dermatology

Non-invasive diagnostic tool for Parkinson's disease by sebum RNA profile with machine learning.

In Scientific reports ; h5-index 158.0

Parkinson's disease (PD) is a progressive neurodegenerative disease presenting with motor and non-motor symptoms, including skin disorders (seborrheic dermatitis, bullous pemphigoid, and rosacea), skin pathological changes (decreased nerve endings and alpha-synuclein deposition), and metabolic changes of sebum. Recently, a transcriptome method using RNA in skin surface lipids (SSL-RNAs) which can be obtained non-invasively with an oil-blotting film was reported as a novel analytic method of sebum. Here we report transcriptome analyses using SSL-RNAs and the potential of these expression profiles with machine learning as diagnostic biomarkers for PD in double cohorts (PD [n = 15, 50], controls [n = 15, 50]). Differential expression analysis between the patients with PD and healthy controls identified more than 100 differentially expressed genes in the two cohorts. In each cohort, several genes related to oxidative phosphorylation were upregulated, and gene ontology analysis using differentially expressed genes revealed functional processes associated with PD. Furthermore, machine learning using the expression information obtained from the SSL-RNAs was able to efficiently discriminate patients with PD from healthy controls, with an area under the receiver operating characteristic curve of 0.806. This non-invasive gene expression profile of SSL-RNAs may contribute to early PD diagnosis based on the neurodegeneration background.

Uehara Yuya, Ueno Shin-Ichi, Amano-Takeshige Haruka, Suzuki Shuji, Imamichi Yoko, Fujimaki Motoki, Ota Noriyasu, Murase Takatoshi, Inoue Takayoshi, Saiki Shinji, Hattori Nobutaka

2021-Sep-20

General General

Enabling precision rehabilitation interventions using wearable sensors and machine learning to track motor recovery.

In NPJ digital medicine

The need to develop patient-specific interventions is apparent when one considers that clinical studies often report satisfactory motor gains only in a portion of participants. This observation provides the foundation for "precision rehabilitation". Tracking and predicting outcomes defining the recovery trajectory is key in this context. Data collected using wearable sensors provide clinicians with the opportunity to do so with little burden on clinicians and patients. The approach proposed in this paper relies on machine learning-based algorithms to derive clinical score estimates from wearable sensor data collected during functional motor tasks. Sensor-based score estimates showed strong agreement with those generated by clinicians. Score estimates of upper-limb impairment severity and movement quality were marked by a coefficient of determination of 0.86 and 0.79, respectively. The application of the proposed approach to monitoring patients' responsiveness to rehabilitation is expected to contribute to the development of patient-specific interventions, aiming to maximize motor gains.

Adans-Dester Catherine, Hankov Nicolas, O’Brien Anne, Vergara-Diaz Gloria, Black-Schaffer Randie, Zafonte Ross, Dy Jennifer, Lee Sunghoon I, Bonato Paolo

2020-Sep-21

General General

Deep Learning and Its Application to Function Approximation for MR in Medicine: An Overview.

In Magnetic resonance in medical sciences : MRMS : an official journal of Japan Society of Magnetic Resonance in Medicine

This article presents an overview of deep learning (DL) and its applications to function approximation for MR in medicine. The aim of this article is to help readers develop various applications of DL. DL has made a large impact on the literature of many medical sciences, including MR. However, its technical details are not easily understandable for non-experts of machine learning (ML).The first part of this article presents an overview of DL and its related technologies, such as artificial intelligence (AI) and ML. AI is explained as a function that can receive many inputs and produce many outputs. ML is a process of fitting the function to training data. DL is a kind of ML, which uses a composite of many functions to approximate the function of interest. This composite function is called a deep neural network (DNN), and the functions composited into a DNN are called layers. This first part also covers the underlying technologies required for DL, such as loss functions, optimization, initialization, linear layers, non-linearities, normalization, recurrent neural networks, regularization, data augmentation, residual connections, autoencoders, generative adversarial networks, model and data sizes, and complex-valued neural networks.The second part of this article presents an overview of the applications of DL in MR and explains how functions represented as DNNs are applied to various applications, such as RF pulse, pulse sequence, reconstruction, motion correction, spectroscopy, parameter mapping, image synthesis, and segmentation.

Takeshima Hidenori

2021-Sep-17

artificial intelligence, deep learning, machine learning

Cardiology Cardiology

Deep learning model to detect significant aortic regurgitation using electrocardiography: Detection model for aortic regurgitation.

In Journal of cardiology

BACKGROUND : Aortic regurgitation (AR) is a common heart disease, with a relatively high prevalence of 4.9% in the Framingham Heart Study. Because the prevalence increases with advancing age, an upward shift in the age distribution may increase the burden of AR. To provide an effective screening method for AR, we developed a deep learning-based artificial intelligence algorithm for the diagnosis of significant AR using electrocardiography (ECG).

METHODS : Our dataset comprised 29,859 paired data of ECG and echocardiography, including 412 AR cases, from January 2015 to December 2019. This dataset was divided into training, validation, and test datasets. We developed a multi-input neural network model, which comprised a two-dimensional convolutional neural network (2D-CNN) using raw ECG data and a fully connected deep neural network (FC-DNN) using ECG features, and compared its performance with the performances of a 2D-CNN model and other machine learning models. In addition, we used gradient-weighted class activation mapping (Grad-CAM) to identify which parts of ECG waveforms had the most effect on algorithm decision making.

RESULTS : The area under the receiver operating characteristic curve of the multi-input model (0.802; 95% CI, 0.762-0.837) was significantly greater than that of the 2D-CNN model alone (0.734; 95% CI, 0.679-0.783; p<0.001) and those of other machine learning models. Grad-CAM demonstrated that the multi-input model tended to focus on the QRS complex in leads I and aVL when detecting AR.

CONCLUSIONS : The multi-input deep learning model using 12-lead ECG data could detect significant AR with modest predictive value.

Sawano Shinnosuke, Kodera Satoshi, Katsushika Susumu, Nakamoto Mitsuhiko, Ninomiya Kota, Shinohara Hiroki, Higashikuni Yasutomi, Nakanishi Koki, Nakao Tomoko, Seki Tomohisa, Takeda Norifumi, Fujiu Katsuhito, Daimon Masao, Akazawa Hiroshi, Morita Hiroyuki, Komuro Issei

2021-Sep-17

Aortic regurgitation, Artificial intelligence, Deep learning, Electrocardiography

Public Health Public Health

Social Determinants in Machine Learning Cardiovascular Disease Prediction Models: A Systematic Review.

In American journal of preventive medicine ; h5-index 75.0

INTRODUCTION : Cardiovascular disease is the leading cause of death worldwide, and cardiovascular disease burden is increasing in low-resource settings and for lower socioeconomic groups. Machine learning algorithms are being developed rapidly and incorporated into clinical practice for cardiovascular disease prediction and treatment decisions. Significant opportunities for reducing death and disability from cardiovascular disease worldwide lie with accounting for the social determinants of cardiovascular outcomes. This study reviews how social determinants of health are being included in machine learning algorithms to inform best practices for the development of algorithms that account for social determinants.

METHODS : A systematic review using 5 databases was conducted in 2020. English language articles from any location published from inception to April 10, 2020, which reported on the use of machine learning for cardiovascular disease prediction that incorporated social determinants of health, were included.

RESULTS : Most studies that compared machine learning algorithms and regression showed increased performance of machine learning, and most studies that compared performance with or without social determinants of health showed increased performance with them. The most frequently included social determinants of health variables were gender, race/ethnicity, marital status, occupation, and income. Studies were largely from North America, Europe, and China, limiting the diversity of the included populations and variance in social determinants of health.

DISCUSSION : Given their flexibility, machine learning approaches may provide an opportunity to incorporate the complex nature of social determinants of health. The limited variety of sources and data in the reviewed studies emphasize that there is an opportunity to include more social determinants of health variables, especially environmental ones, that are known to impact cardiovascular disease risk and that recording such data in electronic databases will enable their use.

Zhao Yuan, Wood Erica P, Mirin Nicholas, Cook Stephanie H, Chunara Rumi

2021-Oct

Cardiology Cardiology

Machine Learning-Derived Echocardiographic Phenotypes Predict Heart Failure Incidence in Asymptomatic Individuals.

In JACC. Cardiovascular imaging

OBJECTIVES : This study sought to identify homogenous echocardiographic phenotypes in community-based cohorts and assess their association with outcomes.

BACKGROUND : Asymptomatic cardiac dysfunction leads to a high risk of long-term cardiovascular morbidity and mortality; however, better echocardiographic classification of asymptomatic individuals remains a challenge.

METHODS : Echocardiographic phenotypes were identified using K-means clustering in the first generation of the STANISLAS (Yearly non-invasive follow-up of Health status of Lorraine insured inhabitants) cohort (N = 827; mean age: 60 ± 5 years; men: 48%), and their associations with vascular function and circulating biomarkers were also assessed. These phenotypes were externally validated in the Malmö Preventive Project cohort (N = 1,394; mean age: 67 ± 6 years; men: 70%), and their associations with the composite of cardiovascular mortality (CVM) or heart failure hospitalization (HFH) were assessed as well.

RESULTS : Three echocardiographic phenotypes were identified as "mostly normal (MN)" (n = 334), "diastolic changes (D)" (n = 323), and "diastolic changes with structural remodeling (D/S)" (n = 170). The D and D/S phenotypes had similar ages, body mass indices, cardiovascular risk factors, vascular impairments, and diastolic function changes. The D phenotype consisted mainly of women and featured increased levels of inflammatory biomarkers, whereas the D/S phenotype, consisted predominantly of men, displayed the highest values of left ventricular mass, volume, and remodeling biomarkers. The phenotypes were predicted based on a simple algorithm including e', left ventricular mass and volume (e'VM algorithm). In the Malmö cohort, subgroups derived from e'VM algorithm were significantly associated with a higher risk of CVM and HFH (adjusted HR in the D phenotype = 1.87; 95% CI: 1.04 to 3.37; adjusted HR in the D/S phenotype = 3.02; 95% CI: 1.71 to 5.34).

CONCLUSIONS : Among asymptomatic, middle-aged individuals, echocardiographic data-driven classification based on the simple e'VM algorithm identified profiles with different long-term HF risk. (4th Visit at 17 Years of Cohort STANISLAS-Stanislas Ancillary Study ESCIF [STANISLASV4]; NCT01391442).

Kobayashi Masatake, Huttin Olivier, Magnusson Martin, Ferreira João Pedro, Bozec Erwan, Huby Anne-Cecile, Preud’homme Gregoire, Duarte Kevin, Lamiral Zohra, Dalleau Kevin, Bresso Emmanuel, Smaïl-Tabbone Malika, Devignes Marie-Dominique, Nilsson Peter M, Leosdottir Margret, Boivin Jean-Marc, Zannad Faiez, Rossignol Patrick, Girerd Nicolas

2021-Sep-08

biomarkers, cardiovascular diseases, cluster analysis, echocardiogram, heart failure, machine learning, prognosis

General General

Artificial Intelligence Predicts Severity of COVID-19 Based on Correlation of Exaggerated Monocyte Activation, Excessive Organ Damage and Hyperinflammatory Syndrome: A Prospective Clinical Study.

In Frontiers in immunology ; h5-index 100.0

Background : Prediction of the severity of COVID-19 at its onset is important for providing adequate and timely management to reduce mortality.

Objective : To study the prognostic value of damage parameters and cytokines as predictors of severity of COVID-19 using an extensive immunologic profiling and unbiased artificial intelligence methods.

Methods : Sixty hospitalized COVID-19 patients (30 moderate and 30 severe) and 17 healthy controls were included in the study. The damage indicators high mobility group box 1 (HMGB1), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), alanine aminotransferase (ALT), extensive biochemical analyses, a panel of 47 cytokines and chemokines were analyzed at weeks 1, 2 and 7 along with clinical complaints and CT scans of the lungs. Unbiased artificial intelligence (AI) methods (logistic regression and Support Vector Machine and Random Forest algorithms) were applied to investigate the contribution of each parameter to prediction of the severity of the disease.

Results : On admission, the severely ill patients had significantly higher levels of LDH, IL-6, monokine induced by gamma interferon (MIG), D-dimer, fibrinogen, glucose than the patients with moderate disease. The levels of macrophage derived cytokine (MDC) were lower in severely ill patients. Based on artificial intelligence analysis, eight parameters (creatinine, glucose, monocyte number, fibrinogen, MDC, MIG, C-reactive protein (CRP) and IL-6 have been identified that could predict with an accuracy of 83-87% whether the patient will develop severe disease.

Conclusion : This study identifies the prognostic factors and provides a methodology for making prediction for COVID-19 patients based on widely accepted biomarkers that can be measured in most conventional clinical laboratories worldwide.

Krysko Olga, Kondakova Elena, Vershinina Olga, Galova Elena, Blagonravova Anna, Gorshkova Ekaterina, Bachert Claus, Ivanchenko Mikhail, Krysko Dmitri V, Vedunova Maria

2021

COVID-19, IL-6, artificial intelligence, macrophage derived cytokine, prediction models

Ophthalmology Ophthalmology

[Attitude of patients to possible telemedicine in ophthalmology : Survey by questionnaire in patients with glaucoma].

In Der Ophthalmologe : Zeitschrift der Deutschen Ophthalmologischen Gesellschaft

BACKGROUND : The COVID-19 pandemic in 2020 and 2021 severely restricted the care of ophthalmology patients. Teleophthalmological services, such as video consultation or medical telephone advice could, at least partially, compensate for the lack of necessary controls in the case of chronic diseases; however, teleophthalmological options are currently still significantly underrepresented in Germany.

OBJECTIVE : In order to determine the willingness of patients to use telemedicine and the virtual clinic, we conducted a survey using a questionnaire on the subject of teleophthalmology in university medicine patients with known glaucoma as a chronic disease during the first wave of the COVID-19 pandemic.

METHODS : A total of100 patients were interviewed. The questionnaire contained 22 questions with multiple choice possible answers. The inclusion criterion was the presence of glaucoma as a chronic disease, age over 18 years, and sufficient linguistic understanding to answer the questions. The data were collected, analyzed and anonymously evaluated.

RESULTS : In the patient survey it could be shown that the respondents with glaucoma are very willing to do teleophthalmology and that this would be utilized. Of the patients surveyed 74.0% would accept telemedicine and virtual clinics. Of the ophthalmological patients surveyed 54.0% stated that their visit to the doctor/clinic could not take place due to SARS-CoV‑2 and 17.0% of the patients stated that the SARS-CoV‑2 pandemic had changed their opinion of telemedicine.

DISCUSSION : The acceptance of telemedicine in patients with chronic open-angle glaucoma seems surprisingly high. This has been increased even further by the SARS-CoV‑2 pandemic. These results reflect a general willingness of patients with chronic eye disease but do not reflect the applicability and acceptance and applicability from a medical point of view; however, this form of virtual consultation is accepted by the majority of patients with glaucoma and could be considered for certain clinical pictures.

Zwingelberg Sarah B, Mercieca Karl, Elksne Eva, Scheffler Stephanie, Prokosch Verena

2021-Sep-20

Artificial intelligence, Glaucoma, Ophthalmology, SARS-CoV‑2, Telemedicine

General General

Assessment of Artificial Intelligence Automatic Multiple Sclerosis Lesion Delineation Tool for Clinical Use.

In Clinical neuroradiology

PURPOSE : To implement and validate an existing algorithm for automatic delineation of white matter lesions on magnetic resonance imaging (MRI) in patients with multiple sclerosis (MS) on a local single-center dataset.

METHODS : We implemented a white matter hyperintensity segmentation model, based on a 2D convolutional neural network, using the conventional T2-weighted fluid attenuated inversion recovery (FLAIR) MRI sequence as input. The model was adapted for delineation of MS lesions by further training on a local dataset of 93 MS patients with a total of 3040 lesions. A quantitative evaluation was performed on ten test patients, in which model-generated masks were compared to manually delineated masks from two expert delineators. A subsequent qualitative evaluation of the implemented model was performed by two expert delineators, in which generated delineation masks on a clinical dataset of 53 patients were rated acceptable (< 10% errors) or unacceptable (> 10% errors) based on the total number of true lesions.

RESULTS : The quantitative evaluation resulted in an average accuracy score (F1) of 0.71, recall of 0.77 and dice similarity coefficient of 0.62. Our implemented model obtained the highest scores in all three metrics, when compared to three out of the box lesion segmentation models. In the clinical evaluation an average of 94% of our 53 model-generated masks were rated acceptable.

CONCLUSION : After adaptation to our local dataset, the implemented segmentation model was able to delineate MS lesions with a high clinical value as rated by delineation experts while outperforming popular out of the box applications. This serves as a promising step towards implementation of automatic lesion delineation in our MS clinic.

Hindsholm Amalie Monberg, Cramer Stig Præstekjær, Simonsen Helle Juhl, Frederiksen Jette Lautrup, Andersen Flemming, Højgaard Liselotte, Ladefoged Claes Nøhr, Lindberg Ulrich

2021-Sep-20

Clinical implementation, Convolutional neural network, Magnetic resonance imaging, White matter hyperintensity

General General

ProFitFun: A Protein Tertiary Structure Fitness Function for Quantifying the Accuracies of Model Structures.

In Bioinformatics (Oxford, England)

MOTIVATION : An accurate estimation of the quality of protein model structures typifies as a cornerstone in protein structure prediction regimes. Despite the recent groundbreaking success in the field of protein structure prediction, there are certain prospects for the improvement in model quality estimation at multiple stages of protein structure prediction and thus, to further push the prediction accuracy. Here, a novel approach, named ProFitFun, for assessing the quality of protein models is proposed by harnessing the sequence and structural features of experimental protein structures in terms of the preferences of backbone dihedral angles and relative surface accessibility of their amino acid residues at the tripeptide level. The proposed approach leverages upon the backbone dihedral angle and surface accessibility preferences of the residues by accounting for its N-terminal and C-terminal neighbors in the protein structure. These preferences are employed to evaluate protein structures through a machine learning approach and tested on an extensive dataset of diverse proteins.

RESULTS : The approach was extensively validated on a large test dataset (n = 25,005) of protein structures, comprising 23,661 models of 82 non-homologous proteins and 1,344 non-homologous experimental structures. Additionally, an external dataset of 40,000 models of 200 non-homologous proteins was also used for the validation of the proposed method. Both datasets were further employed for benchmarking the proposed method with four different state-of-the-art methods for protein structure quality assessment. In the benchmarking, the proposed method outperformed some state of the art methods in terms of Spearman's and Pearson's correlation coefficients, average GDT-TS loss, sum of z-scores, and average absolute difference of predictions over corresponding observed values. The high accuracy of the proposed approach promises a potential use of the sequence and structural features in computational protein design.

AVAILABILITY : http://github.com/KYZ-LSB/ProTerS-FitFun.

SUPPLEMENTARY INFORMATION : Supplementary data are available at Bioinformatics online.

Kaushik Rahul, Zhang Kam Y J

2021-Sep-20

General General

Predictors of functional outcomes in patients with facioscapulohumeral muscular dystrophy.

In Brain : a journal of neurology

Facioscapulohumeral muscular dystrophy (FSHD) is one of the most prevalent muscular dystrophies characterized by considerable variability in severity, rates of progression and functional outcomes. Few studies follow FSHD cohorts long enough to understand predictors of disease progression and functional outcomes, creating gaps in our understanding which impacts clinical care and the design of clinical trials. Efforts to identify molecularly targeted therapies create a need to better understand disease characteristics with predictive value to help refine clinical trial strategies and understand trial outcomes. Here we analyzed a prospective cohort from a large, longitudinally-followed registry of patients with FSHD in the United States to determine predictors of outcomes such as need for wheelchair use. This study analyzed de-identified data from 578 individuals with confirmed FSHD type 1 enrolled in the United States National Registry for FSHD Patients and Family members. Data were collected from January 2002 to September 2019 and included an average of nine years (range 0 to 18) of follow up surveys. Data were analyzed using descriptive epidemiological techniques, and risk of wheelchair use was determined using cox proportional hazards models. Supervised machine learning analysis was completed using Random Forest modeling and included all 189 unique features collected from registry questionnaires. A separate medications-only model was created that included 359 unique medications reported by participants. Here we show that smaller allele sizes were predictive of earlier age at onset, diagnosis and likelihood of wheelchair use. Additionally, we show that women were more likely overall to progress to wheelchair use and at a faster rate as compared to men, independent of genetics. Use of machine learning models that included all reported clinical features showed that the effect of allele size on progression to wheelchair use is small compared to disease duration, which may be important to consider in trial design. Medical comorbidities and medication use add to the risk for need for wheelchair dependence, raising the possibility for better medical management impacting outcomes in FSHD. The findings in this study will require further validation in additional, larger datasets but could have implications for clinical care, and inclusion criteria for future clinical trials in FSHD.

Katz Natalie K, Hogan John, Delbango Ryan, Cernik Colin, Tawil Rabi, Statland Jeffrey M

2021-Sep-20

artificial intelligence, facioscapulohumeral muscular dystrophy, functional outcomes, machine learning, wheelchair use

General General

Students' Personality Contributes More to Academic Performance than Well-Being and Learning Approach-Implications for Sustainable Development and Education.

In European journal of investigation in health, psychology and education

The present study aimed to describe the predictive role of personality dimensions, learning approaches, and well-being in the academic performance of students. In total, 602 students participated in this cross-sectional study and completed a set of questionnaires assessing personality, learning approach, and well-being. Two indexes were calculated to assess affective and non-affective well-being. The results partially support the hypotheses formulated. Results revealed that personality temperament and character dimensions, deep learning approach, and affective well-being were significant predictors of academic performance. A deep approach to learning was a full and partial mediator of the relationship between personality and academic performance. The results improve the understanding of the differential contribution of personality, type of learning approach, and type of well-being to academic performance. Comprehending that personality is the strongest predictor of academic performance, after controlling the type of learning approach and the type of well-being, informs school policies and decision-makers that it is essential to encourage personality development in adolescents to improve academic performance. These results also have implications for educational policies and practices at various levels, including an emphasis on the role of well-being as an educational asset. Understanding the links between personality, well-being, and education is essential to conceptualize education as a vital societal resource for facing current and future challenges, such as sustainable development.

Moreira Paulo, Pedras Susana, Pombo Paula

2020-Dec-06

academic performance, affective well-being, learning approach, non-affective well-being, personality

General General

Automated Size Recognition in Pediatric Emergencies Using Machine Learning and Augmented Reality: Within-Group Comparative Study.

In JMIR formative research

BACKGROUND : Pediatric emergencies involving children are rare events, and the experience of emergency physicians and the results of such emergencies are accordingly poor. Anatomical peculiarities and individual adjustments make treatment during pediatric emergency susceptible to error. Critical mistakes especially occur in the calculation of weight-based drug doses. Accordingly, the need for a ubiquitous assistance service that can, for example, automate dose calculation is high. However, few approaches exist due to the complexity of the problem.

OBJECTIVE : Technically, an assistance service is possible, among other approaches, with an app that uses a depth camera that is integrated in smartphones or head-mounted displays to provide a 3D understanding of the environment. The goal of this study was to automate this technology as much as possible to develop and statistically evaluate an assistance service that does not have significantly worse measurement performance than an emergency ruler (the state of the art).

METHODS : An assistance service was developed that uses machine learning to recognize patients and then automatically determines their size. Based on the size, the weight is automatically derived, and the dosages are calculated and presented to the physician. To evaluate the app, a small within-group design study was conducted with 17 children, who were each measured with the app installed on a smartphone with a built-in depth camera and a state-of-the-art emergency ruler.

RESULTS : According to the statistical results (one-sample t test; P=.42; α=.05), there is no significant difference between the measurement performance of the app and an emergency ruler under the test conditions (indoor, daylight). The newly developed measurement method is thus not technically inferior to the established one in terms of accuracy.

CONCLUSIONS : An assistance service with an integrated augmented reality emergency ruler is technically possible, although some groundwork is still needed. The results of this study clear the way for further research, for example, usability testing.

Schmucker Michael, Haag Martin

2021-Sep-20

augmented reality, emergency medicine, machine learning, mobile applications, mobile phone, resuscitation, user-computer interface

General General

Using Different Machine Learning Models to Classify Patients with Mild and Severe Cases of COVID-19 Based on Multivariate Blood Testing.

In Journal of medical virology

BACKGROUND : COVID19 is a serious respiratory disease. The ever-increasing number of cases is causing heavier loads on the health service system.

METHOD : Using 38 blood test indicators on the first day of admission for the 422 patients diagnosed with COVID-19 (from January 2020 to June 2021) to construct different machine learning models to classify patients with either mild or severe cases of COVID-19.

RESULTS : All models show good performance in the classification between COVID-19 patients with mild and severe disease. The AUC of the random forest model is 0.89, the AUC of the naive Bayes model is 0.90, the AUC of the support vector machine model is 0.86, and the AUC of the KNN model is 0.78, the AUC of the Logistic regression model is 0.84, and the AUC of the artificial neural network model is 0.87, among which the Naive Bayes model has the best performance.

CONCLUSION : Different machine learning models can classify patients with mild and severe cases based on 38 blood test indicators taken on the first day of admission for patients diagnosed with COVID-19. This article is protected by copyright. All rights reserved.

Zhang Rui-Kun, Xiao Qi, Zhu Sheng-Lang, Lin Hai-Yan, Tang Ming

2021-Sep-20

Artificial intelligence < Biostatistics & Bioinformatics, Coronavirus < Virus classification, Infection

Public Health Public Health

Highlighting psychological pain avoidance and decision-making bias as key predictors of suicide attempt in major depressive disorder-A novel investigative approach using machine learning.

In Journal of clinical psychology

OBJECTIVE : Predicting suicide is notoriously difficult and complex, but a serious public health issue. An innovative approach utilizing machine learning (ML) that incorporates features of psychological mechanisms and decision-making characteristics related to suicidality could create an improved model for identifying suicide risk in patients with major depressive disorder (MDD).

METHOD : Forty-four patients with MDD and past suicide attempts (MDD_SA, N = 44); 48 patients with MDD but without past suicide attempts (MDD_NS, N = 48-42 of whom with suicide ideation [MDD_SI, N = 42]), and healthy controls (HCs, N = 51) completed seven psychometric assessments including the Three-dimensional Psychological Pain Scale (TDPPS), and one behavioral assessment, the Balloon Analogue Risk Task (BART). Descriptive statistics, group comparisons, logistic regressions, and ML were used to explore and compare the groups and generate predictors of suicidal acts.

RESULTS : MDD_SA and MDD_NS differed in TDPPS total score, pain arousal and avoidance subscale scores, suicidal ideation scores, and relevant decision-making indicators in BART. Logistic regression tests linked suicide attempts to psychological pain avoidance and a risk decision-making indicator. The resultant key ML model distinguished MDD_SA/MDD_NS with 88.2% accuracy. The model could also distinguish MDD_SA/MDD_SI with 81.25% accuracy. The ML model using hopelessness could classify MDD_SI/HC with 94.4% accuracy.

CONCLUSION : ML analyses showed that motivation to avoid intolerable psychological pain, coupled with impaired decision-making bias toward under-valuing life's worth are highly predictive of suicide attempts. Analyses also demonstrated that suicidal ideation and attempts differed in potential mechanisms, as suicidal ideation was more related to hopelessness. ML algorithms show useful promises as a predictive instrument.

Ji Xinlei, Zhao Jiahui, Fan Lejia, Li Huanhuan, Lin Pan, Zhang Panwen, Fang Shulin, Law Samuel, Yao Shuqiao, Wang Xiang

2021-Sep-20

machine learning, major depressive disorder, psychological pain, risk decision-making, suicide

General General

Mechanism of amorphous phase stabilization in ultrathin films of monoatomic phase change material.

In Nanoscale ; h5-index 139.0

Elemental antimony has been recently proposed as a promising material for phase change memories with improved performances with respect to the most used ternary chalcogenide alloys. The compositional simplification prevents reliability problems due to demixing of the alloy during memory operation. This is made possible by the dramatic stabilization of the amorphous phase once Sb is confined in an ultrathin film 3-5 nm thick. In this work, we shed light on the microscopic origin of this effect by means of large scale molecular dynamics simulations based on an interatomic potential generated with a machine learning technique. The simulations suggest that the dramatic reduction of the crystal growth velocity in the film with respect to the bulk is due to the effect of nanoconfinement on the fast β relaxation dynamics while the slow α relaxation is essentially unaffected.

Dragoni Daniele, Behler Jörg, Bernasconi Marco

2021-Sep-20

General General

Combining Multimodal Behavioral Data of Gait, Speech, and Drawing for Classification of Alzheimer's Disease and Mild Cognitive Impairment.

In Journal of Alzheimer's disease : JAD

BACKGROUND : Gait, speech, and drawing behaviors have been shown to be sensitive to the diagnosis of Alzheimer's disease (AD) and mild cognitive impairment (MCI). However, previous studies focused on only analyzing individual behavioral modalities, although these studies suggested that each of these modalities may capture different profiles of cognitive impairments associated with AD.

OBJECTIVE : We aimed to investigate if combining behavioral data of gait, speech, and drawing can improve classification performance compared with the use of individual modality and if each of these behavioral data can be associated with different cognitive and clinical measures for the diagnosis of AD and MCI.

METHODS : Behavioral data of gait, speech, and drawing were acquired from 118 AD, MCI, and cognitively normal (CN) participants.

RESULTS : Combining all three behavioral modalities achieved 93.0%accuracy for classifying AD, MCI, and CN, and only 81.9%when using the best individual behavioral modality. Each of these behavioral modalities was statistically significantly associated with different cognitive and clinical measures for diagnosing AD and MCI.

CONCLUSION : Our findings indicate that these behaviors provide different and complementary information about cognitive impairments such that classification of AD and MCI is superior to using either in isolation.

Yamada Yasunori, Shinkawa Kaoru, Kobayashi Masatomo, Caggiano Vittorio, Nemoto Miyuki, Nemoto Kiyotaka, Arai Tetsuaki

2021-Sep-16

Drawing, gait analysis, handwriting, machine learning, speech, voice, walking

Surgery Surgery

Automatic Extraction of Hiatal Dimensions in 3-D Transperineal Pelvic Ultrasound Recordings.

In Ultrasound in medicine & biology ; h5-index 42.0

The aims of this work were to create a robust automatic software tool for measurement of the levator hiatal area on transperineal ultrasound (TPUS) volumes and to measure the potential reduction in variability and time taken for analysis in a clinical setting. The proposed tool automatically detects the C-plane (i.e., the plane of minimal hiatal dimensions) from a 3-D TPUS volume and subsequently uses the extracted plane to automatically segment the levator hiatus, using a convolutional neural network. The automatic pipeline was tested using 73 representative TPUS volumes. Reference hiatal outlines were obtained manually by two experts and compared with the pipeline's automated outlines. The Hausdorff distance, area, a clinical quality score, C-plane angle and C-plane Euclidean distance were used to evaluate C-plane detection and quantify levator hiatus segmentation accuracy. A visual Turing test was created to compare the performance of the software with that of the expert, based on the visual assessment of C-plane and hiatal segmentation quality. The overall time taken to extract the hiatal area with both measurement methods (i.e., manual and automatic) was measured. Each metric was calculated both for computer-observer differences and for inter-and intra-observer differences. The automatic method gave results similar to those of the expert when determining the hiatal outline from a TPUS volume. Indeed, the hiatal area measured by the algorithm and by an expert were within the intra-observer variability. Similarly, the method identified the C-plane with an accuracy of 5.76 ± 5.06° and 6.46 ± 5.18 mm in comparison to the inter-observer variability of 9.39 ± 6.21° and 8.48 ± 6.62 mm. The visual Turing test suggested that the automatic method identified the C-plane position within the TPUS volume visually as well as the expert. The average time taken to identify the C-plane and segment the hiatal area manually was 2 min and 35 ± 17 s, compared with 35 ± 4 s for the automatic result. This study presents a method for automatically measuring the levator hiatal area using artificial intelligence-based methodologies whereby the C-plane within a TPUS volume is detected and subsequently traced for the levator hiatal outline. The proposed solution was determined to be accurate, relatively quick, robust and reliable and, importantly, to reduce time and expertise required for pelvic floor disorder assessment.

Williams Helena, Cattani Laura, Van Schoubroeck Dominique, Yaqub Mohammad, Sudre Carole, Vercauteren Tom, D’Hooge Jan, Deprest Jan

2021-Sep-15

Automatic clinical workflow, Deep learning, Levator hiatus, Segmentation, Transperineal ultrasound, Ultrasound

Dermatology Dermatology

Identifying Silver Linings During the Pandemic Through Natural Language Processing.

In Frontiers in psychology ; h5-index 92.0

COVID-19 has presented an unprecedented challenge to human welfare. Indeed, we have witnessed people experiencing a rise of depression, acute stress disorder, and worsening levels of subclinical psychological distress. Finding ways to support individuals' mental health has been particularly difficult during this pandemic. An opportunity for intervention to protect individuals' health & well-being is to identify the existing sources of consolation and hope that have helped people persevere through the early days of the pandemic. In this paper, we identified positive aspects, or "silver linings," that people experienced during the COVID-19 crisis using computational natural language processing methods and qualitative thematic content analysis. These silver linings revealed sources of strength that included finding a sense of community, closeness, gratitude, and a belief that the pandemic may spur positive social change. People's abilities to engage in benefit-finding and leverage protective factors can be bolstered and reinforced by public health policy to improve society's resilience to the distress of this pandemic and potential future health crises.

Lossio-Ventura Juan Antonio, Lee Angela Yuson, Hancock Jeffrey T, Linos Natalia, Linos Eleni

2021

COVID-19, natural language processing, protective factors, sentiment analysis, silver linings, topic modeling

General General

Association Between the Change of Coagulation Parameters and Clinical Prognosis in Acute Ischemic Stroke Patients After Intravenous Thrombolysis With rt-PA.

In Clinical and applied thrombosis/hemostasis : official journal of the International Academy of Clinical and Applied Thrombosis/Hemostasis

Acute ischemic stroke patients with intravenous (IV) recombinant tissue plasminogen activator (rt-PA) thrombolysis have different outcomes. The degree of thrombolysis depends largely on the delicate balance of coagulation and fibrinolysis. Thus, our study aimed to investigate the prognostic value of routine coagulation parameters in acute stroke patients treated with rt-PA. From December 2016 to October 2018, consecutive patients treated with standard-dose IV rt-PA within 4.5 h of stroke onset were collected in Beijing Tiantan Hospital. Routine coagulation parameters, including platelet count, mean platelet volume, platelet distribution width, prothrombin time (PT), activated partial thromboplastin time, thrombin time, and fibrinogen, were measured at baseline (h0) and 24 h (h24) after thrombolysis. The change of coagulation parameters was defined as the (h24-h0)/h0 ratio. The prognosis included short-term outcome at 24 h and functional outcome at 3 months. A total of 267 patients were investigated (188 men and 79 women) with a mean age of 60.88  ± 12.31 years. In total, 9 patients had early neurological deterioration within 24 h, and 99 patients had an unfavorable outcome at the 3-month visit. In multivariate logistic regression, the (h24-h0)/h0 of PT was associated with unfavorable functional outcomes at 3 months (odds ratio: 1.42, 95% confidence interval: 1.02-2.28). While the change of other coagulation parameters failed to show any correlation with short-term or long-term prognosis. In conclusion, the prolongation of PT from baseline to 24 h after IV rt-PA increases the risk of 3-month unfavorable outcomes in acute stroke patients.

Wang Yu, Zhang Jia, Cao Zhentang, Zhang Qian, Zhao Xingquan

acute ischemic stroke, coagulation parameters, prognosis, rt-PA

General General

Diagnostic test accuracy of artificial intelligence analysis of cross-sectional imaging in pulmonary hypertension: a systematic literature review.

In The British journal of radiology

OBJECTIVES : To undertake the first systematic review examining the performance of artificial intelligence (AI) applied to cross-sectional imaging for the diagnosis of acquired pulmonary arterial hypertension (PAH).

METHODS : Searches of Medline, Embase and Web of Science were undertaken on July 1st 2020. Original publications studying AI applied to cross-sectional imaging for the diagnosis of acquired PAH in adults were identified through two-staged double-blinded review. Study quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies and Checklist for Artificial Intelligence in Medicine frameworks. Narrative synthesis was undertaken following Synthesis Without Meta-Analysis guidelines. This review received no funding and was registered in the International Prospective Register of Systematic Reviews (ID:CRD42020196295).

RESULTS : Searches returned 476 citations. Three retrospective observational studies, published between 2016 and 2020, were selected for data-extraction. Two methods applied to cardiac-MRI demonstrated high diagnostic accuracy, with the best model achieving AUC=0.90 (95% CI: 0.85-0.93), 89% sensitivity and 81% specificity. Stronger results were achieved using cardiac-MRI for classification of idiopathic PAH, achieving AUC=0.97 (95% CI: 0.89-1.0), 96% sensitivity and 87% specificity. One study reporting CT-based AI demonstrated lower accuracy, with 64.6% sensitivity and 97.0% specificity.

CONCLUSIONS : Automated methods for identifying PAH on cardiac-MRI are emerging with high diagnostic accuracy. AI applied to cross-sectional imaging may provide non-invasive support to reduce diagnostic delay in PAH. This would be helped by stronger solutions in other modalities.

ADVANCES IN KNOWLEDGE : There is a significant shortage of research in this important area. Early detection of PAH would be supported by further research advances on the promising emerging technologies identified.

Hardacre Conor Joseph, Robertshaw Joseph A, Barratt Shaney L, Adams Hannah L, MacKenzie Ross Robert V, Robinson Graham Re, Suntharalingam Jay, Pauling John D, Rodrigues Jonathan Carl Luis

2021-Sep-19

oncology Oncology

Outcome-based multi-objective optimization of lymphoma radiation therapy plans.

In The British journal of radiology

At its core, radiation therapy (RT) requires balancing therapeutic effects against risk of adverse events in cancer survivors. The radiation oncologist weighs numerous disease and patient-level factors when considering the expected risk-benefit ratio of combined treatment modalities. As part of this, RT plan optimization software is used to find a clinically acceptable RT plan delivering a prescribed dose to the target volume while respecting pre-defined radiation dose-volume constraints for selected organs at risk. The obvious limitation to the current approach is that it is virtually impossible to ensure the selected treatment plan could not be bettered by an alternative plan providing improved disease control and/or reduced risk of adverse events in this individual. Outcome-based optimization refers to a strategy where all planning objectives are defined by modeled estimates of a specific outcome's probability. Noting that various adverse events and disease control are generally incommensurable, leads to the concept of a Pareto-optimal plan: a plan where no single objective can be improved without degrading one or more of the remaining objectives. Further benefits of outcome-based multi objective optimization are that quantitative estimates of risks and benefit are obtained as are the effects of choosing a different trade-off between competing objectives. Furthermore, patient-level risk factors and combined treatment modalities are integrated directly into plan optimization. Here, we present this approach in the clinical setting of multi modality therapy for malignant lymphoma, a malignancy with marked heterogeneity in biology, target localization, and patient characteristics. We discuss future research priorities including the potential of artificial intelligence.

Modiri Arezoo, Vogelius Ivan, Rechner Laura Ann, Nygård Lotte, Bentzen Søren M, Specht Lena

2021-Sep-19

General General

Artificial Visual Perception Nervous System Based on Low-Dimensional Material Photoelectric Memristors.

In ACS nano ; h5-index 203.0

The visual perception system is the most important system for human learning since it receives over 80% of the learning information from the outside world. With the exponential growth of artificial intelligence technology, there is a pressing need for high-energy and area-efficiency visual perception systems capable of processing efficiently the received natural information. Currently, memristors with their elaborate dynamics, excellent scalability, and information (e.g., visual, pressure, sound, etc.) perception ability exhibit tremendous potential for the application of visual perception. Here, we propose a fully memristor-based artificial visual perception nervous system (AVPNS) which consists of a quantum-dot-based photoelectric memristor and a nanosheet-based threshold-switching (TS) memristor. We use a photoelectric and a TS memristor to implement the synapse and leaky integrate-and-fire (LIF) neuron functions, respectively. With the proposed AVPNS we successfully demonstrate the biological image perception, integration and fire, as well as the biosensitization process. Furthermore, the self-regulation process of a speed meeting control system in driverless automobiles can be accurately and conceptually emulated by this system. Our work shows that the functions of the biological visual nervous system may be systematically emulated by a memristor-based hardware system, thus expanding the spectrum of memristor applications in artificial intelligence.

Pei Yifei, Yan Lei, Wu Zuheng, Lu Jikai, Zhao Jianhui, Chen Jingsheng, Liu Qi, Yan Xiaobing

2021-Sep-20

driverless automobile, low-dimensional material, photovoltaic memristor, threshold-switching memristor, visual perception nervous system

General General

Medical student knowledge and critical appraisal of machine learning: a multicentre international cross-sectional study.

In Internal medicine journal

To utilise effectively tools that employ machine learning (ML) in clinical practice medical students and doctors will require a degree of understanding of ML models. To evaluate current levels of understanding, a formative examination and survey was conducted across three centres in Australia, New Zealand and the United States. Of the 245 individuals who participated in the study (response rate = 45.4%), the majority had difficulty with identifying weaknesses in model performance analysis. Further studies examining educational interventions addressing such ML topics are warranted.

Blacketer Charlotte, Parnis Roger, B Franke Kyle, Wagner Morganne, Wang David, Tan Yiran, Oakden-Rayner Luke, Gallagher Steve, Perry Seth W, Licinio Julio, Symonds Ian, Thomas Josephine, Duggan Paul, Bacchi Stephen

2021-Sep

artificial intelligence, curriculum, deep learning, formative examination, medical education, medical school

General General

The complete chloroplast genome and characteristics analysis of Musa basjoo Siebold.

In Molecular biology reports

BACKGROUND : An ornamental plant often seen in gardens and farmhouses, Musa basjoo Siebold can also be used as Chinese herbal medicine. Its pseudostem and leaves are diuretic; its root can be decocted together with ginger and licorice to cure gonorrhea and diabetes; the decoct soup of its pseudostem can help relieve heat, and the decoct soup of its dried flower can treat cerebral hemorrhage. There have not been many chloroplast genome studies on M. basjoo Siebold.

METHODS AND RESULTS : We characterized its complete chloroplast genome using Novaseq 6000 sequencing. This paper shows that the length of the chloroplast genome M. basjoo Siebold is 172,322 bp, with 36.45% GC content. M. basjoo Siebold includes a large single-copy region of 90,160 bp, a small single-copy region of 11,668 bp, and a pair of inverted repeats of 35,247 bp. Comparing the genomic structure and sequence data of closely related species, we have revealed the conserved gene order of the IR and LSC/SSC regions, which has provided a very inspiring discovery for future phylogenetic research.

CONCLUSIONS : Overall, this study has constructed an evolutionary tree of the genus Musa species with the complete chloroplast genome sequence for the first time. As can be seen, there is no obvious multi-branching in the genus, and M. basjoo Siebold and Musa itinerans are the closest relatives.

Liu Fenxiang, Movahedi Ali, Yang Wenguo, Xu Dezhi, Jiang Chuanbei, Xie Jigang, Zhang Yu

2021-Sep-19

Chloroplast genome, Comparative analysis, Musa basjoo Siebold, Phylogeny analysis

General General

Supporting Remote Survey Data Analysis by Co-researchers with Learning Disabilities through Inclusive and Creative Practices and Data Science Approaches.

In DIS. Designing Interactive Systems (Conference)

Through a process of robust co-design, we created a bespoke accessible survey platform to explore the role of co-researchers with learning disabilities (LDs) in research design and analysis. A team of co-researchers used this system to create an online survey to challenge public understanding of LDs [3]. Here, we describe and evaluate the process of remotely co-analyzing the survey data across 30 meetings in a research team consisting of academics and non-academics with diverse abilities amid new COVID-19 lockdown challenges. Based on survey data with >1,500 responses, we first co-analyzed demographics using graphs and art & design approaches. Next, co-researchers co-analyzed the output of machine learning-based structural topic modelling (STM) applied to open-ended text responses. We derived an efficient five-steps STM co-analysis process for creative, inclusive, and critical engagement of data by co-researchers. Co-researchers observed that by trying to understand and impact public opinion, their own perspectives also changed.

Chapko Dorota, Rothstein Pedro, Emeh Lizzie, Frumiento Pino, Kennedy Donald, Mcnicholas David, Orjiekwe Ifeoma, Overton Michaela, Snead Mark, Steward Robyn, Sutton Jenny, Bradshaw Melissa, Jeffreys Evie, Gallia Will, Ewans Sarah, Williams Mark, Grierson Mick

2021-Jun

Human-centered computing → Human computer interaction (HCI), Learning disability, co-design, survey, topic model

Internal Medicine Internal Medicine

KGDAL: Knowledge Graph Guided Double Attention LSTM for Rolling Mortality Prediction for AKI-D Patients.

In ACM-BCB ... ... : the ... ACM Conference on Bioinformatics, Computational Biology and Biomedicine. ACM Conference on Bioinformatics, Computational Biology and Biomedicine

With the rapid accumulation of electronic health record (EHR) data, deep learning (DL) models have exhibited promising performance on patient risk prediction. Recent advances have also demonstrated the effectiveness of knowledge graphs (KG) in providing valuable prior knowledge for further improving DL model performance. However, it is still unclear how KG can be utilized to encode high-order relations among clinical concepts and how DL models can make full use of the encoded concept relations to solve real-world healthcare problems and to interpret the outcomes. We propose a novel knowledge graph guided double attention LSTM model named KGDAL for rolling mortality prediction for critically ill patients with acute kidney injury requiring dialysis (AKI-D). KGDAL constructs a KG-based two-dimension attention in both time and feature spaces. In the experiment with two large healthcare datasets, we compared KGDAL with a variety of rolling mortality prediction models and conducted an ablation study to test the effectiveness, efficacy, and contribution of different attention mechanisms. The results showed that KGDAL clearly outperformed all the compared models. Also, KGDAL-derived patient risk trajectories may assist healthcare providers to make timely decisions and actions. The source code, sample data, and manual of KGDAL are available at https://github.com/lucasliu0928/KGDAL.

Liu Lucas Jing, Ortiz-Soriano Victor, Neyra Javier A, Chen Jin

2021-Aug

Attention Mechanism, Deep Learning, Knowledge Graph, Rolling Mortality Prediction

General General

Attention-Gated Graph Convolutions for Extracting Drug Interaction Information from Drug Labels.

In ACM transactions on computing for healthcare

Preventable adverse events as a result of medical errors present a growing concern in the healthcare system. As drug-drug interactions (DDIs) may lead to preventable adverse events, being able to extract DDIs from drug labels into a machine-processable form is an important step toward effective dissemination of drug safety information. Herein, we tackle the problem of jointly extracting mentions of drugs and their interactions, including interaction outcome, from drug labels. Our deep learning approach entails composing various intermediate representations, including graph-based context derived using graph convolutions (GCs) with a novel attention-based gating mechanism (holistically called GCA), which are combined in meaningful ways to predict on all subtasks jointly. Our model is trained and evaluated on the 2018 TAC DDI corpus. Our GCA model in conjunction with transfer learning performs at 39.20% F1 and 26.09% F1 on entity recognition (ER) and relation extraction (RE), respectively, on the first official test set and at 45.30% F1 and 27.87% F1 on ER and RE, respectively, on the second official test set. These updated results lead to improvements over our prior best by up to 6 absolute F1 points. After controlling for available training data, the proposed model exhibits state-of-the-art performance for this task.

Tran Tung, Kavuluru Ramakanth, Kilicoglu Halil

2021-Mar

Neural networks, drug-drug interactions, multi-task learning, relation extraction

General General

Predicting outcomes of psychotherapy for depression with electronic health record data.

In Journal of affective disorders reports

Predictive analytics with electronic health record (EHR) data holds promise for improving outcomes of psychiatric care. This study evaluated models for predicting outcomes of psychotherapy for depression in a clinical practice setting. EHR data from two large integrated health systems (Kaiser Permanente Colorado and Washington) included 5,554 new psychotherapy episodes with a baseline Patient Health Questionnaire (PHQ-9) score ≥ 10 and a follow-up PHQ-9 14-180 days after treatment initiation. Baseline predictors included demographics and diagnostic, medication, and encounter history. Prediction models for two outcomes-follow-up PHQ-9 score and treatment response (≥ 50% PHQ-9 reduction)-were trained in a random sample of 70% of episodes and validated in the remaining 30%. Two methods were used for modeling: generalized linear regression models with variable selection and random forests. Sensitivity analyses considered alternate predictor, outcome, and model specifications. Predictions of follow-up PHQ-9 scores poorly estimated observed outcomes (mean squared error = 31 for linear regression, 40 for random forest). Predictions of treatment response had low discrimination (AUC = 0.57 for logistic regression, 0.61 for random forest), low classification accuracy, and poor calibration. Sensitivity analyses showed similar results. We note that prediction model performance may vary for settings with different care or EHR documentation practices. In conclusion, prediction models did not accurately predict depression treatment outcomes despite using rich EHR data and advanced analytic techniques. Health systems should proceed cautiously when considering prediction models for psychiatric outcomes using baseline intake information. Transparent research should be conducted to evaluate performance of any model intended for clinical use.

Coley R Yates, Boggs Jennifer M, Beck Arne, Simon Gregory E

2021-Dec

Depression, Machine learning, Measurement-based care, Patient-reported outcomes, Prediction, Quality measures

Public Health Public Health

Microplanning for designing vaccination campaigns in low-resource settings: A geospatial artificial intelligence-based framework.

In Vaccine ; h5-index 70.0

Existing campaign-based healthcare delivery programs used for immunization often fall short of established health coverage targets due to a lack of accurate estimates for population size and location. A microplan, an integrated set of detailed planning components, can be used to identify this information to support programs such as equitable vaccination efforts. Here, we presents a series of steps necessary to create an artificial intelligence-based framework for automated microplanning, and our pilot implementation of this analysis tool across 29 countries of the Americas. Further, we describe our processes for generating a conceptual framework, creating customized catchment areas, and estimating up-to-date populations to support microplanning for health campaigns. Through our application of the present framework, we found that 68 million individuals across the 29 countries are within 5 km of a health facility. The number of health facilities analyzed ranged from 2 in Peru to 789 in Argentina, while the total population within 5 km ranged from 1,233 in Peru to 15,304,439 in Mexico. Our results demonstrate the feasibility of using this methodological framework to support the development of customized microplans for health campaigns using open-source data in multiple countries. The pandemic is demanding an improved capacity to generate successful, efficient immunization campaigns; we believe that the steps described here can increase the automation of microplans in low resource settings.

Augusto Hernandes Rocha Thiago, Grapiuna de Almeida Dante, Shankar Kozhumam Arthi, Cristina da Silva Núbia, Bárbara Abreu Fonseca Thomaz Erika, Christine de Sousa Queiroz Rejane, de Andrade Luciano, Staton Catherine, Ricardo Nickenig Vissoci João

2021-Sep-15

COVID-19, Coronavirus, Health campaign, Microplan, Vaccination, Vaccine

General General

Automatic deep learning system for COVID-19 infection quantification in chest CT.

In Multimedia tools and applications

The paper proposes an automatic deep learning system for COVID-19 infection areas segmentation in chest CT scans. CT imaging proved its ability to detect the COVID-19 disease even for asymptotic patients, which make it a trustworthy alternative for PCR. Coronavirus disease spread globally and PCR screening is the adopted diagnostic testing method for COVID-19 detection. However, PCR is criticized due its low sensitivity ratios, also, it is time-consuming and manual complicated process. The proposed framework includes different steps; it starts to prepare the region of interest by segmenting the lung organ, which then undergoes edge enhancing diffusion filtering (EED) to improve the infection areas contrast and intensity homogeneity. The proposed FCN is implemented using U-net architecture with modified residual block to include concatenation skip connection. The block improves the learning of gradient values by forwarding the infection area features through the network. The proposed system is evaluated using different measures and achieved dice overlapping score of 0.961 and 0.780 for lung and infection areas segmentation, respectively. The proposed system is trained and tested using many 2D CT slices extracted from diverse datasets from different sources, which demonstrate the system generalization and effectiveness. The use of more datasets from different sources helps to enhance the system accuracy and generalization, which can be accomplished based on the data availability in in the future.

Alirr Omar Ibrahim

2021-Sep-13

COVID-19 infection, Chest CT, Deep learning, Segmentation

Surgery Surgery

Salivary Metabolites are Promising Non-Invasive Biomarkers of Hepatocellular Carcinoma and Chronic Liver Disease.

In Liver cancer international

Background : Hepatocellular carcinoma (HCC) is a leading causes of cancer mortality worldwide. Improved tools are needed for detecting HCC so that treatment can begin as early as possible. Current diagnostic approaches and existing biomarkers, such as alpha-fetoprotein (AFP) lack sensitivity, resulting in too many false negative diagnoses. Machine-learning may be able to identify combinations of biomarkers that provide more robust predictions and improve sensitivity for detecting HCC. We sought to evaluate whether metabolites in patient saliva could distinguish those with HCC, cirrhosis, and those with no documented liver disease.

Methods and Results : We tested 125 salivary metabolites from 110 individuals (43 healthy, 37 HCC, 30 cirrhosis) and identified 4 metabolites that displayed significantly different abundance between groups (FDR P <.2). We also developed four tree-based, machine-learning models, optimized to include different numbers of metabolites, that were trained using cross-validation on 99 patients and validated on a withheld test set of 11 patients. A model using 12 metabolites -octadecanol, acetophenone, lauric acid, 1-monopalmitin, dodecanol, salicylaldehyde, glycyl-proline, 1-monostearin, creatinine, glutamine, serine and 4-hydroxybutyric acid- had a cross-validated sensitivity of 84.8%, specificity of 92.4% and correctly classified 90% of the HCC patients in the test cohort. This model outperformed previously reported sensitivities and specificities for AFP (20-100ng/ml) (61%, 86%) and AFP plus ultrasound (62%, 88%).

Conclusions and Impact : Metabolites detectable in saliva may represent products of disease pathology or a breakdown in liver function. Notably, combinations of salivary metabolites derived from machine-learning may serve as promising non-invasive biomarkers for the detection of HCC.

Hershberger Courtney E, Rodarte Alejandro I, Siddiqi Shirin, Moro Amika, Acevedo-Moreno Lou-Anne, Brown J Mark, Allende Daniela S, Aucejo Federico, Rotroff Daniel M

2021-Aug

Metabolomics, cirrhosis, liver cancer, machine learnings, risk factor

Radiology Radiology

Advanced brain ageing in Parkinson's disease is related to disease duration and individual impairment.

In Brain communications

Machine learning can reliably predict individual age from MRI data, revealing that patients with neurodegenerative disorders show an elevated biological age. A surprising gap in the literature, however, pertains to Parkinson's disease. Here, we evaluate brain age in two cohorts of Parkinson's patients and investigated the relationship between individual brain age and clinical characteristics. We assessed 372 patients with idiopathic Parkinson's disease, newly diagnosed cases from the Parkinson's Progression Marker Initiative database and a more chronic local sample, as well as age- and sex-matched healthy controls. Following morphometric preprocessing and atlas-based compression, individual brain age was predicted using a multivariate machine learning model trained on an independent, multi-site reference sample. Across cohorts, healthy controls were well predicted with a mean error of 4.4 years. In turn, Parkinson's patients showed a significant (controlling for age, gender and site) increase in brain age of ∼3 years. While this effect was already present in the newly diagnosed sample, advanced biological age was significantly related to disease duration as well as worse cognitive and motor impairment. While biological age is increased in patients with Parkinson's disease, the effect is at the lower end of what is found for other neurological and psychiatric disorders. We argue that this may reflect a heterochronicity between forebrain atrophy and small but behaviourally salient midbrain pathology. Finally, we point to the need to disentangle physiological ageing trajectories, lifestyle effects and core pathological changes.

Eickhoff Claudia R, Hoffstaedter Felix, Caspers Julian, Reetz Kathrin, Mathys Christian, Dogan Imis, Amunts Katrin, Schnitzler Alfons, Eickhoff Simon B

2021

Parkinson’s, age, atrophy, machine learning, prediction

Internal Medicine Internal Medicine

Reliable and Interpretable Mortality Prediction With Strong Foresight in COVID-19 Patients: An International Study From China and Germany.

In Frontiers in artificial intelligence

Cohort-independent robust mortality prediction model in patients with COVID-19 infection is not yet established. To build up a reliable, interpretable mortality prediction model with strong foresight, we have performed an international, bi-institutional study from China (Wuhan cohort, collected from January to March) and Germany (Würzburg cohort, collected from March to September). A Random Forest-based machine learning approach was applied to 1,352 patients from the Wuhan cohort, generating a mortality prediction model based on their clinical features. The results showed that five clinical features at admission, including lymphocyte (%), neutrophil count, C-reactive protein, lactate dehydrogenase, and α-hydroxybutyrate dehydrogenase, could be used for mortality prediction of COVID-19 patients with more than 91% accuracy and 99% AUC. Additionally, the time-series analysis revealed that the predictive model based on these clinical features is very robust over time when patients are in the hospital, indicating the strong association of these five clinical features with the progression of treatment as well. Moreover, for different preexisting diseases, this model also demonstrated high predictive power. Finally, the mortality prediction model has been applied to the independent Würzburg cohort, resulting in high prediction accuracy (with above 90% accuracy and 85% AUC) as well, indicating the robustness of the model in different cohorts. In summary, this study has established the mortality prediction model that allowed early classification of COVID-19 patients, not only at admission but also along the treatment timeline, not only cohort-independent but also highly interpretable. This model represents a valuable tool for triaging and optimizing the resources in COVID-19 patients.

Bai Tao, Zhu Xue, Zhou Xiang, Grathwohl Denise, Yang Pengshuo, Zha Yuguo, Jin Yu, Chong Hui, Yu Qingyang, Isberner Nora, Wang Dongke, Zhang Lei, Kortüm K Martin, Song Jun, Rasche Leo, Einsele Hermann, Ning Kang, Hou Xiaohua

2021

COVID-19, Wuhan cohort, Würzburg cohort, foresight, interpretability, mortality prediction model, reliability

oncology Oncology

Modality Attention and Sampling Enables Deep Learning with Heterogeneous Marker Combinations in Fluorescence Microscopy.

In Nature machine intelligence

Fluorescence microscopy allows for a detailed inspection of cells, cellular networks, and anatomical landmarks by staining with a variety of carefully-selected markers visualized as color channels. Quantitative characterization of structures in acquired images often relies on automatic image analysis methods. Despite the success of deep learning methods in other vision applications, their potential for fluorescence image analysis remains underexploited. One reason lies in the considerable workload required to train accurate models, which are normally specific for a given combination of markers, and therefore applicable to a very restricted number of experimental settings. We herein propose Marker Sampling and Excite - a neural network approach with a modality sampling strategy and a novel attention module that together enable (i) flexible training with heterogeneous datasets with combinations of markers and (ii) successful utility of learned models on arbitrary subsets of markers prospectively. We show that our single neural network solution performs comparably to an upper bound scenario where an ensemble of many networks is naïvely trained for each possible marker combination separately. In addition, we demonstrate the feasibility of this framework in high-throughput biological analysis by revising a recent quantitative characterization of bone marrow vasculature in 3D confocal microscopy datasets and further confirm the validity of our approach on an additional, significantly different dataset of microvessels in fetal liver tissues. Not only can our work substantially ameliorate the use of deep learning in fluorescence microscopy analysis, but it can also be utilized in other fields with incomplete data acquisitions and missing modalities.

Gomariz Alvaro, Portenier Tiziano, Helbling Patrick M, Isringhausen Stephan, Suessbier Ute, Nombela-Arrieta César, Goksel Orcun

2021-Sep

General General

MirrorME: implementation of an IoT based smart mirror through facial recognition and personalized information recommendation algorithm.

In International journal of information technology : an official journal of Bharati Vidyapeeth's Institute of Computer Applications and Management

We are living in the era of the fourth industrial revolution, which also treated as 4IR or Industry 4.0. Generally, 4IR considered as the mixture of robotics, artificial intelligence (AI), quantum computing, the Internet of Things (IoT) and other frontier technologies. It is obvious that nowadays a plethora of smart devices is providing services to make the daily life of humans easier. However, in the morning most people around the globe use a traditional mirror while preparing themselves for daily tasks. The aim is to build a low-cost intelligent mirror system that can display a variety of details based on user recommendations. Therefore, in this article, Internet of Things (IoT) and AI-based smart mirror is introduced that will support the users to receive the necessary daily update of weather information, date, time, calendar, to-do list, updated news headlines, traffic updates, COVID-19 cases status and so on. Moreover, a face detection method also implemented with the smart mirror to construct the architecture more secure. Our proposed MirrorME application provides a success rate of nearly 87% in interacting with the features of face recognition and voice input. The mirror is capable of delivering multimedia facilities while maintaining high levels of security within the device.

Uddin Khandaker Mohammad Mohi, Dey Samrat Kumar, Parvez Gias Uddin, Mukta Ayesha Siddika, Acharjee Uzzal Kumar

2021-Sep-12

4IR, Artificial intelligence, Authentication, Face detection, IoT, Smart mirror

General General

Experimental Technologies in the Diagnosis and Treatment of COVID-19 in Patients with Comorbidities.

In Journal of healthcare informatics research

The COVID-19 pandemic has impacted the whole world and raised concerns about its effects on different human organ systems. Early detection of COVID-19 may significantly increase the rate of survival; thus, it is critical that the disease is detected early. Emerging technologies have been used to prevent, diagnose, and manage COVID-19 among the populace in the USA and globally. Numerous studies have revealed the growing implementation of novel engineered systems during the intervention at various points of the disease's pathogenesis, especially as it relates to comorbidities and complications related to cardiovascular and respiratory organ systems. In this review, we provide a succinct, but extensive, review of the pathogenesis of COVID-19, particularly as it relates to angiotensin-converting enzyme 2 (ACE2) as a viral entry point. This is followed by a comprehensive analysis of cardiovascular and respiratory comorbidities of COVID-19 and novel technologies that are used to diagnose and manage hospitalized patients. Continuous cardiorespiratory monitoring systems, novel machine learning algorithms for rapidly triaging patients, various imaging modalities, wearable immunosensors, hotspot tracking systems, and other emerging technologies are reviewed. COVID-19 effects on the immune system, associated inflammatory biomarkers, and innovative therapies are also assessed. Finally, with emphasis on the impact of wearable and non-wearable systems, this review highlights future technologies that could help diagnose, monitor, and mitigate disease progression. Technologies that account for an individual's health conditions, comorbidities, and even socioeconomic factors can drastically reduce the high mortality seen among many COVID-19 patients, primarily via disease prevention, early detection, and pertinent management.

Amin Md Shahnoor, Wozniak Marcin, Barbaric Lidija, Pickard Shanel, Yerrabelli Rahul S, Christensen Anton, Coiado Olivia C

2021-Sep-15

Cardiovascular, Immune system, Innovation, Non-wearables, Pulmonary, Wearables

General General

A practical guide to promote informatics-driven efficient biotopographic material development.

In Bioactive materials

Micro/nano topographic structures have shown great utility in many biomedical areas including cell therapies, tissue engineering, and implantable devices. Computer-assisted informatics methods hold great promise for the design of topographic structures with targeted properties for a specific medical application. To benefit from these methods, researchers and engineers require a highly reusable "one structural parameter - one set of cell responses" database. However, existing confounding factors in topographic cell culture devices seriously impede the acquisition of this kind of data. Through carefully dissecting the confounding factors and their possible reasons for emergence, we developed corresponding guideline requirements for topographic cell culture device development to remove or control the influence of such factors. Based on these requirements, we then suggested potential strategies to meet them. In this work, we also experimentally demonstrated a topographic cell culture device with controlled confounding factors based on these guideline requirements and corresponding strategies. A "guideline for the development of topographic cell culture devices" was summarized to instruct researchers to develop topographic cell culture devices with the confounding factors removed or well controlled. This guideline aims to promote the establishment of a highly reusable "one structural parameter - one set of cell responses" database that could facilitate the application of informatics methods, such as artificial intelligence, in the rational design of future biotopographic structures with high efficacy.

Guo Yuanlong, Mi Jiaomei, Ye Chen, Ao Yong, Shi Mengru, Shan Zhengjie, Li Bingzhi, Chen Zetao, Chen Zhuofan, Vasilev Krasimir, Xiao Yin

2022-Feb

Biotopographic materials, Cell culture device, Confounding factor, Database, Informatics

General General

Machine learning for identification of frailty in Canadian primary care practices.

In International journal of population data science

Introduction : Frailty is a medical syndrome, commonly affecting people aged 65 years and over and is characterized by a greater risk of adverse outcomes following illness or injury. Electronic medical records contain a large amount of longitudinal data that can be used for primary care research. Machine learning can fully utilize this wide breadth of data for the detection of diseases and syndromes. The creation of a frailty case definition using machine learning may facilitate early intervention, inform advanced screening tests, and allow for surveillance.

Objectives : The objective of this study was to develop a validated case definition of frailty for the primary care context, using machine learning.

Methods : Physicians participating in the Canadian Primary Care Sentinel Surveillance Network across Canada were asked to retrospectively identify the level of frailty present in a sample of their own patients (total n = 5,466), collected from 2015-2019. Frailty levels were dichotomized using a cut-off of 5. Extracted features included previously prescribed medications, billing codes, and other routinely collected primary care data. We used eight supervised machine learning algorithms, with performance assessed using a hold-out test set. A balanced training dataset was also created by oversampling. Sensitivity analyses considered two alternative dichotomization cut-offs. Model performance was evaluated using area under the receiver-operating characteristic curve, F1, accuracy, sensitivity, specificity, negative predictive value and positive predictive value.

Results : The prevalence of frailty within our sample was 18.4%. Of the eight models developed to identify frail patients, an XGBoost model achieved the highest sensitivity (78.14%) and specificity (74.41%). The balanced training dataset did not improve classification performance. Sensitivity analyses did not show improved performance for cut-offs other than 5.

Conclusion : Supervised machine learning was able to create well performing classification models for frailty. Future research is needed to assess frailty inter-rater reliability, and link multiple data sources for frailty identification.

Aponte-Hao Sylvia, Wong Sabrina T, Thandi Manpreet, Ronksley Paul, McBrien Kerry, Lee Joon, Grandy Mathew, Mangin Dee, Katz Alan, Singer Alexander, Manca Donna, Williamson Tyler

2021

Canada, case definition, electronic health records, electronic medical records, frailty, machine learning, primary care, supervised machine learning

General General

Derivative-free optimization adversarial attacks for graph convolutional networks.

In PeerJ. Computer science

In recent years, graph convolutional networks (GCNs) have emerged rapidly due to their excellent performance in graph data processing. However, recent researches show that GCNs are vulnerable to adversarial attacks. An attacker can maliciously modify edges or nodes of the graph to mislead the model's classification of the target nodes, or even cause a degradation of the model's overall classification performance. In this paper, we first propose a black-box adversarial attack framework based on derivative-free optimization (DFO) to generate graph adversarial examples without using gradient and apply advanced DFO algorithms conveniently. Second, we implement a direct attack algorithm (DFDA) using the Nevergrad library based on the framework. Additionally, we overcome the problem of large search space by redesigning the perturbation vector using constraint size. Finally, we conducted a series of experiments on different datasets and parameters. The results show that DFDA outperforms Nettack in most cases, and it can achieve an average attack success rate of more than 95% on the Cora dataset when perturbing at most eight edges. This demonstrates that our framework can fully exploit the potential of DFO methods in node classification adversarial attacks.

Yang Runze, Long Teng

2021

Adversarial attack, Derivative-free optimization, Graph convolutional network, Machine learning

General General

SSMFN: a fused spatial and sequential deep learning model for methylation site prediction.

In PeerJ. Computer science

Background : Conventional in vivo methods for post-translational modification site prediction such as spectrophotometry, Western blotting, and chromatin immune precipitation can be very expensive and time-consuming. Neural networks (NN) are one of the computational approaches that can predict effectively the post-translational modification site. We developed a neural network model, namely the Sequential and Spatial Methylation Fusion Network (SSMFN), to predict possible methylation sites on protein sequences.

Method : We designed our model to be able to extract spatial and sequential information from amino acid sequences. Convolutional neural networks (CNN) is applied to harness spatial information, while long short-term memory (LSTM) is applied for sequential data. The latent representation of the CNN and LSTM branch are then fused. Afterwards, we compared the performance of our proposed model to the state-of-the-art methylation site prediction models on the balanced and imbalanced dataset.

Results : Our model appeared to be better in almost all measurement when trained on the balanced training dataset. On the imbalanced training dataset, all of the models gave better performance since they are trained on more data. In several metrics, our model also surpasses the PRMePred model, which requires a laborious effort for feature extraction and selection.

Conclusion : Our models achieved the best performance across different environments in almost all measurements. Also, our result suggests that the NN model trained on a balanced training dataset and tested on an imbalanced dataset will offer high specificity and low sensitivity. Thus, the NN model for methylation site prediction should be trained on an imbalanced dataset. Since in the actual application, there are far more negative samples than positive samples.

Lumbanraja Favorisen Rosyking, Mahesworo Bharuno, Cenggoro Tjeng Wawan, Sudigyo Digdo, Pardamean Bens

2021

CNN, Deep Learning, LSTM, Methylation, Prediction, Sequential, Spatial

General General

Integration of statistical inferences and machine learning algorithms for prediction of metritis cure in dairy cows.

In Journal of dairy science

The study's objectives were to identify cow-level and environmental factors associated with metritis cure to predict metritis cure using traditional statistics and machine learning algorithms. The data set used was from a previous study comparing the efficacy of different therapies and self-cure for metritis. Metritis was defined as fetid, watery, reddish-brownish discharge, with or without fever. Cure was defined as an absence of metritis signs 12 d after diagnosis. Cows were randomly allocated to receive a subcutaneous injection of 6.6 mg/kg of ceftiofur crystalline-free acid (Excede, Zoetis) at the day of diagnosis and 3 d later (n = 275); and no treatment at the time of metritis diagnosis (n = 275). The variables days in milk (DIM) at metritis diagnosis, treatment, season of the metritis diagnosis, month of metritis diagnostic, number of lactation, parity, calving score, dystocia, retained fetal membranes, body condition score at d 5 postpartum, vulvovaginal laceration score, the rectal temperature at the metritis diagnosis, fever at diagnosis, milk production from the day before to metritis diagnosis, and milk production slope up to 5, 7, and 9 DIM were offered to univariate logistic regression. Variables included in the multivariable logistic regression model were selected from the univariate analysis according to P-value. Variables were offered to the model to assess the association between these factors and metritis cure. Additionally, the univariate logistic regression variables were offered to a recursive feature elimination to find the optimal subset of features for a machine learning algorithms analysis. Cows without vulvovaginal laceration had 1.91 higher odds of curing of metritis than cows with vulvovaginal laceration. Cows that developed metritis at >7 DIM had 2.09 higher odds of being cured than cows that developed metritis at ≤7 DIM. For rectal temperature, each degree Celsius above 39.4°C led to lower odds to be cured than cows with rectal temperature ≤39.4°C. Furthermore, milk production slope and milk production difference from the day before to the metritis diagnosis were essential variables to predict metritis cure. Cows that had reduced milk production from the day before to the metritis diagnosis had lower odds to be cured than cows with moderate milk production increase. The results from the multivariable logistic regression and receiver operating characteristic analysis indicated that cows developing metritis at >7 DIM, with increase in milk production, and with a rectal temperature ≤39.40°C had increased likelihood of cure of metritis with an accuracy of 75%. The machine learning analysis showed that in addition to these variables, calving-related disorders, season, and month of metritis event were needed to predict whether the cow will cure or not from metritis with an accuracy ≥70% and F1 score (harmonic mean between precision and recall) ≥0.78. Although machine learning algorithms are acknowledged as powerful tools for predictive classification, the current study was unable to replicate its potential benefits. More research is needed to optimize predictive models of metritis cure.

de Oliveira E B, Ferreira F C, Galvão K N, Youn J, Tagkopoulos I, Silva-Del-Rio N, Pereira R V V, Machado V S, Lima F S

2021-Sep-15

ceftiofur, dairy cow, machine learning, metritis cure

General General

COVID-19: Government subsidy models for sustainable energy supply with disruption risks.

In Renewable & sustainable energy reviews

The outbreak of the COVID-19 pandemic poses great challenges to the current government subsidy models in the renewable energy sector for recovering in the post-pandemic economy. Although, many subsidy models have been applied to accelerate renewable energy investment decisions. However, it is important to develop a new model to ensure the sustainability of the renewable energy supply network under disruptions on both the supply and demand sides due to hazardous events. This study investigates different subsidy models (renewable credit, supplier subsidy, and retailer subsidy) to find a win-win subsidy model for sustainable energy supply under disruption risks. The objective is to determine the optimal capacity of renewable energy added to the grid, the optimal wholesale price of the power plant, and the optimal retail price of the aggregator under different subsidy models to maximize the economic, social, and environmental benefits of the whole network. A novel scenario-based robust fuzzy optimization approach is proposed to capture the uncertainties of business-as-usual operations (e.g., some relevant costs and demand) and hazardous events (e.g., COVID-19 pandemic). The proposed model is tested in a case study of the Vietnamese energy market. The results show that for a high negative impact level of hazardous events on the supply side, the renewable credit and supplier subsidy models should be considered to recovery the renewable energy market. Further, the proposed approach has a better performance in improving the power plant's robust profit for most of the hazard scenarios than the robust optimization model.

Tsao Yu-Chung, Thanh Vo-Van, Chang Yi-Ying, Wei Hsi-Hsien

2021-Oct

COVID-19, Government subsidy, Hazardous scenario, Renewable energy, Robust fuzzy model, Sustainable supply

General General

Channel state information estimation for 5G wireless communication systems: recurrent neural networks approach.

In PeerJ. Computer science

In this study, a deep learning bidirectional long short-term memory (BiLSTM) recurrent neural network-based channel state information estimator is proposed for 5G orthogonal frequency-division multiplexing systems. The proposed estimator is a pilot-dependent estimator and follows the online learning approach in the training phase and the offline approach in the practical implementation phase. The estimator does not deal with complete a priori certainty for channels' statistics and attains superior performance in the presence of a limited number of pilots. A comparative study is conducted using three classification layers that use loss functions: mean absolute error, cross entropy function for kth mutually exclusive classes and sum of squared of the errors. The Adam, RMSProp, SGdm, and Adadelat optimisation algorithms are used to evaluate the performance of the proposed estimator using each classification layer. In terms of symbol error rate and accuracy metrics, the proposed estimator outperforms long short-term memory (LSTM) neural network-based channel state information, least squares and minimum mean square error estimators under different simulation conditions. The computational and training time complexities for deep learning BiLSTM- and LSTM-based estimators are provided. Given that the proposed estimator relies on the deep learning neural network approach, where it can analyse massive data, recognise statistical dependencies and characteristics, develop relationships between features and generalise the accrued knowledge for new datasets that it has not seen before, the approach is promising for any 5G and beyond communication system.

Essai Ali Mohamed Hassan, Taha Ibrahim B M

2021

BiLSTM, Channel state information estimator, Deep learning neural networks, Loss functions

General General

Solving musculoskeletal biomechanics with machine learning.

In PeerJ. Computer science

Deep learning is a relatively new computational technique for the description of the musculoskeletal dynamics. The experimental relationships of muscle geometry in different postures are the high-dimensional spatial transformations that can be approximated by relatively simple functions, which opens the opportunity for machine learning (ML) applications. In this study, we challenged general ML algorithms with the problem of approximating the posture-dependent moment arm and muscle length relationships of the human arm and hand muscles. We used two types of algorithms, light gradient boosting machine (LGB) and fully connected artificial neural network (ANN) solving the wrapping kinematics of 33 muscles spanning up to six degrees of freedom (DOF) each for the arm and hand model with 18 DOFs. The input-output training and testing datasets, where joint angles were the input and the muscle length and moment arms were the output, were generated by our previous phenomenological model based on the autogenerated polynomial structures. Both models achieved a similar level of errors: ANN model errors were 0.08 ± 0.05% for muscle lengths and 0.53 ± 0.29% for moment arms, and LGB model made similar errors-0.18 ± 0.06% and 0.13 ± 0.07%, respectively. LGB model reached the training goal with only 103 samples, while ANN required 106 samples; however, LGB models were about 39 times slower than ANN models in the evaluation. The sufficient performance of developed models demonstrates the future applicability of ML for musculoskeletal transformations in a variety of applications, such as in advanced powered prosthetics.

Smirnov Yaroslav, Smirnov Denys, Popov Anton, Yakovenko Sergiy

2021

Biomechanics, Deep neural networks, Hand, Machine learning, Muscle, Real-time

General General

Artificial intelligence maturity model: a systematic literature review.

In PeerJ. Computer science

Organizations in various industries have widely developed the artificial intelligence (AI) maturity model as a systematic approach. This study aims to review state-of-the-art studies related to AI maturity models systematically. It allows a deeper understanding of the methodological issues relevant to maturity models, especially in terms of the objectives, methods employed to develop and validate the models, and the scope and characteristics of maturity model development. Our analysis reveals that most works concentrate on developing maturity models with or without their empirical validation. It shows that the most significant proportion of models were designed for specific domains and purposes. Maturity model development typically uses a bottom-up design approach, and most of the models have a descriptive characteristic. Besides that, maturity grid and continuous representation with five levels are currently trending in maturity model development. Six out of 13 studies (46%) on AI maturity pertain to assess the technology aspect, even in specific domains. It confirms that organizations still require an improvement in their AI capability and in strengthening AI maturity. This review provides an essential contribution to the evolution of organizations using AI to explain the concepts, approaches, and elements of maturity models.

Sadiq Raghad Baker, Safie Nurhizam, Abd Rahman Abdul Hadi, Goudarzi Shidrokh

2021

Artificial Intelligence, Maturity model, Organization, Systematic literature review

General General

Detecting sarcasm in multi-domain datasets using convolutional neural networks and long short term memory network model.

In PeerJ. Computer science

Sarcasm emerges as a common phenomenon across social networking sites because people express their negative thoughts, hatred and opinions using positive vocabulary which makes it a challenging task to detect sarcasm. Although various studies have investigated the sarcasm detection on baseline datasets, this work is the first to detect sarcasm from a multi-domain dataset that is constructed by combining Twitter and News Headlines datasets. This study proposes a hybrid approach where the convolutional neural networks (CNN) are used for feature extraction while the long short-term memory (LSTM) is trained and tested on those features. For performance analysis, several machine learning algorithms such as random forest, support vector classifier, extra tree classifier and decision tree are used. The performance of both the proposed model and machine learning algorithms is analyzed using the term frequency-inverse document frequency, bag of words approach, and global vectors for word representations. Experimental results indicate that the proposed model surpasses the performance of the traditional machine learning algorithms with an accuracy of 91.60%. Several state-of-the-art approaches for sarcasm detection are compared with the proposed model and results suggest that the proposed model outperforms these approaches concerning the precision, recall and F1 scores. The proposed model is accurate, robust, and performs sarcasm detection on a multi-domain dataset.

Jamil Ramish, Ashraf Imran, Rustam Furqan, Saad Eysha, Mehmood Arif, Choi Gyu Sang

2021

Convolutional neural networks, Long short term memory network, Multi-domain sarcastic comments, Sarcasm detection, Social media

Public Health Public Health

Impact of COVID-19 on city-scale transportation and safety: An early experience from Detroit.

In Smart health (Amsterdam, Netherlands)

The COVID-19 pandemic brought unprecedented levels of disruption to the local and regional transportation networks throughout the United States, especially the Motor City---Detroit. That was mainly a result of swift restrictive measures such as statewide quarantine and lock-down orders to confine the spread of the virus and the rising number of COVID-19 confirmed cases and deaths. This work is driven by analyzing five types of real-world data sets from Detroit related to traffic volume, daily cases, weather, social distancing index, and crashes from January 2019 to June 2020. The primary goals of this work are: i) figuring out the impacts of COVID-19 on the transportation network usage (traffic volume) and safety (crashes) for the City of Detroit, ii) determining whether each type of data (e.g. traffic volume data) could be a useful factor in the confirmed-cases prediction, and iii) providing an early future prediction method for COVID-19 rates, which can be a vital contributor to life-saving advanced preventative and preparatory responses. In addressing these problems, the prediction results of six feature groups are presented and analyzed to quantify the prediction effectiveness of each type of data. Then, a deep learning model was developed using long short-term memory networks to predict the number of confirmed cases within the next week. The model demonstrated a promising prediction result with a coefficient of determination ( R 2 ) of up to approximately 0.91. Furthermore, six essential observations with supporting evidence are presented, which will be helpful for decision-makers to take specific measures that aid in preventing the spread of COVID-19 and protecting public health and safety. The proposed approaches could be applied, customized, adjusted, and replicated for analysis of the impact of COVID-19 on a transportation network and prediction of the anticipated COVID-19 cases using a similar data set obtained for other large cities in the USA or from around the world.

Yao Yongtao, Geara Tony G, Shi Weisong

2021-Nov

COVID-19, Daily cases Detroit, Data analysis, Prediction, Quarantine, Social distancing weather, Traffic volume crashes, Transportation networks

General General

DNS dataset for malicious domains detection.

In Data in brief

The Domain Name Service (DNS) is a central point in the functioning of the internet. Just as organizations use domain names to enable the access to their computational services, malicious actors make use of domain names to point to the services under their control. Distinguishing between non-malicious and malicious domain names is extremely important, as it allows to grant or block the access to external services, maximizing the security of the organization and users. Nowadays there are many DNS firewall solutions. Most of these are based on known malicious domain lists that are being constantly updated. However, in this way, it is only possible to block known malicious communications, leaving out many others that can be malicious but are not known. Adopting machine learning to classify domains contributes to the detection of domains that are not yet on the block list. The dataset described in this manuscript is meant for supervised machine learning-based analysis of malicious and non-malicious domain names. The dataset was created from scratch, using publicly DNS logs of both malicious and non-malicious domain names. Using the domain name as input, 34 features were obtained. Features like the domain name entropy, number of strange characters and domain name length were obtained directly from the domain name. Other features like, domain name creation date, Internet Protocol (IP), open ports, geolocation were obtained from data enrichment processes (e.g. Open Source Intelligence (OSINT)). The class was determined considering the data source (malicious DNS log files and non-malicious DNS log files). The dataset consists of data from approximately 90000 domain names and it is balanced between 50% non-malicious and 50% of malicious domain names.

Marques Cláudio, Malta Silvestre, Magalhães João Paulo

2021-Oct

Cybersecurity, DNS, Firewall, Machine learning

Ophthalmology Ophthalmology

Automated Grading of Diabetic Retinopathy with Ultra-Widefield Fluorescein Angiography and Deep Learning.

In Journal of diabetes research ; h5-index 44.0

Purpose : The objective of this study was to establish diagnostic technology to automatically grade the severity of diabetic retinopathy (DR) according to the ischemic index and leakage index with ultra-widefield fluorescein angiography (UWFA) and the Early Treatment Diabetic Retinopathy Study (ETDRS) 7-standard field (7-SF).

Methods : This is a cross-sectional study. UWFA samples from 280 diabetic patients and 119 normal patients were used to train and test an artificial intelligence model to differentiate PDR and NPDR based on the ischemic index and leakage index with UWFA. A panel of retinal specialists determined the ground truth for our data set before experimentation. A confusion matrix as a metric was used to measure the precision of our algorithm, and a simple linear regression function was implemented to explore the discrimination of indexes on the DR grades. In addition, the model was tested with simulated 7-SF.

Results : The model classification of DR in the original UWFA images achieved 88.50% accuracy and 73.68% accuracy in the simulated 7-SF images. A simple linear regression function demonstrated that there is a significant relationship between the ischemic index and leakage index and the severity of DR. These two thresholds were set to classify the grade of DR, which achieved 76.8% accuracy.

Conclusions : The optimization of the cycle generative adversarial network (CycleGAN) and convolutional neural network (CNN) model classifier achieved DR grading based on the ischemic index and leakage index with UWFA and simulated 7-SF and provided accurate inference results. The classification accuracy with UWFA is slightly higher than that of simulated 7-SF.

Wang Xiaoling, Ji Zexuan, Ma Xiao, Zhang Ziyue, Yi Zuohuizi, Zheng Hongmei, Fan Wen, Chen Changzheng

2021

Dermatology Dermatology

Application of Genetic Algorithm-Based Support Vector Machine in Identification of Gene Expression Signatures for Psoriasis Classification: A Hybrid Model.

In BioMed research international ; h5-index 102.0

Background : Psoriasis is a chronic autoimmune disease impairing significantly the quality of life of the patient. The diagnosis of the disease is done via a visual inspection of the lesional skin by dermatologists. Classification of psoriasis using gene expression is an important issue for the early and effective treatment of the disease. Therefore, gene expression data and selection of suitable gene signatures are effective sources of information.

Methods : We aimed to develop a hybrid classifier for the diagnosis of psoriasis based on two machine learning models of the genetic algorithm and support vector machine (SVM). The method also conducts gene signature selection. A publically available gene expression dataset was used to test the model.

Results : A number of 181 probe sets were selected among the original 54,675 probes using the hybrid model with a prediction accuracy of 100% over the test set. A number of 10 hub genes were identified using the protein-protein interaction network. Nine out of 10 identified genes were found in significant modules.

Conclusions : The results showed that the genetic algorithm improved the SVM classifier performance significantly implying the ability of the proposed model in terms of detecting relevant gene expression signatures as the best features.

Tapak Leili, Afshar Saeid, Afrasiabi Mahlagha, Ghasemi Mohammad Kazem, Alirezaei Pedram

2021

Radiology Radiology

DeepStrain: A Deep Learning Workflow for the Automated Characterization of Cardiac Mechanics.

In Frontiers in cardiovascular medicine

Myocardial strain analysis from cinematic magnetic resonance imaging (cine-MRI) data provides a more thorough characterization of cardiac mechanics than volumetric parameters such as left-ventricular ejection fraction, but sources of variation including segmentation and motion estimation have limited its wider clinical use. We designed and validated a fast, fully-automatic deep learning (DL) workflow to generate both volumetric parameters and strain measures from cine-MRI data consisting of segmentation and motion estimation convolutional neural networks. The final motion network design, loss function, and associated hyperparameters are the result of a thorough ad hoc implementation that we carefully planned specific for strain quantification, tested, and compared to other potential alternatives. The optimal configuration was trained using healthy and cardiovascular disease (CVD) subjects (n = 150). DL-based volumetric parameters were correlated (>0.98) and without significant bias relative to parameters derived from manual segmentations in 50 healthy and CVD test subjects. Compared to landmarks manually-tracked on tagging-MRI images from 15 healthy subjects, landmark deformation using DL-based motion estimates from paired cine-MRI data resulted in an end-point-error of 2.9 ± 1.5 mm. Measures of end-systolic global strain from these cine-MRI data showed no significant biases relative to a tagging-MRI reference method. On 10 healthy subjects, intraclass correlation coefficient for intra-scanner repeatability was good to excellent (>0.75) for all global measures and most polar map segments. In conclusion, we developed and evaluated the first end-to-end learning-based workflow for automated strain analysis from cine-MRI data to quantitatively characterize cardiac mechanics of healthy and CVD subjects.

Morales Manuel A, van den Boomen Maaike, Nguyen Christopher, Kalpathy-Cramer Jayashree, Rosen Bruce R, Stultz Collin M, Izquierdo-Garcia David, Catana Ciprian

2021

cine-MRI, deep learning, motion estimation, myocardial strain, segmentation

General General

Automatic Detection of Gaze and Body Orientation in Elementary School Classrooms.

In Frontiers in robotics and AI

Detecting the direction of the gaze and orientation of the body of both teacher and students is essential to estimate who is paying attention to whom. It also provides vital clues for understanding their unconscious, non-verbal behavior. These are called "honest signals" since they are unconscious subtle patterns in our interaction with other people that help reveal the focus of our attention. Inside the classroom, they provide important clues about teaching practices and students' responses to different conscious and unconscious teaching strategies. Scanning this non-verbal behavior in the classroom can provide important feedback to the teacher in order for them to improve their teaching practices. This type of analysis usually requires sophisticated eye-tracking equipment, motion sensors, or multiple cameras. However, for this to be a useful tool in the teacher's daily practice, an alternative must be found using only a smartphone. A smartphone is the only instrument that a teacher always has at their disposal and is nowadays considered truly ubiquitous. Our study looks at data from a group of first-grade classrooms. We show how video recordings on a teacher's smartphone can be used in order to estimate the direction of the teacher and students' gaze, as well as their body orientation. Using the output from the OpenPose software, we run Machine Learning (ML) algorithms to train an estimator to recognize the direction of the students' gaze and body orientation. We found that the level of accuracy achieved is comparable to that of human observers watching frames from the videos. The mean square errors (RMSE) of the predicted pitch and yaw angles for head and body directions are on average 11% lower than the RMSE between human annotators. However, our solution is much faster, avoids the tedium of doing it manually, and makes it possible to design solutions that give the teacher feedback as soon as they finish the class.

Araya Roberto, Sossa-Rivera Jorge

2021

body orientation detection, gaze detection, non-verbal behavior, student attention, teaching practices

General General

Adoption of New Technologies: Artificial Intelligence.

In Gastrointestinal endoscopy clinics of North America

Over the past decade, artificial intelligence (AI) has been broadly applied to many aspects of human life, with recent groundbreaking successes in facial recognition, natural language processing, autonomous driving, and medical imaging. Gastroenterology has applied AI to a vast array of clinical problems, and some of the earliest prospective trials examining AI in medicine have been in computer vision applied to endoscopy. Evidence is mounting for 2 broad areas of AI as applied to gastroenterology: computer-aided detection and computer-aided diagnosis.

Glissen Brown Jeremy R, Berzin Tyler M

2021-Oct

Computer-aided detection, Computer-aided diagnosis, Cost-effectiveness, Deep learning, Machine learning, Operations, Polyp detection, Regulations

General General

MFBCNNC: Momentum factor biogeography convolutional neural network for COVID-19 detection via chest X-ray images.

In Knowledge-based systems

Aim : By October 6, 2020, Coronavirus disease 2019 (COVID-19) was diagnosed worldwide, reaching 3,355,7427 people and 1,037,862 deaths. Detection of COVID-19 and pneumonia by the chest X-ray images is of great significance to control the development of the epidemic situation. The current COVID-19 and pneumonia detection system may suffer from two shortcomings: the selection of hyperparameters in the models is not appropriate, and the generalization ability of the model is poor.

Method : To solve the above problems, our team proposed an improved intelligent global optimization algorithm, which is based on the biogeography-based optimization to automatically optimize the hyperparameters value of the models according to different detection objectives. In the optimization progress, after selecting the immigration of suitable index vector and the emigration of suitable index vector, we proposed adding a comparison operation to compare the value of them. According to the different numerical relationships between them, the corresponding operations are performed to improve the migration operation of biogeography-based optimization. The improved algorithm (momentum factor biogeography-based optimization) can better perform the automatic optimization operation. In addition, our team also proposed two frameworks: biogeography convolutional neural network and momentum factor biogeography convolutional neural network. And two methods for detection COVID-19 based on the proposed frameworks.

Results : Our method used three convolutional neural networks (LeNet-5, VGG-16, and ResNet-18) as the basic classification models for chest X-ray images detection of COVID-19, Normal, and Pneumonia. The accuracy of LeNet-5, VGG-16, and ResNet-18 is improved by 1.56%, 1.48%, and 0.73% after using biogeography-based optimization to optimize the hyperparameters of the models. The accuracy of LeNet-5, VGG-16, and ResNet-18 is improved by 2.87%, 6.31%, and 1.46% after using the momentum factor biogeography-based optimization to optimize the hyperparameters of the models.

Conclusion : Under the same experimental conditions, the performance of the momentum factor biogeography-based optimization is superior to the biogeography-based optimization in optimizing the hyperparameters of the convolutional neural networks. Experimental results show that the momentum factor biogeography-based optimization can improve the detection performance of the state-of-the-art approaches in terms of overall accuracy. In future research, we will continue to use and improve other global optimization algorithms to enhance the application ability of deep learning in medical pathological image detection.

Sun Junding, Li Xiang, Tang Chaosheng, Wang Shui-Hua, Zhang Yu-Dong

2021-Sep-15

Biogeography-based optimization, COVID-19, Convolutional neural network, Deep learning, Pneumonia

Surgery Surgery

A Novel CpG Methylation Risk Indicator for Predicting Prognosis in Bladder Cancer.

In Frontiers in cell and developmental biology

Purpose : Bladder cancer (BLCA) is one of the most common cancers worldwide. In a large proportion of BLCA patients, disease recurs and/or progress after resection, which remains a major clinical issue in BLCA management. Therefore, it is vital to identify prognostic biomarkers for treatment stratification. We investigated the efficiency of CpG methylation for the potential to be a prognostic biomarker for patients with BLCA.

Patients and Methods : Overall, 357 BLCA patients from The Cancer Genome Atlas (TCGA) were randomly separated into the training and internal validation cohorts. Least absolute shrinkage and selector operation (LASSO) and support vector machine-recursive feature elimination (SVM-RFE) were used to select candidate CpGs and build the methylation risk score model, which was validated for its prognostic value in the validation cohort by Kaplan-Meier analysis. Hazard curves were generated to reveal the risk nodes throughout the follow-up. Gene Set Enrichment Analysis (GSEA) was used to reveal the potential biological pathways associated with the methylation model. Quantitative real-time polymerase chain reaction (PCR) and western blotting were performed to verify the expression level of the methylated genes.

Results : After incorporating the CpGs obtained by the two algorithms, CpG methylation of eight genes corresponding to TNFAIP8L3, KRTDAP, APC, ZC3H3, COL9A2, SLCO4A1, POU3F3, and ADARB2 were prominent candidate predictors in establishing a methylation risk score for BLCA (MRSB), which was used to divide the patients into high- and low-risk progression groups (p < 0.001). The effectiveness of the MRSB was validated in the internal cohort (p < 0.001). In the MRSB high-risk group, the hazard curve exhibited an initial wide, high peak within 10 months after treatment, whereas some gentle peaks around 2 years were noted. Furthermore, a nomogram comprising MRSB, age, sex, and tumor clinical stage was developed to predict the individual progression risk, and it performed well. Survival analysis implicated the effectiveness of MRSB, which remains significant in all the subgroup analysis based on the clinical features. A functional analysis of MRSB and the corresponding genes revealed potential pathways affecting tumor progression. Validation of quantitative real-time PCR and western blotting revealed that TNFAIP8L3 was upregulated in the BLCA tissues.

Conclusion : We developed the MRSB, an eight-gene-based methylation signature, which has great potential to be used to predict the post-surgery progression risk of BLCA.

Guo Yufeng, Yin Jianjian, Dai Yuanheng, Guan Yudong, Chen Pinjin, Chen Yongqiang, Huang Chenzheng, Lu Yong-Jie, Zhang Lirong, Song Dongkui

2021

DNA methylation, LASSO, SVM-RFE, bladder cancer, machine learning, prognosis

Radiology Radiology

Surveillance Strategy for Barcelona Clinic Liver Cancer B Hepatocellular Carcinoma Achieving Complete Response: An Individualized Risk-Based Machine Learning Study.

In Frontiers in bioengineering and biotechnology

Background: For patients with complete response (CR) of Barcelona Clinical Liver Cancer (BCLC) stage B hepatocellular carcinoma (HCC), there is no consensus regarding the monitoring strategy. Optimal surveillance strategies that can detect early progression of HCC within a limited visit after treatment have not yet been investigated. A retrospective, real-world study was conducted to investigate surveillance strategies for BCLC stage B HCC (BBHCC) patients with CR after curative treatment to support clinical decision making. Methods: From January 2007 to December 2019, 546 BBHCC patients with CR after radical treatment were collected at Sun Yat-sen University Cancer Center. Seventy percent of patients were subjected to the train cohort randomly; the remaining patients comprised the validation cohort to verify the proposed arrangements. The random survival forest method was applied to calculate the disease progression hazard per month, and follow-up schedules were arranged to maximize the capability of progression detection at each visit. The primary endpoint of the study was the delayed-detection months for disease progression. Results: The cumulative 1, 2, and 3-years risk-adjusted probabilities for the train/validation cohorts were 32.8%/33.7%, 54.0%/56.3%, and 64.0%/67.4%, respectively, with peaks around approximately the 9th month. The surveillance regime was primarily concentrated in the first year posttreatment. The delayed-detection months gradually decreased when the total follow-up times increased from 6 to 11. Compared with controls, our schedule reduced delayed detection. Typically, the benefits of our surveillance regimes were obvious when the patients were followed seven times according to our schedule. The optional schedules were 5, 7, 9, 11, 17, 23, and 30 months. Conclusion: The proposed new surveillance schedule may provide a new perspective concerning follow-up for BBHCC patients with CR.

Chen Qi-Feng, Dai Lin, Wu Ying, Huang Zilin, Chen Minshan, Zhao Ming

2021

Barcelona clinic liver cancer B, complete response, hepatocellular carcinoma, machine learning, surveillance strategy

General General

A Structured Pathway Toward Disruption: A Novel HealthTec Innovation Design Curriculum With Entrepreneurship in Mind.

In Frontiers in public health

The typical curriculum of training and educating future clinicians, biomedical engineers, health IT, and artificial intelligence experts lacks needed twenty first-century skills like problem-solving, stakeholder empathy, curiosity stimulation, entrepreneurship, and health economics, which are essential generators and are pre-requirements for creating intentional disruptive innovations. Moreover, the translation from research to a valuable and affordable product/process innovation is not formalized by the current teachings that focus on short-term rather than long-term developments, leading to inaccurate and incremental forecasting on the future of healthcare and longevity. The Stanford Biodesign approach of unmet clinical need detection would be an excellent starting methodology for health-related innovation work, although unfortunately not widely taught yet. We have developed a novel lecture titled HealthTec Innovation Design (HTID) offered in an interdisciplinary setup to medical students and biomedical engineers. It teaches a future-oriented view and the application and effects of exponential trends. We implemented a novel approach using the Purpose Launchpad meta-methodology combined with other innovation generation tools to define, experiment, and validate existing project ideas. As part of the process of defining the novel curriculum, we used experimentation methods, like a global science fiction event to create a comic book with Future Health stories and an Innovation Think Tank Certification Program of a large medical technology company that is focused on identifying future health opportunities. We conducted before and after surveys and concluded that the proposed initiatives were impactful in developing an innovative design thinking approach. Participants' awareness and enthusiasm were raised, including their willingness to implement taught skills, values, and methods in their working projects. We conclude that a new curriculum based on HTID is essential and needed to move the needle of healthcare activities from treating sickness to maintaining health.

Fritzsche Holger, Barbazzeni Beatrice, Mahmeen Mohd, Haider Sultan, Friebe Michael

2021

biodesign, bioengineering education, design thinking, disruptive technologies, exponential medicine, future of health, health democratization, twenty-first century skills

General General

Automated Quantitative Stress Perfusion Cardiac Magnetic Resonance in Pediatric Patients.

In Frontiers in pediatrics

Background: Myocardial ischemia occurs in pediatrics, as a result of both congenital and acquired heart diseases, and can lead to further adverse cardiac events if untreated. The aim of this work is to assess the feasibility of fully automated, high resolution, quantitative stress myocardial perfusion cardiac magnetic resonance (CMR) in a cohort of pediatric patients and to evaluate its agreement with the coronary anatomical status of the patients. Methods: Fourteen pediatric patients, with 16 scans, who underwent dual-bolus stress perfusion CMR were retrospectively analyzed. All patients also had anatomical coronary assessment with either CMR, CT, or X-ray angiography. The perfusion CMR images were automatically processed and quantified using an analysis pipeline previously developed in adults. Results: Automated perfusion quantification was successful in 15/16 cases. The coronary perfusion territories supplied by vessels affected by a medium/large aneurysm or stenosis (according to the AHA guidelines), induced by Kawasaki disease, an anomalous origin, or interarterial course had significantly reduced myocardial blood flow (MBF) (median (interquartile range), 1.26 (1.05, 1.67) ml/min/g) as compared to territories supplied by unaffected coronaries [2.57 (2.02, 2.69) ml/min/g, p < 0.001] and territories supplied by vessels with a small aneurysm [2.52 (2.45, 2.83) ml/min/g, p = 0.002]. Conclusion: Automatic CMR-derived MBF quantification is feasible in pediatric patients, and the technology could be potentially used for objective non-invasive assessment of ischemia in children with congenital and acquired heart diseases.

Scannell Cian M, Hasaneen Hadeer, Greil Gerald, Hussain Tarique, Razavi Reza, Lee Jack, Pushparajah Kuberan, Duong Phuoc, Chiribiri Amedeo

2021

Kawasaki disease, automated quantitative stress perfusion, cardiac magnetic resonance, deep learning, pediatrics

Ophthalmology Ophthalmology

Who needs myopia control?

In International journal of ophthalmology

Myopia has become a major visual disorder among school-aged children in East Asia due to its rising prevalence over the past few decades and will continue to be a leading health issue with an annual incidence as high as 20%-30%. Although various interventions have been proposed for myopia control, consensus in treatment strategies has yet to be fully developed. Atropine and orthokeratology stand out for their effectiveness in myopia progression control, but children with rapid progression of myopia require treatment with higher concentrations of atropine that are associated with increased rates of side effects, or with orthokeratology that carries risk of significant complication. Therefore, improved risk assessment for myopia onset and progression in children is critical in clinical decision-making. Besides traditional prediction models based on genetic effects and environmental exposures within populations, individualized prediction using machine learning and data based on age-specific refraction is promising. Although emerging treatments for myopia are promising and some have been incorporated into clinical practice, identifying populations who require and benefit from intervention remains the most important initial step for clinical practice.

Chen Yan-Xian, Liao Chi-Mei, Tan Zachary, He Ming-Guang

2021

intervention, myopia control, prediction

General General

Diagnosis of Pediatric Pneumonia with Ensemble of Deep Convolutional Neural Networks in Chest X-Ray Images.

In Arabian journal for science and engineering

Pneumonia is a fatal disease that appears in the lungs and is caused by viral or bacterial infection. Diagnosis of pneumonia in chest X-ray images can be difficult and error-prone because of its similarity with other infections in the lungs. The aim of this study is to develop a computer-aided pneumonia detection system to facilitate the diagnosis decision process. Therefore, a convolutional neural network (CNN) ensemble method was proposed for the automatic diagnosis of pneumonia which is seen in children. In this context, seven well-known CNN models (VGG-16, VGG-19, ResNet-50, Inception-V3, Xception, MobileNet, and SqueezeNet) pre-trained on the ImageNet dataset were trained with the appropriate transfer learning and fine-tuning strategies on the chest X-ray dataset. Among the seven different models, the three most successful ones were selected for the ensemble method. The final results were obtained by combining the predictions of CNN models with the ensemble method during the test. In addition, a CNN model was trained from scratch, and the results of this model were compared with the proposed ensemble method. The proposed ensemble method achieved remarkable results with an AUC of 95.21 and a sensitivity of 97.76 on the test data. Also, the proposed ensemble method achieved classification accuracy of 90.71 in chest X-ray images as normal, viral pneumonia, and bacterial pneumonia.

Ayan Enes, Karabulut Bergen, Ünver Halil Murat

2021-Sep-12

Convolutional neural networks, Deep learning, Medical image analysis, Pneumonia, Transfer learning

General General

Determination of COVID-19 Vaccine Hesitancy Among University Students.

In Cureus

Introduction With the sudden outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-COV-2), vaccines appear to be the most efficient measure in combating spread. However, vaccines are only effective if a community collectively uptakes vaccination. This approach is growing increasingly difficult with the emergence of 'Vaccine Hesitancy.' This paper aims to determine the association between university curricula and the degree of hesitancy for the COVID-19 vaccine. Methods The online questionnaire assessed demographic data, prior knowledge of vaccines, attitude towards COVID-19 vaccines using an adapted version of the WHO Strategic Advisory Group of Experts (SAGE) Working Group's Vaccine Hesitancy Survey (VHS) and factors likely to motivate vaccine uptake. By using binary scoring, the degree of hesitancy among students was determined. Exploratory Factor Analysis (EFA) on VHS revealed underlying causes of hesitancy. To analyze the dependence between hesitancy and curriculum, a chi-squared test was conducted. Results Medical students scored higher for prior knowledge of vaccines (M = 3.54) as opposed to non-medical students (M = 3.49). Medical students responded favorably to COVID-19 vaccines with only 1.37% showing hesitancy for all nine items of VHS, compared to 2.55% of non-medical students. EFA produced three subscales within the VHS: lack of confidence, risk factor concern, and misinformation. The lack of confidence factor accounted for 65% of the data obtained. The chi-square test solidified that vaccine hesitancy is dependent on curriculum. Conclusion The majority of non-medical students showed hesitancy towards obtaining COVID-19 vaccines compared to medical students who were more willing, largely owing to their knowledge and understanding of vaccines.

Sadaqat Waliya, Habib Shanzay, Tauseef Ambreen, Akhtar Sheharyar, Hayat Meryum, Shujaat Syeda A, Mahmood Amina

2021-Aug

covid-19, curriculum, hesitancy, university students, vaccine

General General

Accuracy of health-related information regarding COVID-19 on Twitter during a global pandemic.

In World medical & health policy

This study was performed to analyze the accuracy of health-related information on Twitter during the coronavirus disease 2019 (COVID-19) pandemic. Authors queried Twitter on three dates for information regarding COVID-19 and five terms (cure, emergency or emergency room, prevent or prevention, treat or treatments, vitamins or supplements) assessing the first 25 results with health-related information. Tweets were authoritative if written by governments, hospitals, or physicians. Two physicians assessed each tweet for accuracy. Metrics were compared between accurate and inaccurate tweets using χ 2 analysis and Mann-Whitney U. A total of 25.4% of tweets were inaccurate. Accurate tweets were more likely written by Twitter authenticated authors (49.8% vs. 20.9%, 28.9% difference, 95% confidence interval [CI]: 17.7-38.2) with accurate tweet authors having more followers (19,491 vs. 7346; 3446 difference, 95% CI: 234-14,054) versus inaccurate tweet authors. Likes, retweets, tweet length, botometer scores, writing grade level, and rank order did not differ between accurate and inaccurate tweets. We found 1/4 of health-related COVID-19 tweets inaccurate indicating that the public should not rely on COVID-19 health information written on Twitter. Ideally, improved government regulatory authority, public/private industry oversight, independent fact-checking, and artificial intelligence algorithms are needed to ensure inaccurate information on Twitter is removed.

Swetland Sarah B, Rothrock Ava N, Andris Halle, Davis Bennett, Nguyen Linh, Davis Phil, Rothrock Steven G

2021-Jul-29

COVID‐19, pandemic, social media

General General

Why cannot long-term cascade be predicted? Exploring temporal dynamics in information diffusion processes.

In Royal Society open science

Predicting information cascade plays a crucial role in various applications such as advertising campaigns, emergency management and infodemic controlling. However, predicting the scale of an information cascade in the long-term could be difficult. In this study, we take Weibo, a Twitter-like online social platform, as an example, exhaustively extract predictive features from the data, and use a conventional machine learning algorithm to predict the information cascade scales. Specifically, we compare the predictive power (and the loss of it) of different categories of features in short-term and long-term prediction tasks. Among the features that describe the user following network, retweeting network, tweet content and early diffusion dynamics, we find that early diffusion dynamics are the most predictive ones in short-term prediction tasks but lose most of their predictive power in long-term tasks. In-depth analyses reveal two possible causes of such failure: the bursty nature of information diffusion and feature temporal drift over time. Our findings further enhance the comprehension of the information diffusion process and may assist in the control of such a process.

Cao Ren-Meng, Liu Xiao Fan, Xu Xiao-Ke

2021-Sep

cascade prediction, information diffusion, online social network

General General

Data-driven operation of the resilient electric grid: A case of COVID-19.

In Journal of engineering (Stevenage, England)

Electrical energy is a vital part of modern life, and expectations for grid resilience to allow a continuous and reliable energy supply has tremendously increased even during adverse events (e.g. Ukraine cyberattack, Hurricane Maria). The global pandemic COVID-19 has raised the electric energy reliability risk due to potential workforce disruptions, supply chain interruptions, and increased possible cybersecurity threats. Additionally, the pandemic introduces a significant degree of uncertainty to the grid operation in the presence of other challenges including aging power grids, high proliferation of distributed generation, market mechanism, and active distribution network. This situation increases the need for measures for the resiliency of power grids to mitigate the impact of the pandemic as well as simultaneous extreme events including cyberattacks and adverse weather events. Solutions to manage such an adverse scenario will be multi-fold: (a) emergency planning and organisational support, (b) following safety protocol, (c) utilising enhanced automation and sensing for situational awareness, and (d) integration of advanced technologies and data points for ML-driven enhanced decision support. Enhanced digitalisation and automation resulted in better network visibility at various levels, including generation, transmission, and distribution. These data or information can be employed to take advantage of advanced machine learning techniques for automation and increased power grid resilience. In this paper, the resilience of power grids in the face of pandemics is explored and various machine learning tools that can be helpful to augment human operators are discused by: (a) reviewing the impact of COVID-19 on power grid operations and actions taken by operators/organisations to minimise the impact of COVID-19, and (b) presenting recently developed tools and concepts of machine learning and artificial intelligence that can be applied to increase the resiliency of power systems in normal and extreme scenarios such as the COVID-19 pandemic.

Noorazar H, Srivastava A, Pannala S, K Sadanandan Sajan

2021-Aug-09

Pathology Pathology

The Future of the Private Gastroenterology Practice.

In Gastrointestinal endoscopy clinics of North America

The future private gastroenterology practice will be a large multidisciplinary practice including a clinic, AEC, pathology services, infusion services, anesthesia services, pharmacy services, and imaging centers. Delivery of gastrointestinal (GI) services will be a team-based clinic with AEC access and improved quality of care. Competing technologies will drive practices to promote the value of colonoscopy as the best screening test for colon cancer. Artificial intelligence (AI) may significantly alter our approach to clinic and endoscopic services. The creative and intellectual capital of practice leaders will continue to define the private GI practice of the future.

Shiels Aaron J, Vicari Joseph J

2021-Oct

Ancillaries, Artificial intelligence, Competing technology, Cost, Efficiency, Hospital service, Reimbursement, Telehealth

General General

Adoption of Improved Neural Network Blade Pattern Recognition in Prevention and Control of Corona Virus Disease-19 Pandemic.

In Pattern recognition letters

To explore the adoption effect of improved neural network blade pattern in corona virus disease (COVID)-19, comparative analysis is implemented. First, the following hypotheses are proposed. I: in addition to the confirmed cases and deaths, people suspected of being infected are also involved in the spread of the epidemic. II: patients who have been cured may also develop secondary infections, so it is considered that there is still a link between cured cases and the spread of the epidemic. III: only the relevant data of the previous day is used to predict the epidemic prevention and control of the next day. Then, the epidemic data from February 1st to February 15th in X province were selected as the control. The combined neural network model is used for prevention and control prediction, and the prediction results of the traditional neural network model are compared. The results show that the predictions of the daily new cases by the five neural network models have little difference with the actual value, and the trend is basically consistent. However, there are still differences in some time nodes. The errors of neural network 1 on the 6th and network 3 on the 13th are large. The accuracy of the combined neural network prediction model is high, and there is little difference between the result and the actual value at each time node. The prediction of the cumulative number of diagnoses per day of the five neural network models is also analyzed, and the results are relatively ideal. In addition, the accuracy of the combined neural network prediction model is high, and the difference between the result and the actual value at each time node is relatively small. It is found that the standard deviations of neural networks 2 and 3 are relatively high through the comparison of the deviations. The deviation means of the five models were all relatively low, and the mean deviation and standard deviation of the combined neural network model are the lowest. It is found that the accuracy of prediction on the epidemic spread in this province is good by comparing the performance of each neural network model. Regarding various indicators, the prediction accuracy of the combined neural network model is higher than that of the other four models, and its performance is also the best. Finally, the MSE of the improved neural network model is lower compared with the traditional neural network model. Moreover, with the change of learning times, the change trend of MSE is constant (P <0.05 for all). In short, the improved neural network blade model has better performance compared with that of the traditional neural network blade model. The prediction results of the epidemic situation are accurate, and the application effect is remarkable, so the proposed model is worthy of further promotion and application in the medical field.

Ma Yanli, Li Zhonghua, Gou Jixiang, Ding Lihua, Yang Dong, Feng Guiliang

2021-Sep-15

artificial intelligence, corona virus disease (COVID)-19, improved neural network blade model, neural network model

oncology Oncology

Applying digital storytelling in the medical oncology curriculum: Effects on students' achievement and critical thinking.

In Annals of medicine and surgery (2012)

Background : Digital storytelling (DST), which combines traditional storytelling with digital tools, can provide a narrative pedagogy that promotes critical thinking (CT). However, we found no previous study in medical education.

Materials and methods : The aim of the study was to investigate if DST can promote CT and, if so, which CT skills were improved. Thirty-two students participated in a non-equivalent control group pretest-posttest research study, with 16 in each group. The participants were fifth-year medical students on a hematology rotation. We compared the routine instructional method (control group) with DST (intervention group). The measures of CT used for the pre- and post-test in both groups was the Health Science Reasoning Test (HRST) and knowledge test. We also evaluated the satisfaction of the students in DST group. We used Paired and independent t-tests for comparing the mean scores. To eliminate the confounding effect of pre-test on the results of the intervention, the ANCOVA test was used.

Results : There was no significant difference in the overall CT pretest scores (P-value = 0.51) between the control and intervention groupsbut the difference was significant for the post-test scores (P-value = 0.03). Although post-test scores showed a significant increase (P-value = 0.002) compared to pre-test scores in the intervention group, no significant increase was observed in the control group (P-value = 0.26). Most students considered that DST improved their CT, deep learning, communication skills and team-working.

Conclusions : The study demonstrated that DST promoted CT. We recommend the use of DST to promote CT in clinical education placements.

Zarei Afagh, Mojtahedzadeh Rita, Mohammadi Aeen, Sandars John, Hossein Emami Seyed Amir

2021-Oct

Critical thinking, Digital storytelling, Medical education

General General

Comparing Machine Learning Methods to Improve Fall Risk Detection in Elderly with Osteoporosis from Balance Data.

In Journal of healthcare engineering

Falls are a multifactorial cause of injuries for older people. Subjects with osteoporosis are particularly vulnerable to falls. We study the performance of different computational methods to identify people with osteoporosis who experience a fall by analysing balance parameters. Balance parameters, from eyes open and closed posturographic studies, and prospective registration of falls were obtained from a sample of 126 community-dwelling older women with osteoporosis (age 74.3 ± 6.3) using World Health Organization Questionnaire for the study of falls during a follow-up of 2.5 years. We analyzed model performance to determine falls of every developed model and to validate the relevance of the selected parameter sets. The principal findings of this research were (1) models built using oversampling methods with either IBk (KNN) or Random Forest classifier can be considered good options for a predictive clinical test and (2) feature selection for minority class (FSMC) method selected previously unnoticed balance parameters, which implies that intelligent computing methods can extract useful information with attributes which otherwise are disregarded by experts. Finally, the results obtained suggest that Random Forest classifier using the oversampling method to balance the data independent of the set of variables used got the best overall performance in measures of sensitivity (>0.71), specificity (>0.18), positive predictive value (PPV >0.74), and negative predictive value (NPV >0.66) independent of the set of variables used. Although the IBk classifier was built with oversampling data considering information from both eyes opened and closed, using all variables got the best performance (sensitivity >0.81, specificity >0.19, PPV = 0.97, and NPV = 0.66).

Cuaya-Simbro German, Perez-Sanpablo Alberto-I, Morales Eduardo-F, Quiñones Uriostegui Ivett, Nuñez-Carrera Lidia

2021

General General

Passive Fetal Movement Signal Detection System Based on Intelligent Sensing Technology.

In Journal of healthcare engineering

Fetal movement (FM) is an essential physiological parameter to determine the health status of the fetus. To address the problems of harrowing FM signal extraction and the low recognition rate of traditional machine learning classifiers in FM signal detection, this paper develops a passive FM signal detection system based on intelligent sensing technology. FM signals are obtained from the abdomen of the pregnant woman by using accelerometers. The FM signals are extracted and identified according to the clinical nature of the features hidden in the amplitude and waveform of the FM signals that fluctuate in duration. The system consists of four main stages: (i) FM signal preprocessing, (ii) maternal artifact signal preidentification, (iii) FM signal identification, and (iv) FM classification. Firstly, Kalman filtering is used to reconstruct the FM signal in a continuous low-amplitude noise background. Secondly, the maternal artifact signal is identified using an amplitude threshold algorithm. Then, an innovative dictionary learning algorithm is used to construct a dictionary of FM features, and orthogonal matching pursuit and adaptive filtering algorithms are used to identify the FM signals, respectively. Finally, mask fusion classification is performed based on the multiaxis recognition results. Experiments are conducted to evaluate the performance of the proposed FM detection system using publicly available and self-built accelerated FM datasets. The classification results showed that the orthogonal matching pursuit algorithm was more effective than the adaptive filtering algorithm in identifying FM signals, with a positive prediction value of 89.74%. The proposed FM detection system has great potential and promise for wearable FM health monitoring.

Liang Sensong, Peng Jiansheng, Xu Yong

2021

oncology Oncology

All around suboptimal health - a joint position paper of the Suboptimal Health Study Consortium and European Association for Predictive, Preventive and Personalised Medicine.

In The EPMA journal

First two decades of the twenty-first century are characterised by epidemics of non-communicable diseases such as many hundreds of millions of patients diagnosed with cardiovascular diseases and the type 2 diabetes mellitus, breast, lung, liver and prostate malignancies, neurological, sleep, mood and eye disorders, amongst others. Consequent socio-economic burden is tremendous. Unprecedented decrease in age of maladaptive individuals has been reported. The absolute majority of expanding non-communicable disorders carry a chronic character, over a couple of years progressing from reversible suboptimal health conditions to irreversible severe pathologies and cascading collateral complications. The time-frame between onset of SHS and clinical manifestation of associated disorders is the operational area for an application of reliable risk assessment tools and predictive diagnostics followed by the cost-effective targeted prevention and treatments tailored to the person. This article demonstrates advanced strategies in bio/medical sciences and healthcare focused on suboptimal health conditions in the frame-work of Predictive, Preventive and Personalised Medicine (3PM/PPPM). Potential benefits in healthcare systems and for society at large include but are not restricted to an improved life-quality of major populations and socio-economical groups, advanced professionalism of healthcare-givers and sustainable healthcare economy. Amongst others, following medical areas are proposed to strongly benefit from PPPM strategies applied to the identification and treatment of suboptimal health conditions:Stress overload associated pathologiesMale and female healthPlanned pregnanciesPeriodontal healthEye disordersInflammatory disorders, wound healing and pain management with associated complicationsMetabolic disorders and suboptimal body weightCardiovascular pathologiesCancersStroke, particularly of unknown aetiology and in young individualsSleep medicineSports medicineImproved individual outcomes under pandemic conditions such as COVID-19.

Wang Wei, Yan Yuxiang, Guo Zheng, Hou Haifeng, Garcia Monique, Tan Xuerui, Anto Enoch Odame, Mahara Gehendra, Zheng Yulu, Li Bo, Kang Timothy, Zhong Zhaohua, Wang Youxin, Guo Xiuhua, Golubnitschaja Olga

2021-Sep-13

Adolescence, Artificial intelligence (AI), Behavioural patterns, Big data management, Body mass index (BMI), COVID-19, Cancers, Cardiovascular disease, Communicable, Dietary habits, Epidemics, Glycan, Health economy, Health policy, Individualised patient profile, Lifestyle, Liquid biopsy, Medical ethics, Microbiome, Modifiable preventable risks, Mood disorders, Multi-level diagnostics, Multi-parametric analysis, Natural substances, Neurologic diseases, Non-communicable diseases, Omics, Pandemics, Periodontal health, Predictive preventive personalised medicine (PPPM/3PM), Risk assessment, Sleep medicine, Stress overload, Suboptimal health status (SHS), Traditional medicine

General General

GEOMScope: Large Field-of-view 3D Lensless Microscopy with Low Computational Complexity.

In Laser & photonics reviews

Imaging systems with miniaturized device footprint, real-time processing speed and high resolution three-dimensional (3D) visualization are critical to broad biomedical applications such as endoscopy. Most of existing imaging systems rely on bulky lenses and mechanically refocusing to perform 3D imaging. Here, we demonstrate GEOMScope, a lensless single-shot 3D microscope that forms image through a single layer of thin microlens array and reconstructs objects through an innovative algorithm combining geometrical-optics-based pixel back projection and background suppressions. We verify the effectiveness of GEOMScope on resolution target, fluorescent particles and volumetric objects. Comparing to other widefield lensless imaging devices, we significantly reduce the required computational resource and increase the reconstruction speed by orders of magnitude. This enables us to image and recover large volume 3D object in high resolution with near real-time processing speed. Such a low computational complexity is attributed to the joint design of imaging optics and reconstruction algorithms, and a joint application of geometrical optics and machine learning in the 3D reconstruction. More broadly, the excellent performance of GEOMScope in imaging resolution, volume, and reconstruction speed implicates that geometrical optics could greatly benefit and play an important role in computational imaging.

Tian Feng, Hu Junjie, Yang Weijian

2021-Aug

3D imaging, 3D microscopy, computational imaging, geometrical optics, lensless imaging, light field, microlens array

General General

Decoding Optical Data with Machine Learning.

In Laser & photonics reviews

Optical spectroscopy and imaging techniques play important roles in many fields such as disease diagnosis, biological study, information technology, optical science, and materials science. Over the past decade, machine learning (ML) has proved promising in decoding complex data, enabling rapid and accurate analysis of optical spectra and images. This review aims to shed light on various ML algorithms for optical data analysis with a focus on their applications in a wide range of fields. The goal of this work is to sketch the validity of ML-based optical data decoding. The review concludes with an outlook on unaddressed problems and opportunities in this emerging subject that interfaces optics, data science and ML.

Fang Jie, Swain Anand, Unni Rohit, Zheng Yuebing

2021-Feb

data decoding, machine learning, optical data, optics

General General

Automated Detection of COVID-19 Cough.

In Biomedical signal processing and control

Easy detection of COVID-19 is a challenge. Quick biological tests do not give enough accuracy. Success in the fight against new outbreaks depends not only on the efficiency of the tests used, but also on the cost, time elapsed and the number of tests that can be done massively. Our proposal provides a solution to this challenge. The main objective is to design a freely available, quick and efficient methodology for the automatic detection of COVID-19 in raw audio files. Our proposal is based on automated extraction of time-frequency cough features and selection of the more significant ones to be used to diagnose COVID-19 using a supervised machine-learning algorithm. Random Forest has performed better than the other models analysed in this study. An accuracy close to 90% was obtained. This study demonstrates the feasibility of the automatic diagnose of COVID-19 from coughs, and its applicability to detecting new outbreaks.

Tena Alberto, Clarià Francesc, Solsona Francesc

2021-Sep-13

COVID-19, automated cough detection, diagnosis, signal processing, time-frequency.

General General

Strategies to Improve the Quality and Freshness of Human Bone Marrow-Derived Mesenchymal Stem Cells for Neurological Diseases.

In Stem cells international

Human bone marrow-derived mesenchymal stem cells (hBM-MSCs) have been studied for their application to manage various neurological diseases, owing to their anti-inflammatory, immunomodulatory, paracrine, and antiapoptotic ability, as well as their homing capacity to specific regions of brain injury. Among mesenchymal stem cells, such as BM-MSCs, adipose-derived MSCs, and umbilical cord MSCs, BM-MSCs have many merits as cell therapeutic agents based on their widespread availability and relatively easy attainability and in vitro handling. For stem cell-based therapy with BM-MSCs, it is essential to perform ex vivo expansion as low numbers of MSCs are obtained in bone marrow aspirates. Depending on timing, before hBM-MSC transplantation into patients, after detaching them from the culture dish, cell viability, deformability, cell size, and membrane fluidity are decreased, whereas reactive oxygen species generation, lipid peroxidation, and cytosolic vacuoles are increased. Thus, the quality and freshness of hBM-MSCs decrease over time after detachment from the culture dish. Especially, for neurological disease cell therapy, the deformability of BM-MSCs is particularly important in the brain for the development of microvessels. As studies on the traditional characteristics of hBM-MSCs before transplantation into the brain are very limited, omics and machine learning approaches are needed to evaluate cell conditions with indepth and comprehensive analyses. Here, we provide an overview of hBM-MSCs, the application of these cells to various neurological diseases, and improvements in their quality and freshness based on integrated omics after detachment from the culture dish for successful cell therapy.

Lee Da Yeon, Lee Sung Eun, Kwon Do Hyeon, Nithiyanandam Saraswathy, Lee Mi Ha, Hwang Ji Su, Basith Shaherin, Ahn Jung Hwan, Shin Tae Hwan, Lee Gwang

2021

General General

Study on 3D Clothing Color Application Based on Deep Learning-Enabled Macro-Micro Adversarial Network and Human Body Modeling.

In Computational intelligence and neuroscience

In real life, people's life gradually tends to be simple, so the convenience of online shopping makes more and more research begin to explore the convenience optimization of shopping, in which the fitting system is the research product. However, due to the immaturity of the virtual fitting system, there are a lot of problems, such as the expression of clothing color is not clear or deviation. In view of this, this paper proposes a 3D clothing color display model based on deep learning to support human modeling-driven. Firstly, the macro-micro adversarial network (MMAN) based on deep learning is used to analyze the original image, and then, the results are preprocessed. Finally, the 3D model with the original image color is constructed by using UV mapping. The experimental results show that the accuracy of the MMAN algorithm reaches 0.972, the established three-dimensional model is emotional enough, the expression of the clothing color is clear, and the difference between the color difference and the original image is within 0.01, and the subjective evaluation of volunteers is more than 90 points. The above results show that it is effective to use deep learning to build a 3D model with the original picture clothing color, which has great guiding significance for the research of character model modeling and simulation.

Liu Jingmiao, Ren Yu, Qin Xiaotong

2021

General General

Data Management and Modeling in Plant Biology.

In Frontiers in plant science

The study of plant-environment interactions is a multidisciplinary research field. With the emergence of quantitative large-scale and high-throughput techniques, amount and dimensionality of experimental data have strongly increased. Appropriate strategies for data storage, management, and evaluation are needed to make efficient use of experimental findings. Computational approaches of data mining are essential for deriving statistical trends and signatures contained in data matrices. Although, current biology is challenged by high data dimensionality in general, this is particularly true for plant biology. Plants as sessile organisms have to cope with environmental fluctuations. This typically results in strong dynamics of metabolite and protein concentrations which are often challenging to quantify. Summarizing experimental output results in complex data arrays, which need computational statistics and numerical methods for building quantitative models. Experimental findings need to be combined by computational models to gain a mechanistic understanding of plant metabolism. For this, bioinformatics and mathematics need to be combined with experimental setups in physiology, biochemistry, and molecular biology. This review presents and discusses concepts at the interface of experiment and computation, which are likely to shape current and future plant biology. Finally, this interface is discussed with regard to its capabilities and limitations to develop a quantitative model of plant-environment interactions.

Krantz Maria, Zimmer David, Adler Stephan O, Kitashova Anastasia, Klipp Edda, Mühlhaus Timo, Nägele Thomas

2021

differential equations, genome-scale networks, machine learning, mathematical modeling, metabolic regulation, omics analysis, plant-environment interactions

Public Health Public Health

The future of zoonotic risk prediction.

In Philosophical transactions of the Royal Society of London. Series B, Biological sciences

In the light of the urgency raised by the COVID-19 pandemic, global investment in wildlife virology is likely to increase, and new surveillance programmes will identify hundreds of novel viruses that might someday pose a threat to humans. To support the extensive task of laboratory characterization, scientists may increasingly rely on data-driven rubrics or machine learning models that learn from known zoonoses to identify which animal pathogens could someday pose a threat to global health. We synthesize the findings of an interdisciplinary workshop on zoonotic risk technologies to answer the following questions. What are the prerequisites, in terms of open data, equity and interdisciplinary collaboration, to the development and application of those tools? What effect could the technology have on global health? Who would control that technology, who would have access to it and who would benefit from it? Would it improve pandemic prevention? Could it create new challenges? This article is part of the theme issue 'Infectious disease macroecology: parasite diversity and dynamics across the globe'.

Carlson Colin J, Farrell Maxwell J, Grange Zoe, Han Barbara A, Mollentze Nardus, Phelan Alexandra L, Rasmussen Angela L, Albery Gregory F, Bett Bernard, Brett-Major David M, Cohen Lily E, Dallas Tad, Eskew Evan A, Fagre Anna C, Forbes Kristian M, Gibb Rory, Halabi Sam, Hammer Charlotte C, Katz Rebecca, Kindrachuk Jason, Muylaert Renata L, Nutter Felicia B, Ogola Joseph, Olival Kevin J, Rourke Michelle, Ryan Sadie J, Ross Noam, Seifert Stephanie N, Sironen Tarja, Standley Claire J, Taylor Kishana, Venter Marietjie, Webala Paul W

2021-Nov-08

access and benefit sharing, epidemic risk, global health, machine learning, viral ecology, zoonotic risk

General General

Software system to predict the infection in COVID-19 patients using deep learning and web of things.

In Software: practice & experience

Since the end of 2019, computed tomography (CT) images have been used as an important substitute for the time-consuming Reverse Transcriptase polymerase chain reaction (RT-PCR) test; a new coronavirus 2019 (COVID-19) disease has been detected and has quickly spread through many countries across the world. Medical imaging such as computed tomography provides great potential due to growing skepticism toward the sensitivity of RT-PCR as a screening tool. For this purpose, automated image segmentation is highly desired for a clinical decision aid and disease monitoring. However, there is limited publicly accessible COVID-19 image knowledge, leading to the overfitting of conventional approaches. To address this issue, the present paper focuses on data augmentation techniques to create synthetic data. Further, a framework has been proposed using WoT and traditional U-Net with EfficientNet B0 to segment the COVID Radiopedia and Medseg datasets automatically. The framework achieves an F-score of 0.96, which is best among state-of-the-art methods. The performance of the proposed framework also computed using Sensitivity, Specificity, and Dice-coefficient, achieves 84.5%, 93.9%, and 65.0%, respectively. Finally, the proposed work is validated using three quality of service (QoS) parameters such as server latency, response time, and network latency which improves the performance by 8%, 7%, and 10%, respectively.

Singh Ashima, Kaur Amrita, Dhillon Arwinder, Ahuja Sahil, Vohra Harpreet

2021-Jun-24

COVID‐19, EfficientNet B0, SARSCoV‐2, U‐net, WoT, deep learning, segmentation

General General

A Deep Learning-Based Method for Automatic Assessment of Stomatal Index in Wheat Microscopic Images of Leaf Epidermis.

In Frontiers in plant science

The stomatal index of the leaf is the ratio of the number of stomata to the total number of stomata and epidermal cells. Comparing with the stomatal density, the stomatal index is relatively constant in environmental conditions and the age of the leaf and, therefore, of diagnostic characteristics for a given genotype or species. Traditional assessment methods involve manual counting of the number of stomata and epidermal cells in microphotographs, which is labor-intensive and time-consuming. Although several automatic measurement algorithms of stomatal density have been proposed, no stomatal index pipelines are currently available. The main aim of this research is to develop an automated stomatal index measurement pipeline. The proposed method employed Faster regions with convolutional neural networks (R-CNN) and U-Net and image-processing techniques to count stomata and epidermal cells, and subsequently calculate the stomatal index. To improve the labeling speed, a semi-automatic strategy was employed for epidermal cell annotation in each micrograph. Benchmarking the pipeline on 1,000 microscopic images of leaf epidermis in the wheat dataset (Triticum aestivum L.), the average counting accuracies of 98.03 and 95.03% for stomata and epidermal cells, respectively, and the final measurement accuracy of the stomatal index of 95.35% was achieved. R 2 values between automatic and manual measurement of stomata, epidermal cells, and stomatal index were 0.995, 0.983, and 0.895, respectively. The average running time (ART) for the entire pipeline could be as short as 0.32 s per microphotograph. The proposed pipeline also achieved a good transferability on the other families of the plant using transfer learning, with the mean counting accuracies of 94.36 and 91.13% for stomata and epidermal cells and the stomatal index accuracy of 89.38% in seven families of the plant. The pipeline is an automatic, rapid, and accurate tool for the stomatal index measurement, enabling high-throughput phenotyping, and facilitating further understanding of the stomatal and epidermal development for the plant physiology community. To the best of our knowledge, this is the first deep learning-based microphotograph analysis pipeline for stomatal index assessment.

Zhu Chuancheng, Hu Yusong, Mao Hude, Li Shumin, Li Fangfang, Zhao Congyuan, Luo Lin, Liu Weizhen, Yuan Xiaohui

2021

cell counting, convolutional network, stomata detection, stomatal index, transfer learning

General General

High-Throughput Phenotyping and Random Regression Models Reveal Temporal Genetic Control of Soybean Biomass Production.

In Frontiers in plant science

Understanding temporal accumulation of soybean above-ground biomass (AGB) has the potential to contribute to yield gains and the development of stress-resilient cultivars. Our main objectives were to develop a high-throughput phenotyping method to predict soybean AGB over time and to reveal its temporal quantitative genomic properties. A subset of the SoyNAM population (n = 383) was grown in multi-environment trials and destructive AGB measurements were collected along with multispectral and RGB imaging from 27 to 83 days after planting (DAP). We used machine-learning methods for phenotypic prediction of AGB, genomic prediction of breeding values, and genome-wide association studies (GWAS) based on random regression models (RRM). RRM enable the study of changes in genetic variability over time and further allow selection of individuals when aiming to alter the general response shapes over time. AGB phenotypic predictions were high (R 2 = 0.92-0.94). Narrow-sense heritabilities estimated over time ranged from low to moderate (from 0.02 at 44 DAP to 0.28 at 33 DAP). AGB from adjacent DAP had highest genetic correlations compared to those DAP further apart. We observed high accuracies and low biases of prediction indicating that genomic breeding values for AGB can be predicted over specific time intervals. Genomic regions associated with AGB varied with time, and no genetic markers were significant in all time points evaluated. Thus, RRM seem a powerful tool for modeling the temporal genetic architecture of soybean AGB and can provide useful information for crop improvement. This study provides a basis for future studies to combine phenotyping and genomic analyses to understand the genetic architecture of complex longitudinal traits in plants.

Freitas Moreira Fabiana, Rojas de Oliveira Hinayah, Lopez Miguel Angel, Abughali Bilal Jamal, Gomes Guilherme, Cherkauer Keith Aric, Brito Luiz Fernando, Rainey Katy Martin

2021

Glycine max, digital agriculture, longitudinal traits, phenomics, plant breeding, quantitative genetics, time series

General General

Artificial Intelligence Predicts Severity of COVID-19 Based on Correlation of Exaggerated Monocyte Activation, Excessive Organ Damage and Hyperinflammatory Syndrome: A Prospective Clinical Study.

In Frontiers in immunology ; h5-index 100.0

Background : Prediction of the severity of COVID-19 at its onset is important for providing adequate and timely management to reduce mortality.

Objective : To study the prognostic value of damage parameters and cytokines as predictors of severity of COVID-19 using an extensive immunologic profiling and unbiased artificial intelligence methods.

Methods : Sixty hospitalized COVID-19 patients (30 moderate and 30 severe) and 17 healthy controls were included in the study. The damage indicators high mobility group box 1 (HMGB1), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), alanine aminotransferase (ALT), extensive biochemical analyses, a panel of 47 cytokines and chemokines were analyzed at weeks 1, 2 and 7 along with clinical complaints and CT scans of the lungs. Unbiased artificial intelligence (AI) methods (logistic regression and Support Vector Machine and Random Forest algorithms) were applied to investigate the contribution of each parameter to prediction of the severity of the disease.

Results : On admission, the severely ill patients had significantly higher levels of LDH, IL-6, monokine induced by gamma interferon (MIG), D-dimer, fibrinogen, glucose than the patients with moderate disease. The levels of macrophage derived cytokine (MDC) were lower in severely ill patients. Based on artificial intelligence analysis, eight parameters (creatinine, glucose, monocyte number, fibrinogen, MDC, MIG, C-reactive protein (CRP) and IL-6 have been identified that could predict with an accuracy of 83-87% whether the patient will develop severe disease.

Conclusion : This study identifies the prognostic factors and provides a methodology for making prediction for COVID-19 patients based on widely accepted biomarkers that can be measured in most conventional clinical laboratories worldwide.

Krysko Olga, Kondakova Elena, Vershinina Olga, Galova Elena, Blagonravova Anna, Gorshkova Ekaterina, Bachert Claus, Ivanchenko Mikhail, Krysko Dmitri V, Vedunova Maria

2021

COVID-19, IL-6, artificial intelligence, macrophage derived cytokine, prediction models

Dermatology Dermatology

Identifying Silver Linings During the Pandemic Through Natural Language Processing.

In Frontiers in psychology ; h5-index 92.0

COVID-19 has presented an unprecedented challenge to human welfare. Indeed, we have witnessed people experiencing a rise of depression, acute stress disorder, and worsening levels of subclinical psychological distress. Finding ways to support individuals' mental health has been particularly difficult during this pandemic. An opportunity for intervention to protect individuals' health & well-being is to identify the existing sources of consolation and hope that have helped people persevere through the early days of the pandemic. In this paper, we identified positive aspects, or "silver linings," that people experienced during the COVID-19 crisis using computational natural language processing methods and qualitative thematic content analysis. These silver linings revealed sources of strength that included finding a sense of community, closeness, gratitude, and a belief that the pandemic may spur positive social change. People's abilities to engage in benefit-finding and leverage protective factors can be bolstered and reinforced by public health policy to improve society's resilience to the distress of this pandemic and potential future health crises.

Lossio-Ventura Juan Antonio, Lee Angela Yuson, Hancock Jeffrey T, Linos Natalia, Linos Eleni

2021

COVID-19, natural language processing, protective factors, sentiment analysis, silver linings, topic modeling

Surgery Surgery

Estimation of Subglottal Pressure, Vocal Fold Collision Pressure, and Intrinsic Laryngeal Muscle Activation From Neck-Surface Vibration Using a Neural Network Framework and a Voice Production Model.

In Frontiers in physiology

The ambulatory assessment of vocal function can be significantly enhanced by having access to physiologically based features that describe underlying pathophysiological mechanisms in individuals with voice disorders. This type of enhancement can improve methods for the prevention, diagnosis, and treatment of behaviorally based voice disorders. Unfortunately, the direct measurement of important vocal features such as subglottal pressure, vocal fold collision pressure, and laryngeal muscle activation is impractical in laboratory and ambulatory settings. In this study, we introduce a method to estimate these features during phonation from a neck-surface vibration signal through a framework that integrates a physiologically relevant model of voice production and machine learning tools. The signal from a neck-surface accelerometer is first processed using subglottal impedance-based inverse filtering to yield an estimate of the unsteady glottal airflow. Seven aerodynamic and acoustic features are extracted from the neck surface accelerometer and an optional microphone signal. A neural network architecture is selected to provide a mapping between the seven input features and subglottal pressure, vocal fold collision pressure, and cricothyroid and thyroarytenoid muscle activation. This non-linear mapping is trained solely with 13,000 Monte Carlo simulations of a voice production model that utilizes a symmetric triangular body-cover model of the vocal folds. The performance of the method was compared against laboratory data from synchronous recordings of oral airflow, intraoral pressure, microphone, and neck-surface vibration in 79 vocally healthy female participants uttering consecutive /pæ/ syllable strings at comfortable, loud, and soft levels. The mean absolute error and root-mean-square error for estimating the mean subglottal pressure were 191 Pa (1.95 cm H2O) and 243 Pa (2.48 cm H2O), respectively, which are comparable with previous studies but with the key advantage of not requiring subject-specific training and yielding more output measures. The validation of vocal fold collision pressure and laryngeal muscle activation was performed with synthetic values as reference. These initial results provide valuable insight for further vocal fold model refinement and constitute a proof of concept that the proposed machine learning method is a feasible option for providing physiologically relevant measures for laboratory and ambulatory assessment of vocal function.

Ibarra Emiro J, Parra Jesús A, Alzamendi Gabriel A, Cortés Juan P, Espinoza Víctor M, Mehta Daryush D, Hillman Robert E, Zañartu Matías

2021

ambulatory monitoring, clinical voice assessment, neck-surface accelerometer, neural networks, subglottal pressure estimation, voice production model

General General

Multi-Modal Pain Intensity Assessment Based on Physiological Signals: A Deep Learning Perspective.

In Frontiers in physiology

Traditional pain assessment approaches ranging from self-reporting methods, to observational scales, rely on the ability of an individual to accurately assess and successfully report observed or experienced pain episodes. Automatic pain assessment tools are therefore more than desirable in cases where this specific ability is negatively affected by various psycho-physiological dispositions, as well as distinct physical traits such as in the case of professional athletes, who usually have a higher pain tolerance as regular individuals. Hence, several approaches have been proposed during the past decades for the implementation of an autonomous and effective pain assessment system. These approaches range from more conventional supervised and semi-supervised learning techniques applied on a set of carefully hand-designed feature representations, to deep neural networks applied on preprocessed signals. Some of the most prominent advantages of deep neural networks are the ability to automatically learn relevant features, as well as the inherent adaptability of trained deep neural networks to related inference tasks. Yet, some significant drawbacks such as requiring large amounts of data to train deep models and over-fitting remain. Both of these problems are especially relevant in pain intensity assessment, where labeled data is scarce and generalization is of utmost importance. In the following work we address these shortcomings by introducing several novel multi-modal deep learning approaches (characterized by specific supervised, as well as self-supervised learning techniques) for the assessment of pain intensity based on measurable bio-physiological data. While the proposed supervised deep learning approach is able to attain state-of-the-art inference performances, our self-supervised approach is able to significantly improve the data efficiency of the proposed architecture by automatically generating physiological data and simultaneously performing a fine-tuning of the architecture, which has been previously trained on a significantly smaller amount of data.

Thiam Patrick, Hihn Heinke, Braun Daniel A, Kestler Hans A, Schwenker Friedhelm

2021

deep neural networks, information fusion, pain intensity assessment, physiological signals, signal processing

General General

Clinically-Driven Virtual Patient Cohorts Generation: An Application to Aorta.

In Frontiers in physiology

The combination of machine learning methods together with computational modeling and simulation of the cardiovascular system brings the possibility of obtaining very valuable information about new therapies or clinical devices through in-silico experiments. However, the application of machine learning methods demands access to large cohorts of patients. As an alternative to medical data acquisition and processing, which often requires some degree of manual intervention, the generation of virtual cohorts made of synthetic patients can be automated. However, the generation of a synthetic sample can still be computationally demanding to guarantee that it is clinically meaningful and that it reflects enough inter-patient variability. This paper addresses the problem of generating virtual patient cohorts of thoracic aorta geometries that can be used for in-silico trials. In particular, we focus on the problem of generating a cohort of patients that meet a particular clinical criterion, regardless the access to a reference sample of that phenotype. We formalize the problem of clinically-driven sampling and assess several sampling strategies with two goals, sampling efficiency, i.e., that the generated individuals actually belong to the target population, and that the statistical properties of the cohort can be controlled. Our results show that generative adversarial networks can produce reliable, clinically-driven cohorts of thoracic aortas with good efficiency. Moreover, non-linear predictors can serve as an efficient alternative to the sometimes expensive evaluation of anatomical or functional parameters of the organ of interest.

Romero Pau, Lozano Miguel, Martínez-Gil Francisco, Serra Dolors, Sebastián Rafael, Lamata Pablo, García-Fernández Ignacio

2021

clinically-driven sampling, digital twin, generative adversarial network, in-silico trials, support vector machine, synthetic population, thoracic-aorta, virtual cohort

General General

Research on a Segmentation Algorithm for the Tujia Brocade Images Based on Unsupervised Gaussian Mixture Clustering.

In Frontiers in neurorobotics

Tujia brocades are important carriers of Chinese Tujia national culture and art. It records the most detailed and real cultural history of Tujia nationality and is one of the National Intangible Cultural Heritage. Classic graphic elements are separated from Tujia brocade patterns to establish the Tujia brocade graphic element database, which is used for the protection and inheritance of traditional national culture. Tujia brocade dataset collected a total of more than 200 clear Tujia brocade patterns and was divided into seven categories, according to traditional meanings. The weave texture of a Tujia brocade is coarse, and the textural features of the background are obvious, so classical segmentation algorithms cannot achieve good segmentation effects. At the same time, deep learning technology cannot be used because there is no standard Tujia brocade dataset. Based on the above problems, this study proposes a method based on an unsupervised clustering algorithm for the segmentation of Tujia brocades. First, the cluster number K is calculated by fusing local binary patterns (LBP) and gray-level co-occurrence matrix (GLCM) characteristic values. Second, clustering and segmentation are conducted on each input Tujia brocade image by adopting a Gaussian mixture model (GMM) to obtain a preliminary segmentation image, wherein the image yielded after preliminary segmentation is rough. Then, a method based on voting optimization and dense conditional random field (DenseCRF) (CRF denotes conditional random filtering) is adopted to optimize the image after preliminary segmentation and obtain the image segmentation results. Finally, the desired graphic element contour is extracted through interactive cutting. The contributions of this study include: (1) a calculation method for the cluster number K wherein the experimental results show that the effect of the clustering number K chosen in this paper is ideal; (2) an optimization method for the noise points of Tujia brocade patterns based on voting, which can effectively eliminate isolated noise points from brocade patterns.

He Shuqi

2021

DenseCRF, GMM, K auto-selection based on information fusion, Tujia brocade segmentation, optimization based on the vote

General General

Medical Image Interpolation Using Recurrent Type-2 Fuzzy Neural Network.

In Frontiers in neuroinformatics

Image interpolation is an essential process for image processing and computer graphics in wide applications to medical imaging. For image interpolation used in medical diagnosis, the two-dimensional (2D) to three-dimensional (3D) transformation can significantly reduce human error, leading to better decisions. This research proposes the type-2 fuzzy neural networks method which is a hybrid of the fuzzy logic and neural networks as well as recurrent type-2 fuzzy neural networks (RT2FNNs) for advancing a novel 2D to 3D strategy. The ability of the proposed methods in the approximation of the function for image interpolation is investigated. The results report that both proposed methods are reliable for medical diagnosis. However, the RT2FNN model outperforms the type-2 fuzzy neural networks model. The average squares error for the recurrent network and the typical network reported 0.016 and 0.025, respectively. On the other hand, the number of fuzzy rules for the recurrent network and the typical network reported 16 and 22, respectively.

Tavoosi Jafar, Zhang Chunwei, Mohammadzadeh Ardashir, Mobayen Saleh, Mosavi Amir H

2021

2D to 3D, artificial intelligence, brain MRI, image interpolation, machine learning, recurrent neural network, type-2 fuzzy system

General General

Distinguishing Self, Other, and Autonomy From Visual Feedback: A Combined Correlation and Acceleration Transfer Analysis.

In Frontiers in human neuroscience ; h5-index 79.0

In cognitive science, Theory of Mind (ToM) is the mental faculty of assessing intentions and beliefs of others and requires, in part, to distinguish incoming sensorimotor (SM) signals and, accordingly, attribute these to either the self-model, the model of the other, or one pertaining to the external world, including inanimate objects. To gain an understanding of this mechanism, we perform a computational analysis of SM interactions in a dual-arm robotic setup. Our main contribution is that, under the common fate principle, a correlation analysis of the velocities of visual pivots is shown to be sufficient to characterize "the self" (including proximo-distal arm-joint dependencies) and to assess motor to sensory influences, and "the other" by computing clusters in the correlation dependency graph. A correlational analysis, however, is not sufficient to assess the non-symmetric/directed dependencies required to infer autonomy, the ability of entities to move by themselves. We subsequently validate 3 measures that can potentially quantify a metric for autonomy: Granger causality (GC), transfer entropy (TE), as well as a novel "Acceleration Transfer" (AT) measure, which is an instantaneous measure that computes the estimated instantaneous transfer of acceleration between visual features, from which one can compute a directed SM graph. Subsequently, autonomy is characterized by the sink nodes in this directed graph. This study results show that although TE can capture the directional dependencies, a rectified subtraction operation denoted, in this study, as AT is both sufficient and computationally cheaper.

Demirel Berkay, Moulin-Frier Clément, Arsiwalla Xerxes D, Verschure Paul F M J, Sánchez-Fibla Martí

2021

agency, attention, autonomy, cognitive development, computational cognition, developmental psychology, sensorimotor learning, theory of mind

General General

A roadmap towards predicting species interaction networks (across space and time).

In Philosophical transactions of the Royal Society of London. Series B, Biological sciences

Networks of species interactions underpin numerous ecosystem processes, but comprehensively sampling these interactions is difficult. Interactions intrinsically vary across space and time, and given the number of species that compose ecological communities, it can be tough to distinguish between a true negative (where two species never interact) from a false negative (where two species have not been observed interacting even though they actually do). Assessing the likelihood of interactions between species is an imperative for several fields of ecology. This means that to predict interactions between species-and to describe the structure, variation, and change of the ecological networks they form-we need to rely on modelling tools. Here, we provide a proof-of-concept, where we show how a simple neural network model makes accurate predictions about species interactions given limited data. We then assess the challenges and opportunities associated with improving interaction predictions, and provide a conceptual roadmap forward towards predictive models of ecological networks that is explicitly spatial and temporal. We conclude with a brief primer on the relevant methods and tools needed to start building these models, which we hope will guide this research programme forward. This article is part of the theme issue 'Infectious disease macroecology: parasite diversity and dynamics across the globe'.

Strydom Tanya, Catchen Michael D, Banville Francis, Caron Dominique, Dansereau Gabriel, Desjardins-Proulx Philippe, Forero-Muñoz Norma R, Higino Gracielle, Mercier Benjamin, Gonzalez Andrew, Gravel Dominique, Pollock Laura, Poisot Timothée

2021-Nov-08

biogeography, deep learning, ecological forecasting, ecological networks, machine learning

Public Health Public Health

Microplanning for designing vaccination campaigns in low-resource settings: A geospatial artificial intelligence-based framework.

In Vaccine ; h5-index 70.0

Existing campaign-based healthcare delivery programs used for immunization often fall short of established health coverage targets due to a lack of accurate estimates for population size and location. A microplan, an integrated set of detailed planning components, can be used to identify this information to support programs such as equitable vaccination efforts. Here, we presents a series of steps necessary to create an artificial intelligence-based framework for automated microplanning, and our pilot implementation of this analysis tool across 29 countries of the Americas. Further, we describe our processes for generating a conceptual framework, creating customized catchment areas, and estimating up-to-date populations to support microplanning for health campaigns. Through our application of the present framework, we found that 68 million individuals across the 29 countries are within 5 km of a health facility. The number of health facilities analyzed ranged from 2 in Peru to 789 in Argentina, while the total population within 5 km ranged from 1,233 in Peru to 15,304,439 in Mexico. Our results demonstrate the feasibility of using this methodological framework to support the development of customized microplans for health campaigns using open-source data in multiple countries. The pandemic is demanding an improved capacity to generate successful, efficient immunization campaigns; we believe that the steps described here can increase the automation of microplans in low resource settings.

Augusto Hernandes Rocha Thiago, Grapiuna de Almeida Dante, Shankar Kozhumam Arthi, Cristina da Silva Núbia, Bárbara Abreu Fonseca Thomaz Erika, Christine de Sousa Queiroz Rejane, de Andrade Luciano, Staton Catherine, Ricardo Nickenig Vissoci João

2021-Sep-15

COVID-19, Coronavirus, Health campaign, Microplan, Vaccination, Vaccine

Surgery Surgery

Discovery of temporal structure intricacy in arterial blood pressure waveforms representing acuity of liver transplant and forecasting short term surgical outcome via unsupervised manifold learning

ArXiv Preprint

Background: Arterial blood pressure (ABP) waveform evolves across each consecutive pulse during the liver transplant surgery. We hypothesized that the quantification of the waveform evolution reflects 1) the acuity of the recipient undergoing liver transplant and 2) the intraoperative dynamics that forecasts short-term surgical outcomes. Methods: In this prospective observational single cohort study on living donor liver transplant surgery, we extracted the waveform morphological evolution from the ABP data with the unsupervised manifold learning waveform analysis. Two quantitative indices, trend movement and fluctuation movement, were developed to represent the slow-varying and fast-varying dynamics respectively. We investigated the associations with the liver disease acuity represented with the Model for End-Stage Liver Disease (MELD) score and the primary outcomes, the early allograft failure (EAF), as well as the recently developed EAF scores, including the Liver Graft Assessment Following Transplantation (L-GrAFT) score, the Early Allograft Failure Simplified Estimation (EASE) score, and the Model for Early Allograft Function (MEAF) score. Results: Sixty recipients were enrolled. The presurgical trend movement was correlated with the MELD scores. It decreased in the anhepatic phase. The neohepatic trend movement correlated with the L-GrAFT scores, the EASE score, and the MEAF score. Regarding the constituent of the EAF scores, the trend movement most correlated with the postoperative day 7 bilirubin. Conclusions: The ABP waveform evolution intricacy in the presurgical phase reflects recipients' acuity condition while that in the neohepatic phase reveal the short-term surgical outcome calculated from laboratory data in postoperative day 7-10. The waveform evolution reflects the intraoperative contribution to the early outcome.

Shen-Chih Wang, Chien-Kun Ting, Cheng-Yen Chen, Chin-Su Liu, Niang-Cheng Lin, Che-Chuan Loon, Hau-Tieng Wu, Yu-Ting Lin

2021-09-21

General General

The Application of the Principles of Responsible AI on Social Media Marketing for Digital Health.

In Information systems frontiers : a journal of research and innovation

Social media enables medical professionals and authorities to share, disseminate, monitor, and manage health-related information digitally through online communities such as Twitter and Facebook. Simultaneously, artificial intelligence (AI) powered social media offers digital capabilities for organizations to select, screen, detect and predict problems with possible solutions through digital health data. Both the patients and healthcare professionals have benefited from such improvements. However, arising ethical concerns related to the use of AI raised by stakeholders need scrutiny which could help organizations obtain trust, minimize privacy invasion, and eventually facilitate the responsible success of AI-enabled social media operations. This paper examines the impact of responsible AI on businesses using insights from analysis of 25 in-depth interviews of health care professionals. The exploratory analysis conducted revealed that abiding by the responsible AI principles can allow healthcare businesses to better take advantage of the improved effectiveness of their social media marketing initiatives with their users. The analysis is further used to offer research propositions and conclusions, and the contributions and limitations of the study have been discussed.

Liu Rui, Gupta Suraksha, Patel Parth

2021-Sep-13

Consumer trust theory, Digital health, Information sharing theory, Responsible AI, Social media marketing, Technology acceptance model

General General

Automatic deep learning system for COVID-19 infection quantification in chest CT.

In Multimedia tools and applications

The paper proposes an automatic deep learning system for COVID-19 infection areas segmentation in chest CT scans. CT imaging proved its ability to detect the COVID-19 disease even for asymptotic patients, which make it a trustworthy alternative for PCR. Coronavirus disease spread globally and PCR screening is the adopted diagnostic testing method for COVID-19 detection. However, PCR is criticized due its low sensitivity ratios, also, it is time-consuming and manual complicated process. The proposed framework includes different steps; it starts to prepare the region of interest by segmenting the lung organ, which then undergoes edge enhancing diffusion filtering (EED) to improve the infection areas contrast and intensity homogeneity. The proposed FCN is implemented using U-net architecture with modified residual block to include concatenation skip connection. The block improves the learning of gradient values by forwarding the infection area features through the network. The proposed system is evaluated using different measures and achieved dice overlapping score of 0.961 and 0.780 for lung and infection areas segmentation, respectively. The proposed system is trained and tested using many 2D CT slices extracted from diverse datasets from different sources, which demonstrate the system generalization and effectiveness. The use of more datasets from different sources helps to enhance the system accuracy and generalization, which can be accomplished based on the data availability in in the future.

Alirr Omar Ibrahim

2021-Sep-13

COVID-19 infection, Chest CT, Deep learning, Segmentation

General General

COVID-19: Government subsidy models for sustainable energy supply with disruption risks.

In Renewable & sustainable energy reviews

The outbreak of the COVID-19 pandemic poses great challenges to the current government subsidy models in the renewable energy sector for recovering in the post-pandemic economy. Although, many subsidy models have been applied to accelerate renewable energy investment decisions. However, it is important to develop a new model to ensure the sustainability of the renewable energy supply network under disruptions on both the supply and demand sides due to hazardous events. This study investigates different subsidy models (renewable credit, supplier subsidy, and retailer subsidy) to find a win-win subsidy model for sustainable energy supply under disruption risks. The objective is to determine the optimal capacity of renewable energy added to the grid, the optimal wholesale price of the power plant, and the optimal retail price of the aggregator under different subsidy models to maximize the economic, social, and environmental benefits of the whole network. A novel scenario-based robust fuzzy optimization approach is proposed to capture the uncertainties of business-as-usual operations (e.g., some relevant costs and demand) and hazardous events (e.g., COVID-19 pandemic). The proposed model is tested in a case study of the Vietnamese energy market. The results show that for a high negative impact level of hazardous events on the supply side, the renewable credit and supplier subsidy models should be considered to recovery the renewable energy market. Further, the proposed approach has a better performance in improving the power plant's robust profit for most of the hazard scenarios than the robust optimization model.

Tsao Yu-Chung, Thanh Vo-Van, Chang Yi-Ying, Wei Hsi-Hsien

2021-Oct

COVID-19, Government subsidy, Hazardous scenario, Renewable energy, Robust fuzzy model, Sustainable supply

Dermatology Dermatology

A Deep Learning Model to Predict Knee Osteoarthritis Based on Nonimage Longitudinal Medical Record.

In Journal of multidisciplinary healthcare

Purpose : To develop deep learning model (Deep-KOA) that can predict the risk of knee osteoarthritis (KOA) within the next year by using the previous three years nonimage-based electronic medical record (EMR) data.

Patients and Methods : We randomly selected information of two million patients from the Taiwan National Health Insurance Research Database (NHIRD) from January 1, 1999 to December 31, 2013. During the study period, 132,594 patients were diagnosed with KOA, while 1,068,464 patients without KOA were chosen randomly as control. We constructed a feature matrix by using the three-year history of sequential diagnoses, drug prescriptions, age, and sex. Deep learning methods of convolutional neural network (CNN) and artificial neural network (ANN) were used together to develop a risk prediction model. We used the area under the receiver operating characteristic (AUROC), sensitivity, specificity, and precision to evaluate the performance of Deep-KOA. Then, we explored the important features using stepwise feature selection.

Results : This study included 132,594 KOA patients, 83,111 females (62.68%), 49,483 males (37.32%), mean age 64.2 years, and 1,068,464 non-KOA patients, 545,902 females (51.09%), 522,562 males (48.91%), mean age 51.00 years. The Deep-KOA achieved an overall AUROC, sensitivity, specificity, and precision of 0.97, 0.89, 0.93, and 0.80 respectively. The discriminative analysis of Deep-KOA showed important features from several diseases such as disorders of the eye and adnexa, acute respiratory infection, other metabolic and immunity disorders, and diseases of the musculoskeletal and connective tissue. Age and sex were not found as the most discriminative features, with AUROC of 0.9593 (-0.76% loss) and 0.9644 (-0.25% loss) respectively. Whereas medications including antacid, cough suppressant, and expectorants were identified as discriminative features.

Conclusion : Deep-KOA was developed to predict the risk of KOA within one year earlier, which may provide clues for clinical decision support systems to target patients with high risk of KOA to get precision prevention program.

Ningrum Dina Nur Anggraini, Kung Woon-Man, Tzeng I-Shiang, Yuan Sheng-Po, Wu Chieh-Chen, Huang Chu-Ya, Muhtar Muhammad Solihuddin, Nguyen Phung-Anh, Li Jack Yu-Chuan, Wang Yao-Chin

2021

artificial intelligence, clinical decision support system, medical informatics application, precision medicine

Pathology Pathology

Artificial intelligence application in diagnostic gastrointestinal endoscopy - Deus ex machina?

In World journal of gastroenterology ; h5-index 103.0

The close relationship of medicine with technology and the particular interest in this symbiosis in recent years has led to the development of several computed artificial intelligence (AI) systems aimed at various areas of medicine. A number of studies have demonstrated that those systems allow accurate diagnoses with histological precision, thus facilitating decision-making by clinicians in real time. In the field of gastroenterology, AI has been applied in the diagnosis of pathologies of the entire digestive tract and their attached glands, and are increasingly accepted for the detection of colorectal polyps and confirming their histological classification. Studies have shown high accuracy, sensitivity, and specificity in relation to expert endoscopists, and mainly in relation to those with less experience. Other applications that are increasingly studied and with very promising results are the investigation of dysplasia in patients with Barrett's esophagus and the endoscopic and histological assessment of colon inflammation in patients with ulcerative colitis. In some cases AI is thus better than or at least equal to human abilities. However, additional studies are needed to reinforce the existing data, and mainly to determine the applicability of this technology in other indications. This review summarizes the state of the art of AI in gastroenterological pathology.

Correia Fábio Pereira, Lourenço Luís Carvalho

2021-Aug-28

Artificial intelligence, Colorectal polyps, Computer-aided diagnosis, Deep learning, Dysplasia, gastrointestinal endoscopy

Radiology Radiology

Application of artificial intelligence in preoperative imaging of hepatocellular carcinoma: Current status and future perspectives.

In World journal of gastroenterology ; h5-index 103.0

Hepatocellular carcinoma (HCC) is the most common primary malignant liver tumor in China. Preoperative diagnosis of HCC is challenging because of atypical imaging manifestations and the diversity of focal liver lesions. Artificial intelligence (AI), such as machine learning (ML) and deep learning, has recently gained attention for its capability to reveal quantitative information on images. Currently, AI is used throughout the entire radiomics process and plays a critical role in multiple fields of medicine. This review summarizes the applications of AI in various aspects of preoperative imaging of HCC, including segmentation, differential diagnosis, prediction of histopathology, early detection of recurrence after curative treatment, and evaluation of treatment response. We also review the limitations of previous studies and discuss future directions for diagnostic imaging of HCC.

Feng Bing, Ma Xiao-Hong, Wang Shuang, Cai Wei, Liu Xia-Bi, Zhao Xin-Ming

2021-Aug-28

Artificial intelligence, Diagnosis, Hepatocellular carcinoma, Radiomics, Recurrence, Treatment

General General

Latent Class Analysis Reveals COVID-19-related ARDS Subgroups with Differential Responses to Corticosteroids.

In American journal of respiratory and critical care medicine ; h5-index 108.0

Rationale Two distinct subphenotypes have been identified in acute respiratory distress syndrome (ARDS), but the presence of subgroups in ARDS associated with COVID-19 is unknown. The objective of this study was to identify clinically relevant, novel subgroups in COVID-19-related ARDS, and compare them to previously described ARDS subphenotypes. Methods Eligible participants were adults with COVID-19 and ARDS at Columbia University Irving Medical Center. Latent class analysis (LCA) was used to identify subgroups with baseline clinical, respiratory, and laboratory data serving as partitioning variables. A previously-developed machine learning model was used to classify patients as the hypoinflammatory and hyperinflammatory subphenotypes. Baseline characteristics and clinical outcomes were compared between subgroups. Heterogeneity of treatment effect (HTE) for corticosteroid-use in subgroups was tested. Measurements and Main Results From 3/2-4/30/2020, 483 patients with COVID-19-related ARDS met study criteria. A two-class LCA model best fit the population (p=0.0075). Class 2 (23%) had higher pro-inflammatory markers, troponin, creatinine and lactate, lower bicarbonate and lower blood pressure than Class 1 (77%). 90-day mortality was higher in Class 2 versus Class 1 (75% vs 48%; p<0.0001). Considerable overlap was observed between these subgroups and ARDS subphenotypes. SARS-CoV-2 RT-PCR cycle threshold was associated with mortality in the hypoinflammatory but not the hyperinflammatory phenotype. HTE to corticosteroids was observed (p=0.0295), with improved mortality in the hyperinflammatory phenotype and worse mortality in the hypoinflammatory phenotype, with the caveat that corticosteroid treatment was not randomized. Conclusions We identified two COVID-19-related ARDS subgroups with differential outcomes, similar to previously described ARDS subphenotypes. SARS-CoV-2 PCR cycle threshold had differential value for predicting mortality in the subphenotypes. The subphenotypes had differential treatment responses to corticosteroids. This article is open access and distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives License 4.0 (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Sinha Pratik, Furfaro David, Cummings Matthew J, Abrams Darryl, Delucchi Kevin, Maddali Manoj V, He June, Thompson Alison, Murn Michael, Fountain John, Rosen Amanda, Robbins-Juarez Shelief Y, Adan Matthew A, Satish Tejus, Madhavan Mahesh, Gupta Aakriti, Lyashchenko Alexander K, Agerstrand Cara, Yip Natalie H, Burkart Kristin M, Beitler Jeremy R, Baldwin Matthew R, Calfee Carolyn S, Brodie Daniel, O’Donnell Max R

2021-Sep-20

ARDS, COVID-19, Latent class analysis, Phenotyping

Surgery Surgery

Radiomics and machine learning applications in rectal cancer: Current update and future perspectives.

In World journal of gastroenterology ; h5-index 103.0

The high incidence of rectal cancer in both sexes makes it one of the most common tumors, with significant morbidity and mortality rates. To define the best treatment option and optimize patient outcome, several rectal cancer biological variables must be evaluated. Currently, medical imaging plays a crucial role in the characterization of this disease, and it often requires a multimodal approach. Magnetic resonance imaging is the first-choice imaging modality for local staging and restaging and can be used to detect high-risk prognostic factors. Computed tomography is widely adopted for the detection of distant metastases. However, conventional imaging has recognized limitations, and many rectal cancer characteristics remain assessable only after surgery and histopathology evaluation. There is a growing interest in artificial intelligence applications in medicine, and imaging is by no means an exception. The introduction of radiomics, which allows the extraction of quantitative features that reflect tumor heterogeneity, allows the mining of data in medical images and paved the way for the identification of potential new imaging biomarkers. To manage such a huge amount of data, the use of machine learning algorithms has been proposed. Indeed, without prior explicit programming, they can be employed to build prediction models to support clinical decision making. In this review, current applications and future perspectives of artificial intelligence in medical imaging of rectal cancer are presented, with an imaging modality-based approach and a keen eye on unsolved issues. The results are promising, but the road ahead for translation in clinical practice is rather long.

Stanzione Arnaldo, Verde Francesco, Romeo Valeria, Boccadifuoco Francesca, Mainenti Pier Paolo, Maurea Simone

2021-Aug-28

Artificial intelligence, Deep learning, Machine learning, Radiogenomics, Radiomics, Rectal cancer

Public Health Public Health

Identification of high-risk COVID-19 patients using machine learning.

In PloS one ; h5-index 176.0

The current COVID-19 public health crisis, caused by SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2), has produced a devastating toll both in terms of human life loss and economic disruption. In this paper we present a machine-learning algorithm capable of identifying whether a given patient (actually infected or suspected to be infected) is more likely to survive than to die, or vice-versa. We train this algorithm with historical data, including medical history, demographic data, as well as COVID-19-related information. This is extracted from a database of confirmed and suspected COVID-19 infections in Mexico, constituting the official COVID-19 data compiled and made publicly available by the Mexican Federal Government. We demonstrate that the proposed method can detect high-risk patients with high accuracy, in each of four identified clinical stages, thus improving hospital capacity planning and timely treatment. Furthermore, we show that our method can be extended to provide optimal estimators for hypothesis-testing techniques commonly-used in biological and medical statistics. We believe that our work could be of use in the context of the current pandemic in assisting medical professionals with real-time assessments so as to determine health care priorities.

Quiroz-Juárez Mario A, Torres-Gómez Armando, Hoyo-Ulloa Irma, León-Montiel Roberto de J, U’Ren Alfred B

2021

General General

Intelligent fish farm-the future of aquaculture.

In Aquaculture international : journal of the European Aquaculture Society

With the continuous expansion of aquaculture scale and density, contemporary aquaculture methods have been forced to overproduce resulting in the accelerated imbalance rate of water environment, the frequent occurrence of fish diseases, and the decline of aquatic product quality. Moreover, due to the fact that the average age profile of agricultural workers in many parts of the world are on the higher side, fishery production will face the dilemma of shortage of labor, and aquaculture methods are in urgent need of change. Modern information technology has gradually penetrated into various fields of agriculture, and the concept of intelligent fish farm has also begun to take shape. The intelligent fish farm tries to deal with the precise work of increasing oxygen, optimizing feeding, reducing disease incidences, and accurately harvesting through the idea of "replacing human with machine," so as to liberate the manpower completely and realize the green and sustainable aquaculture. This paper reviews the application of fishery intelligent equipment, IoT, edge computing, 5G, and artificial intelligence algorithms in modern aquaculture, and analyzes the existing problems and future development prospects. Meanwhile, based on different business requirements, the design frameworks for key functional modules in the construction of intelligent fish farm are proposed.

Wang Cong, Li Zhen, Wang Tan, Xu Xianbao, Zhang Xiaoshuan, Li Daoliang

2021-Sep-13

Artificial intelligence, Intelligent equipment, Internet of Things, Machine vision, Unmanned boat

General General

Ethics, Integrity and Retributions of Digital Detection Surveillance Systems on Infectious Diseases: Systematic literature review.

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : The COVID-19 pandemic has raised the importance of the deployment of digital detection surveillance systems to support early warning and monitoring of infectious diseases (ID). These opportunities create a "double-edge sword" as the ethical governance of such approaches often lag behind technological achievements.

OBJECTIVE : The aim was to investigate ethical issues identified from utilizing AI-augmented surveillance or early warning systems to monitor and detect common or novel ID outbreaks.

METHODS : We searched relevant articles in a number of databases that addressed ethical issues of using artificial intelligence, digital surveillance systems, early warning systems and/or big data analytics technology for detecting, monitoring, or tracing ID according to PRISMA guidelines, and further identified and analysed them with a theoretical framework.

RESULTS : This systematic review identified 29 articles presented in six major themes clustered under individual, organizational and societal levels, including: awareness of implementing digital surveillance, digital integrity, trust, privacy and confidentiality, civil rights, and governance. Whilst these measures were understandable during a pandemic, the public were concerned about receiving inadequate information, unclear governance frameworks, and lack of privacy protection, data integrity and autonomy when utilizing ID digital surveillance. The barriers to engagement could widen existing healthcare disparities or digital divides by underrepresenting vulnerable and at-risk populations, and expose patients' highly sensitive data such as their movements and contacts to outside sources, impinging significantly upon basic human and civil rights.

CONCLUSIONS : Our findings inform ethical considerations for service delivery models for medical practitioners and policymakers implicated in the use of digital surveillance for ID spread and a basis for the global governance structure.

CLINICALTRIAL :

Zhao Yan Ivy, Ma Ye Xuan, Yu Man Wai Cecilia, Liu Jia, Dong Wei Nan, Pang Qin, Lu Xiao Qin, Molassiotis Alex, Holroyd Eleanor, Wong Chi Wai William

2021-Sep-14

General General

Examining embedded apparatuses of AI in Facebook and TikTok.

In AI & society

In popular discussions, the nuances of AI are often abridged as "the algorithm", as the specific arrangements of machine learning (ML), deep learning (DL) and automated decision-making on social media platforms are typically shrouded in proprietary secrecy punctuated by press releases and transparency initiatives. What is clear, however, is that AI embedded on social media functions to recommend content, personalize ads, aggregate news stories, and moderate problematic material. It is also increasingly apparent that individuals are concerned with the uses, implications, and fairness of algorithmic systems. Perhaps in response to concerns about "the algorithm" by individuals and governments, social media platforms utilize transparency initiatives and official statements, in part, to deflect official regulation. In the following paper, I draw from transparency initiatives and statements from representatives of Facebook and TikTok as case studies of how AI is embedded in these platforms, with attention to the promotion of AI content moderation as a solution to the circulation of problematic material and misinformation. This examination considers the complexity of embedded AI as a material-discursive apparatus, predicated on discursive techniques-what is seeable, sayable, knowable in a given time period-as well as the material arrangements-algorithms, datasets, users, platforms, infrastructures, moderators, etc. As such, the use of AI as part of the immensely popular platforms Facebook and TikTok demonstrates that AI does not exist in isolation, instead functioning as human-machine ensemble reliant on strategies of acceptance via discursive techniques and the changing material arrangements of everyday embeddedness.

Grandinetti Justin

2021-Sep-12

AI, Algorithms, Material-discursive, Platforms, Transparency

Ophthalmology Ophthalmology

[Attitude of patients to possible telemedicine in ophthalmology : Survey by questionnaire in patients with glaucoma].

In Der Ophthalmologe : Zeitschrift der Deutschen Ophthalmologischen Gesellschaft

BACKGROUND : The COVID-19 pandemic in 2020 and 2021 severely restricted the care of ophthalmology patients. Teleophthalmological services, such as video consultation or medical telephone advice could, at least partially, compensate for the lack of necessary controls in the case of chronic diseases; however, teleophthalmological options are currently still significantly underrepresented in Germany.

OBJECTIVE : In order to determine the willingness of patients to use telemedicine and the virtual clinic, we conducted a survey using a questionnaire on the subject of teleophthalmology in university medicine patients with known glaucoma as a chronic disease during the first wave of the COVID-19 pandemic.

METHODS : A total of100 patients were interviewed. The questionnaire contained 22 questions with multiple choice possible answers. The inclusion criterion was the presence of glaucoma as a chronic disease, age over 18 years, and sufficient linguistic understanding to answer the questions. The data were collected, analyzed and anonymously evaluated.

RESULTS : In the patient survey it could be shown that the respondents with glaucoma are very willing to do teleophthalmology and that this would be utilized. Of the patients surveyed 74.0% would accept telemedicine and virtual clinics. Of the ophthalmological patients surveyed 54.0% stated that their visit to the doctor/clinic could not take place due to SARS-CoV‑2 and 17.0% of the patients stated that the SARS-CoV‑2 pandemic had changed their opinion of telemedicine.

DISCUSSION : The acceptance of telemedicine in patients with chronic open-angle glaucoma seems surprisingly high. This has been increased even further by the SARS-CoV‑2 pandemic. These results reflect a general willingness of patients with chronic eye disease but do not reflect the applicability and acceptance and applicability from a medical point of view; however, this form of virtual consultation is accepted by the majority of patients with glaucoma and could be considered for certain clinical pictures.

Zwingelberg Sarah B, Mercieca Karl, Elksne Eva, Scheffler Stephanie, Prokosch Verena

2021-Sep-20

Artificial intelligence, Glaucoma, Ophthalmology, SARS-CoV‑2, Telemedicine

General General

MFBCNNC: Momentum factor biogeography convolutional neural network for COVID-19 detection via chest X-ray images.

In Knowledge-based systems

Aim : By October 6, 2020, Coronavirus disease 2019 (COVID-19) was diagnosed worldwide, reaching 3,355,7427 people and 1,037,862 deaths. Detection of COVID-19 and pneumonia by the chest X-ray images is of great significance to control the development of the epidemic situation. The current COVID-19 and pneumonia detection system may suffer from two shortcomings: the selection of hyperparameters in the models is not appropriate, and the generalization ability of the model is poor.

Method : To solve the above problems, our team proposed an improved intelligent global optimization algorithm, which is based on the biogeography-based optimization to automatically optimize the hyperparameters value of the models according to different detection objectives. In the optimization progress, after selecting the immigration of suitable index vector and the emigration of suitable index vector, we proposed adding a comparison operation to compare the value of them. According to the different numerical relationships between them, the corresponding operations are performed to improve the migration operation of biogeography-based optimization. The improved algorithm (momentum factor biogeography-based optimization) can better perform the automatic optimization operation. In addition, our team also proposed two frameworks: biogeography convolutional neural network and momentum factor biogeography convolutional neural network. And two methods for detection COVID-19 based on the proposed frameworks.

Results : Our method used three convolutional neural networks (LeNet-5, VGG-16, and ResNet-18) as the basic classification models for chest X-ray images detection of COVID-19, Normal, and Pneumonia. The accuracy of LeNet-5, VGG-16, and ResNet-18 is improved by 1.56%, 1.48%, and 0.73% after using biogeography-based optimization to optimize the hyperparameters of the models. The accuracy of LeNet-5, VGG-16, and ResNet-18 is improved by 2.87%, 6.31%, and 1.46% after using the momentum factor biogeography-based optimization to optimize the hyperparameters of the models.

Conclusion : Under the same experimental conditions, the performance of the momentum factor biogeography-based optimization is superior to the biogeography-based optimization in optimizing the hyperparameters of the convolutional neural networks. Experimental results show that the momentum factor biogeography-based optimization can improve the detection performance of the state-of-the-art approaches in terms of overall accuracy. In future research, we will continue to use and improve other global optimization algorithms to enhance the application ability of deep learning in medical pathological image detection.

Sun Junding, Li Xiang, Tang Chaosheng, Wang Shui-Hua, Zhang Yu-Dong

2021-Sep-15

Biogeography-based optimization, COVID-19, Convolutional neural network, Deep learning, Pneumonia

General General

Using Different Machine Learning Models to Classify Patients with Mild and Severe Cases of COVID-19 Based on Multivariate Blood Testing.

In Journal of medical virology

BACKGROUND : COVID19 is a serious respiratory disease. The ever-increasing number of cases is causing heavier loads on the health service system.

METHOD : Using 38 blood test indicators on the first day of admission for the 422 patients diagnosed with COVID-19 (from January 2020 to June 2021) to construct different machine learning models to classify patients with either mild or severe cases of COVID-19.

RESULTS : All models show good performance in the classification between COVID-19 patients with mild and severe disease. The AUC of the random forest model is 0.89, the AUC of the naive Bayes model is 0.90, the AUC of the support vector machine model is 0.86, and the AUC of the KNN model is 0.78, the AUC of the Logistic regression model is 0.84, and the AUC of the artificial neural network model is 0.87, among which the Naive Bayes model has the best performance.

CONCLUSION : Different machine learning models can classify patients with mild and severe cases based on 38 blood test indicators taken on the first day of admission for patients diagnosed with COVID-19. This article is protected by copyright. All rights reserved.

Zhang Rui-Kun, Xiao Qi, Zhu Sheng-Lang, Lin Hai-Yan, Tang Ming

2021-Sep-20

Artificial intelligence < Biostatistics & Bioinformatics, Coronavirus < Virus classification, Infection

Radiology Radiology

Diffusion and quantification of diffusion of prostate cancer.

In The British journal of radiology

For assessing a cancer treatment, and for detecting and characterizing cancer, Diffusion-weighted imaging (DWI) is commonly used. The key in DWI's use extracranially has been due to the emergence of of high-gradient amplitude and multichannel coils, parallelimaging, and echo-planar imaging. The benefit has been fewer motion artefacts and high-quality prostate images.Recently, new techniques have been developed to improve the signal-to-noise ratio of DWI with fewer artefacts, allowing an increase in spatial resolution. For apparent diffusion coefficient quantification, non-Gaussian diffusion models have been proposed as additional tools for prostate cancer detection and evaluation of its aggressiveness. More recently, radiomics and machine learning for prostate magnetic resonance imaging have emerged as novel techniques for the non-invasive characterisation of prostate cancer. This review presents recent developments in prostate DWI and discusses its potential use in clinical practice.

Ueno Yoshiko, Tamada Tsutomu, Sofue Keitaro, Murakami Takamichi

2021-Sep-19

Public Health Public Health

The future of zoonotic risk prediction.

In Philosophical transactions of the Royal Society of London. Series B, Biological sciences

In the light of the urgency raised by the COVID-19 pandemic, global investment in wildlife virology is likely to increase, and new surveillance programmes will identify hundreds of novel viruses that might someday pose a threat to humans. To support the extensive task of laboratory characterization, scientists may increasingly rely on data-driven rubrics or machine learning models that learn from known zoonoses to identify which animal pathogens could someday pose a threat to global health. We synthesize the findings of an interdisciplinary workshop on zoonotic risk technologies to answer the following questions. What are the prerequisites, in terms of open data, equity and interdisciplinary collaboration, to the development and application of those tools? What effect could the technology have on global health? Who would control that technology, who would have access to it and who would benefit from it? Would it improve pandemic prevention? Could it create new challenges? This article is part of the theme issue 'Infectious disease macroecology: parasite diversity and dynamics across the globe'.

Carlson Colin J, Farrell Maxwell J, Grange Zoe, Han Barbara A, Mollentze Nardus, Phelan Alexandra L, Rasmussen Angela L, Albery Gregory F, Bett Bernard, Brett-Major David M, Cohen Lily E, Dallas Tad, Eskew Evan A, Fagre Anna C, Forbes Kristian M, Gibb Rory, Halabi Sam, Hammer Charlotte C, Katz Rebecca, Kindrachuk Jason, Muylaert Renata L, Nutter Felicia B, Ogola Joseph, Olival Kevin J, Rourke Michelle, Ryan Sadie J, Ross Noam, Seifert Stephanie N, Sironen Tarja, Standley Claire J, Taylor Kishana, Venter Marietjie, Webala Paul W

2021-Nov-08

access and benefit sharing, epidemic risk, global health, machine learning, viral ecology, zoonotic risk

General General

IL13Pred: A method for predicting immunoregulatory cytokine IL-13 inducing peptides for managing COVID-19 severity.

bioRxiv Preprint

Interleukin 13 (IL-13) is an immunoregulatory cytokine that is primarily released by activated T-helper 2 cells. It induces the pathogenesis of many allergic diseases, such as airway hyperresponsiveness, glycoprotein hypersecretion and goblet cell hyperplasia. IL-13 also inhibits tumor immunosurveillance, which leads to carcinogenesis. In recent studies, elevated IL-13 serum levels have been shown in severe COVID-19 patients. Thus it is important to predict IL-13 inducing peptides or regions in a protein for designing safe protein therapeutics particularly immunotherapeutic. This paper describes a method developed for predicting, designing and scanning IL-13 inducing peptides. The dataset used in this study contain experimentally validated 313 IL-13 inducing peptides and 2908 non-inducing homo-sapiens peptides extracted from the immune epitope database (IEDB). We have extracted 95 key features using SVC-L1 technique from the originally generated 9165 features using Pfeature. Further, these key features were ranked based on their prediction ability, and top 10 features were used for building machine learning prediction models. In this study, we have deployed various machine learning techniques to develop models for predicting IL-13 inducing peptides. These models were trained, test and evaluated using five-fold cross-validation techniques; best model were evaluated on independent dataset. Our best model based on XGBoost achieves a maximum AUC of 0.83 and 0.80 on the training and independent dataset, respectively. Our analysis indicate that certain SARS-COV2 variants are more prone to induce IL-13 in COVID-19 patients. A standalone package as well as a web server named IL-13Pred has been developed for predicting IL-13 inducing peptides (https://webs.iiitd.edu.in/raghava/il13pred/).

Jain, S.; Dhall, A.; Patiyal, S.; Raghava, G. P. S.

2021-09-21

General General

Vaccine allocation policy optimization and budget sharing mechanism using Thompson sampling

ArXiv Preprint

The optimal allocation of vaccines to population subgroups over time is a challenging health care management problem. In the context of a pandemic, the interaction between vaccination policies adopted by multiple agents and the cooperation (or lack thereof) creates a complex environment that affects the global transmission dynamics of the disease. In this study, we take the perspective of decision-making agents that aim to minimize the size of their susceptible populations and must allocate vaccine under limited supply. We assume that vaccine efficiency rates are unknown to agents and we propose an optimization policy based on Thompson sampling to learn mean vaccine efficiency rates over time. Furthermore, we develop a budget-balanced resource sharing mechanism to promote cooperation among agents. We apply the proposed framework to the COVID-19 pandemic. We use a raster model of the world where agents represent the main countries worldwide and interact in a global mobility network to generate multiple problem instances. Our numerical results show that the proposed vaccine allocation policy achieves a larger reduction in the number of susceptible individuals, infections and deaths globally compared to a population-based policy. In addition, we show that, under a fixed global vaccine allocation budget, most countries can reduce their national number of infections and deaths by sharing their budget with countries with which they have a relatively high mobility exchange. The proposed framework can be used to improve policy-making in health care management by national and global health authorities.

David Rey, Ahmed W Hammad, Meead Saberi

2021-09-21

General General

Internet of things-enabled real-time health monitoring system using deep learning.

In Neural computing & applications

Smart healthcare monitoring systems are proliferating due to the Internet of Things (IoT)-enabled portable medical devices. The IoT and deep learning in the healthcare sector prevent diseases by evolving healthcare from face-to-face consultation to telemedicine. To protect athletes' life from life-threatening severe conditions and injuries in training and competitions, real-time monitoring of physiological indicators is critical. In this research work, we present a deep learning-based IoT-enabled real-time health monitoring system. The proposed system uses wearable medical devices to measure vital signs and apply various deep learning algorithms to extract valuable information. For this purpose, we have taken Sanda athletes as our case study. The deep learning algorithms help physicians properly analyze these athletes' conditions and offer the proper medications to them, even if the doctors are away. The performance of the proposed system is extensively evaluated using a cross-validation test by considering various statistical-based performance measurement metrics. The proposed system is considered an effective tool that diagnoses dreadful diseases among the athletes, such as brain tumors, heart disease, cancer, etc. The performance results of the proposed system are evaluated in terms of precision, recall, AUC, and F1, respectively.

Wu Xingdong, Liu Chao, Wang Lijun, Bilal Muhammad

2021-Sep-15

Deep learning, Deep neural network, Diseases, Healthcare system, Internet of things

General General

Supporting Remote Survey Data Analysis by Co-researchers with Learning Disabilities through Inclusive and Creative Practices and Data Science Approaches.

In DIS. Designing Interactive Systems (Conference)

Through a process of robust co-design, we created a bespoke accessible survey platform to explore the role of co-researchers with learning disabilities (LDs) in research design and analysis. A team of co-researchers used this system to create an online survey to challenge public understanding of LDs [3]. Here, we describe and evaluate the process of remotely co-analyzing the survey data across 30 meetings in a research team consisting of academics and non-academics with diverse abilities amid new COVID-19 lockdown challenges. Based on survey data with >1,500 responses, we first co-analyzed demographics using graphs and art & design approaches. Next, co-researchers co-analyzed the output of machine learning-based structural topic modelling (STM) applied to open-ended text responses. We derived an efficient five-steps STM co-analysis process for creative, inclusive, and critical engagement of data by co-researchers. Co-researchers observed that by trying to understand and impact public opinion, their own perspectives also changed.

Chapko Dorota, Rothstein Pedro, Emeh Lizzie, Frumiento Pino, Kennedy Donald, Mcnicholas David, Orjiekwe Ifeoma, Overton Michaela, Snead Mark, Steward Robyn, Sutton Jenny, Bradshaw Melissa, Jeffreys Evie, Gallia Will, Ewans Sarah, Williams Mark, Grierson Mick

2021-Jun

Human-centered computing → Human computer interaction (HCI), Learning disability, co-design, survey, topic model

Surgery Surgery

Adaptive kernel selection network with attention constraint for surgical instrument classification.

In Neural computing & applications

Computer vision (CV) technologies are assisting the health care industry in many respects, i.e., disease diagnosis. However, as a pivotal procedure before and after surgery, the inventory work of surgical instruments has not been researched with the CV-powered technologies. To reduce the risk and hazard of surgical tools' loss, we propose a study of systematic surgical instrument classification and introduce a novel attention-based deep neural network called SKA-ResNet which is mainly composed of: (a) A feature extractor with selective kernel attention module to automatically adjust the receptive fields of neurons and enhance the learnt expression and (b) A multi-scale regularizer with KL-divergence as the constraint to exploit the relationships between feature maps. Our method is easily trained end-to-end in only one stage with few additional calculation burdens. Moreover, to facilitate our study, we create a new surgical instrument dataset called SID19 (with 19 kinds of surgical tools consisting of 3800 images) for the first time. Experimental results show the superiority of SKA-ResNet for the classification of surgical tools on SID19 when compared with state-of-the-art models. The classification accuracy of our method reaches up to 97.703%, which is well supportive for the inventory and recognition study of surgical tools. Also, our method can achieve state-of-the-art performance on four challenging fine-grained visual classification datasets.

Hou Yaqing, Zhang Wenkai, Liu Qian, Ge Hongwei, Meng Jun, Zhang Qiang, Wei Xiaopeng

2021-Sep-13

Attention mechanism, Deep learning, Fine-grained classification, Health care

Internal Medicine Internal Medicine

Reliable and Interpretable Mortality Prediction With Strong Foresight in COVID-19 Patients: An International Study From China and Germany.

In Frontiers in artificial intelligence

Cohort-independent robust mortality prediction model in patients with COVID-19 infection is not yet established. To build up a reliable, interpretable mortality prediction model with strong foresight, we have performed an international, bi-institutional study from China (Wuhan cohort, collected from January to March) and Germany (Würzburg cohort, collected from March to September). A Random Forest-based machine learning approach was applied to 1,352 patients from the Wuhan cohort, generating a mortality prediction model based on their clinical features. The results showed that five clinical features at admission, including lymphocyte (%), neutrophil count, C-reactive protein, lactate dehydrogenase, and α-hydroxybutyrate dehydrogenase, could be used for mortality prediction of COVID-19 patients with more than 91% accuracy and 99% AUC. Additionally, the time-series analysis revealed that the predictive model based on these clinical features is very robust over time when patients are in the hospital, indicating the strong association of these five clinical features with the progression of treatment as well. Moreover, for different preexisting diseases, this model also demonstrated high predictive power. Finally, the mortality prediction model has been applied to the independent Würzburg cohort, resulting in high prediction accuracy (with above 90% accuracy and 85% AUC) as well, indicating the robustness of the model in different cohorts. In summary, this study has established the mortality prediction model that allowed early classification of COVID-19 patients, not only at admission but also along the treatment timeline, not only cohort-independent but also highly interpretable. This model represents a valuable tool for triaging and optimizing the resources in COVID-19 patients.

Bai Tao, Zhu Xue, Zhou Xiang, Grathwohl Denise, Yang Pengshuo, Zha Yuguo, Jin Yu, Chong Hui, Yu Qingyang, Isberner Nora, Wang Dongke, Zhang Lei, Kortüm K Martin, Song Jun, Rasche Leo, Einsele Hermann, Ning Kang, Hou Xiaohua

2021

COVID-19, Wuhan cohort, Würzburg cohort, foresight, interpretability, mortality prediction model, reliability

General General

Adoption of Improved Neural Network Blade Pattern Recognition in Prevention and Control of Corona Virus Disease-19 Pandemic.

In Pattern recognition letters

To explore the adoption effect of improved neural network blade pattern in corona virus disease (COVID)-19, comparative analysis is implemented. First, the following hypotheses are proposed. I: in addition to the confirmed cases and deaths, people suspected of being infected are also involved in the spread of the epidemic. II: patients who have been cured may also develop secondary infections, so it is considered that there is still a link between cured cases and the spread of the epidemic. III: only the relevant data of the previous day is used to predict the epidemic prevention and control of the next day. Then, the epidemic data from February 1st to February 15th in X province were selected as the control. The combined neural network model is used for prevention and control prediction, and the prediction results of the traditional neural network model are compared. The results show that the predictions of the daily new cases by the five neural network models have little difference with the actual value, and the trend is basically consistent. However, there are still differences in some time nodes. The errors of neural network 1 on the 6th and network 3 on the 13th are large. The accuracy of the combined neural network prediction model is high, and there is little difference between the result and the actual value at each time node. The prediction of the cumulative number of diagnoses per day of the five neural network models is also analyzed, and the results are relatively ideal. In addition, the accuracy of the combined neural network prediction model is high, and the difference between the result and the actual value at each time node is relatively small. It is found that the standard deviations of neural networks 2 and 3 are relatively high through the comparison of the deviations. The deviation means of the five models were all relatively low, and the mean deviation and standard deviation of the combined neural network model are the lowest. It is found that the accuracy of prediction on the epidemic spread in this province is good by comparing the performance of each neural network model. Regarding various indicators, the prediction accuracy of the combined neural network model is higher than that of the other four models, and its performance is also the best. Finally, the MSE of the improved neural network model is lower compared with the traditional neural network model. Moreover, with the change of learning times, the change trend of MSE is constant (P <0.05 for all). In short, the improved neural network blade model has better performance compared with that of the traditional neural network blade model. The prediction results of the epidemic situation are accurate, and the application effect is remarkable, so the proposed model is worthy of further promotion and application in the medical field.

Ma Yanli, Li Zhonghua, Gou Jixiang, Ding Lihua, Yang Dong, Feng Guiliang

2021-Sep-15

artificial intelligence, corona virus disease (COVID)-19, improved neural network blade model, neural network model

General General

MirrorME: implementation of an IoT based smart mirror through facial recognition and personalized information recommendation algorithm.

In International journal of information technology : an official journal of Bharati Vidyapeeth's Institute of Computer Applications and Management

We are living in the era of the fourth industrial revolution, which also treated as 4IR or Industry 4.0. Generally, 4IR considered as the mixture of robotics, artificial intelligence (AI), quantum computing, the Internet of Things (IoT) and other frontier technologies. It is obvious that nowadays a plethora of smart devices is providing services to make the daily life of humans easier. However, in the morning most people around the globe use a traditional mirror while preparing themselves for daily tasks. The aim is to build a low-cost intelligent mirror system that can display a variety of details based on user recommendations. Therefore, in this article, Internet of Things (IoT) and AI-based smart mirror is introduced that will support the users to receive the necessary daily update of weather information, date, time, calendar, to-do list, updated news headlines, traffic updates, COVID-19 cases status and so on. Moreover, a face detection method also implemented with the smart mirror to construct the architecture more secure. Our proposed MirrorME application provides a success rate of nearly 87% in interacting with the features of face recognition and voice input. The mirror is capable of delivering multimedia facilities while maintaining high levels of security within the device.

Uddin Khandaker Mohammad Mohi, Dey Samrat Kumar, Parvez Gias Uddin, Mukta Ayesha Siddika, Acharjee Uzzal Kumar

2021-Sep-12

4IR, Artificial intelligence, Authentication, Face detection, IoT, Smart mirror

General General

Software system to predict the infection in COVID-19 patients using deep learning and web of things.

In Software: practice & experience

Since the end of 2019, computed tomography (CT) images have been used as an important substitute for the time-consuming Reverse Transcriptase polymerase chain reaction (RT-PCR) test; a new coronavirus 2019 (COVID-19) disease has been detected and has quickly spread through many countries across the world. Medical imaging such as computed tomography provides great potential due to growing skepticism toward the sensitivity of RT-PCR as a screening tool. For this purpose, automated image segmentation is highly desired for a clinical decision aid and disease monitoring. However, there is limited publicly accessible COVID-19 image knowledge, leading to the overfitting of conventional approaches. To address this issue, the present paper focuses on data augmentation techniques to create synthetic data. Further, a framework has been proposed using WoT and traditional U-Net with EfficientNet B0 to segment the COVID Radiopedia and Medseg datasets automatically. The framework achieves an F-score of 0.96, which is best among state-of-the-art methods. The performance of the proposed framework also computed using Sensitivity, Specificity, and Dice-coefficient, achieves 84.5%, 93.9%, and 65.0%, respectively. Finally, the proposed work is validated using three quality of service (QoS) parameters such as server latency, response time, and network latency which improves the performance by 8%, 7%, and 10%, respectively.

Singh Ashima, Kaur Amrita, Dhillon Arwinder, Ahuja Sahil, Vohra Harpreet

2021-Jun-24

COVID‐19, EfficientNet B0, SARSCoV‐2, U‐net, WoT, deep learning, segmentation

General General

Experimental Technologies in the Diagnosis and Treatment of COVID-19 in Patients with Comorbidities.

In Journal of healthcare informatics research

The COVID-19 pandemic has impacted the whole world and raised concerns about its effects on different human organ systems. Early detection of COVID-19 may significantly increase the rate of survival; thus, it is critical that the disease is detected early. Emerging technologies have been used to prevent, diagnose, and manage COVID-19 among the populace in the USA and globally. Numerous studies have revealed the growing implementation of novel engineered systems during the intervention at various points of the disease's pathogenesis, especially as it relates to comorbidities and complications related to cardiovascular and respiratory organ systems. In this review, we provide a succinct, but extensive, review of the pathogenesis of COVID-19, particularly as it relates to angiotensin-converting enzyme 2 (ACE2) as a viral entry point. This is followed by a comprehensive analysis of cardiovascular and respiratory comorbidities of COVID-19 and novel technologies that are used to diagnose and manage hospitalized patients. Continuous cardiorespiratory monitoring systems, novel machine learning algorithms for rapidly triaging patients, various imaging modalities, wearable immunosensors, hotspot tracking systems, and other emerging technologies are reviewed. COVID-19 effects on the immune system, associated inflammatory biomarkers, and innovative therapies are also assessed. Finally, with emphasis on the impact of wearable and non-wearable systems, this review highlights future technologies that could help diagnose, monitor, and mitigate disease progression. Technologies that account for an individual's health conditions, comorbidities, and even socioeconomic factors can drastically reduce the high mortality seen among many COVID-19 patients, primarily via disease prevention, early detection, and pertinent management.

Amin Md Shahnoor, Wozniak Marcin, Barbaric Lidija, Pickard Shanel, Yerrabelli Rahul S, Christensen Anton, Coiado Olivia C

2021-Sep-15

Cardiovascular, Immune system, Innovation, Non-wearables, Pulmonary, Wearables

Surgery Surgery

Eyelid Measurements: Smartphone-Based Artificial Intelligence-Assisted Prediction.

In JMIR mHealth and uHealth

BACKGROUND : Margin reflex distance 1(MRD1), margin reflex distance 2 (MRD2), and levator muscle function (LF) are crucial for ptosis evaluation and management. Manual measurements of MRD1, MRD2, and LF are time-consuming, subjective, and prone to human error. Smartphone-based artificial intelligence (AI) image processing is a potential solution to overcome these limitations.

OBJECTIVE : We proposed the first smartphone-based AI-assisted image processing algorithm for MRD1, MRD2, and LF measurements.

METHODS : This observational study included 822 eyes of 411 volunteers aged over 18 years from August 1, 2020, to April 30, 2021. Six orbital photographs (bilateral primary gaze, up-gaze, and down-gaze) were taken using a smartphone (iPhone 11 pro max). The gold standard measurements and normalized eye photographs were obtained from these orbital photographs and compiled using AI-assisted software to create MRD1, MRD2 and LF models.

RESULTS : The Pearson correlation coefficients between the gold standard measurements and the predicted values obtained with the MRD1 and MRD2 models were excellent (r = 0.91, and 0.88, respectively) and with the LF model were good (r = 0.73). The intraclass correlation coefficient results showed excellent agreement between the gold standard measurements and the values predicted by the MRD1and MRD2 models (0.90, and 0.84, respectively), and substantial agreement with the LF model (0.69). The mean absolute errors were 0.35 mm, 0.37 mm, and 1.06 mm for MRD1, MRD2, and LF models, respectively. The 95% limits of agreement were -0.94 to 0.94 mm for the MRD1 model; -0.92 to 1.03 mm for the MRD2 model; and -0.63 to 2.53 mm for the LF model.

CONCLUSIONS : In this study, we proposed the first smartphone-based AI-assisted image processing algorithm for eyelid measurements. MRD1, MRD2, and LF measures can be taken in a quick, objective, and convenient manner. Furthermore, by using a smartphone, the examiner can check these measurements anywhere and at any time, which facilitates data collection.

CLINICALTRIAL :

Chen Hung-Chang, Tzeng Shin-Shi, Hsiao Yen-Chang, Chen Ruei-Feng, Hung Erh-Chien, Lee Oscar K

2021-Sep-19

Public Health Public Health

Impact of COVID-19 on city-scale transportation and safety: An early experience from Detroit.

In Smart health (Amsterdam, Netherlands)

The COVID-19 pandemic brought unprecedented levels of disruption to the local and regional transportation networks throughout the United States, especially the Motor City---Detroit. That was mainly a result of swift restrictive measures such as statewide quarantine and lock-down orders to confine the spread of the virus and the rising number of COVID-19 confirmed cases and deaths. This work is driven by analyzing five types of real-world data sets from Detroit related to traffic volume, daily cases, weather, social distancing index, and crashes from January 2019 to June 2020. The primary goals of this work are: i) figuring out the impacts of COVID-19 on the transportation network usage (traffic volume) and safety (crashes) for the City of Detroit, ii) determining whether each type of data (e.g. traffic volume data) could be a useful factor in the confirmed-cases prediction, and iii) providing an early future prediction method for COVID-19 rates, which can be a vital contributor to life-saving advanced preventative and preparatory responses. In addressing these problems, the prediction results of six feature groups are presented and analyzed to quantify the prediction effectiveness of each type of data. Then, a deep learning model was developed using long short-term memory networks to predict the number of confirmed cases within the next week. The model demonstrated a promising prediction result with a coefficient of determination ( R 2 ) of up to approximately 0.91. Furthermore, six essential observations with supporting evidence are presented, which will be helpful for decision-makers to take specific measures that aid in preventing the spread of COVID-19 and protecting public health and safety. The proposed approaches could be applied, customized, adjusted, and replicated for analysis of the impact of COVID-19 on a transportation network and prediction of the anticipated COVID-19 cases using a similar data set obtained for other large cities in the USA or from around the world.

Yao Yongtao, Geara Tony G, Shi Weisong

2021-Nov

COVID-19, Daily cases Detroit, Data analysis, Prediction, Quarantine, Social distancing weather, Traffic volume crashes, Transportation networks

Radiology Radiology

A Radiomic-based Machine Learning Algorithm to Reliably Differentiate Benign Renal Masses from Renal Cell Carcinoma.

In European urology focus

BACKGROUND : A substantial proportion of patients undergo treatment for renal masses where active surveillance or observation may be more appropriate.

OBJECTIVE : To determine whether radiomic-based machine learning platforms can distinguish benign from malignant renal masses.

DESIGN, SETTING, AND PARTICIPANTS : A prospectively maintained single-institutional renal mass registry was queried to identify patients with a computed tomography-proven clinically localized renal mass who underwent partial or radical nephrectomy.

INTERVENTION : Radiomic analysis of preoperative scans was performed. Clinical and radiomic variables of importance were identified through decision tree analysis, which were incorporated into Random Forest and REAL Adaboost predictive models.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS : The primary outcome was the degree of congruity between the virtual diagnosis and final pathology. Subanalyses were performed for small renal masses and patients who had percutaneous renal mass biopsies as part of their workup. Receiver operating characteristic curves were used to evaluate each model's discriminatory function.

RESULTS AND LIMITATIONS : A total of 684 patients met the selection criteria. Of them, 76% had renal cell carcinoma; 57% had small renal masses, of which 73% were malignant. Predictive modeling differentiated benign pathology from malignant with an area under the curve (AUC) of 0.84 (95% confidence interval [CI] 0.79-0.9). In small renal masses, radiomic analysis yielded a discriminatory AUC of 0.77 (95% CI 0.69-0.85). When negative and nondiagnostic biopsies were supplemented with radiomic analysis, accuracy increased from 83.3% to 93.4%.

CONCLUSIONS : Radiomic-based predictive modeling may distinguish benign from malignant renal masses. Clinical factors did not substantially improve the diagnostic accuracy of predictive models. Enhanced diagnostic predictability may improve patient selection before surgery and increase the utilization of active surveillance protocols.

PATIENT SUMMARY : Not all kidney tumors are cancerous, and some can be watched. We evaluated a new method that uses radiographic features invisible to the naked eye to distinguish benign masses from true cancers and found that it can do so with acceptable accuracy.

Nassiri Nima, Maas Marissa, Cacciamani Giovanni, Varghese Bino, Hwang Darryl, Lei Xiaomeng, Aron Monish, Desai Mihir, Oberai Assad A, Cen Steven Y, Gill Inderbir S, Duddalwar Vinay A

2021-Sep-16

Artificial Intelligence, Benign renal mass, Radiomics, Renal cell carcinoma, Small renal mass

General General

Determination of COVID-19 Vaccine Hesitancy Among University Students.

In Cureus

Introduction With the sudden outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-COV-2), vaccines appear to be the most efficient measure in combating spread. However, vaccines are only effective if a community collectively uptakes vaccination. This approach is growing increasingly difficult with the emergence of 'Vaccine Hesitancy.' This paper aims to determine the association between university curricula and the degree of hesitancy for the COVID-19 vaccine. Methods The online questionnaire assessed demographic data, prior knowledge of vaccines, attitude towards COVID-19 vaccines using an adapted version of the WHO Strategic Advisory Group of Experts (SAGE) Working Group's Vaccine Hesitancy Survey (VHS) and factors likely to motivate vaccine uptake. By using binary scoring, the degree of hesitancy among students was determined. Exploratory Factor Analysis (EFA) on VHS revealed underlying causes of hesitancy. To analyze the dependence between hesitancy and curriculum, a chi-squared test was conducted. Results Medical students scored higher for prior knowledge of vaccines (M = 3.54) as opposed to non-medical students (M = 3.49). Medical students responded favorably to COVID-19 vaccines with only 1.37% showing hesitancy for all nine items of VHS, compared to 2.55% of non-medical students. EFA produced three subscales within the VHS: lack of confidence, risk factor concern, and misinformation. The lack of confidence factor accounted for 65% of the data obtained. The chi-square test solidified that vaccine hesitancy is dependent on curriculum. Conclusion The majority of non-medical students showed hesitancy towards obtaining COVID-19 vaccines compared to medical students who were more willing, largely owing to their knowledge and understanding of vaccines.

Sadaqat Waliya, Habib Shanzay, Tauseef Ambreen, Akhtar Sheharyar, Hayat Meryum, Shujaat Syeda A, Mahmood Amina

2021-Aug

covid-19, curriculum, hesitancy, university students, vaccine

Surgery Surgery

Multi-step validation of a deep learning-based system for the quantification of bowel preparation: a prospective, observational study.

In The Lancet. Digital health

BACKGROUND : Inadequate bowel preparation is associated with a decrease in adenoma detection rate (ADR). A deep learning-based bowel preparation assessment system based on the Boston bowel preparation scale (BBPS) has been previously established to calculate the automatic BBPS (e-BBPS) score (ranging 0-20). The aims of this study were to investigate whether there was a statistically inverse relationship between the e-BBPS score and the ADR, and to determine the threshold of e-BBPS score for adequate bowel preparation in colonoscopy screening.

METHODS : In this prospective, observational study, we trained and internally validated the e-BBPS system using retrospective colonoscopy images and videos from the Endoscopy Center of Wuhan University, annotated by endoscopists. We externally validated the system using colonoscopy images and videos from the First People's Hospital of Yichang and the Third Hospital of Wuhan. To prospectively validate the system, we recruited consecutive patients at Renmin Hospital of Wuhan University aged between 18 and 75 years undergoing colonoscopy. The exclusion criteria included: contraindication to colonoscopy, family polyposis syndrome, inflammatory bowel disease, history of surgery for colorectal or colorectal cancer, known or suspected bowel obstruction or perforation, patients who were pregnant or lactating, inability to receive caecal intubation, and lumen obstruction. We did colonoscopy procedures and collected withdrawal videos, which were reviewed and the e-BBPS system was applied to all colon segments. The primary outcome of this study was ADR, defined as the proportion of patients with one or more conventional adenomas detected during colonoscopy. We calculated the ADR of each e-BBPS score and did a correlation analysis using Spearman analysis.

FINDINGS : From May 11 to Aug 10, 2020, 616 patients underwent screening colonoscopies, which evaluated. There was a significant inverse correlation between the e-BBPS score and ADR (Spearman's rank -0·976, p<0·010). The ADR for the e-BBPS scores 1-8 was 28·57%, 28·68%, 26·79%, 19·19%, 17·57%, 17·07%, 14·81%, and 0%, respectively. According to the 25% ADR standard for screening colonoscopy, an e-BBPS score of 3 was set as a threshold to guarantee an ADR of more than 25%, and so high-quality endoscopy. Patients with scores of more than 3 had a significantly lower ADR than those with a score of 3 or less (ADR 15·93% vs 28·03%, p<0·001, 95% CI 0·28-0·66, odds ratio 0·43).

INTERPRETATION : The e-BBPS system has potential to provide a more objective and refined threshold for the quantification of adequate bowel preparation.

FUNDING : Project of Hubei Provincial Clinical Research Center for Digestive Disease Minimally Invasive Incision and Hubei Province Major Science and Technology Innovation Project.

Zhou Wei, Yao Liwen, Wu Huiling, Zheng Biqing, Hu Shan, Zhang Lihui, Li Xun, He Chunping, Wang Zhengqiang, Li Yanxia, Huang Chao, Guo Mingwen, Zhang Xiaoqing, Zhu Qingxi, Wu Lianlian, Deng Yunchao, Zhang Jun, Tan Wei, Li Chao, Zhang Chenxia, Gong Rongrong, Du Hongliu, Zhou Jie, Sharma Prateek, Yu Honggang

2021-Sep-16

General General

Accuracy of health-related information regarding COVID-19 on Twitter during a global pandemic.

In World medical & health policy

This study was performed to analyze the accuracy of health-related information on Twitter during the coronavirus disease 2019 (COVID-19) pandemic. Authors queried Twitter on three dates for information regarding COVID-19 and five terms (cure, emergency or emergency room, prevent or prevention, treat or treatments, vitamins or supplements) assessing the first 25 results with health-related information. Tweets were authoritative if written by governments, hospitals, or physicians. Two physicians assessed each tweet for accuracy. Metrics were compared between accurate and inaccurate tweets using χ 2 analysis and Mann-Whitney U. A total of 25.4% of tweets were inaccurate. Accurate tweets were more likely written by Twitter authenticated authors (49.8% vs. 20.9%, 28.9% difference, 95% confidence interval [CI]: 17.7-38.2) with accurate tweet authors having more followers (19,491 vs. 7346; 3446 difference, 95% CI: 234-14,054) versus inaccurate tweet authors. Likes, retweets, tweet length, botometer scores, writing grade level, and rank order did not differ between accurate and inaccurate tweets. We found 1/4 of health-related COVID-19 tweets inaccurate indicating that the public should not rely on COVID-19 health information written on Twitter. Ideally, improved government regulatory authority, public/private industry oversight, independent fact-checking, and artificial intelligence algorithms are needed to ensure inaccurate information on Twitter is removed.

Swetland Sarah B, Rothrock Ava N, Andris Halle, Davis Bennett, Nguyen Linh, Davis Phil, Rothrock Steven G

2021-Jul-29

COVID‐19, pandemic, social media

General General

A neural network predicting the amplitude of the N2pc in individual EEG datasets.

In Journal of neural engineering ; h5-index 52.0

OBJECTIVE : The N2pc is a small amplitude transient interhemispheric voltage asymmetry used in cognitive neuroscience to investigate subject's allocation of selective visuo-spatial attention. N2pc is typically estimated by coherent averaging of EEG sweeps but, in absence of explicit normative indications, the number of sweeps is often based on arbitrariness or personal experience. With the final aim of reducing duration and cost of experimental protocols, here we developed a new approach to reliably predict N2pc amplitude from a minimal EEG dataset.

APPROACH : First, features predictive of N2pc amplitude were identified in the time-frequency domain. Then, an artificial neural network (NN) was trained to predict N2pc mean amplitude at the individual level. By resorting to simulated data, accuracy of the NN was assessed by computing the mean squared error (MSE) and the amplitude discretization error (ADE) and compared to the standard time averaging (TA) technique. The NN was then tested against two real datasets consisting of 14 and 12 subjects, respectively.

MAIN RESULT : In simulated scenarios entailing different number of sweeps (between 10 and 100), the MSE obtained with the proposed method resulted, on average, 1/5 of that obtained with the TA technique. Implementation on real EEG datasets showed that N2pc amplitude could be reliably predicted with as few as 40 EEG sweeps per cell of the experimental design.

SIGNIFICANCE : The developed approach allows to reduce duration and cost of experiments involving the N2pc, for instance in studies investigating attention deficits in pathological subjects.

Marturano Francesca, Brigadoi Sabrina, Doro Mattia, Dell’Acqua Roberto, Sparacino Giovanni

2021-Sep-20

EEG/ERP, N2pc, artificial neural network, machine-learning, time-frequency analysis

Internal Medicine Internal Medicine

A Machine Learning Model for Evaluating Imported Disease Screening Strategies in Immigrant Populations.

In The American journal of tropical medicine and hygiene

Given the high prevalence of imported diseases in immigrant populations, it has postulated the need to establish screening programs that allow their early diagnosis and treatment. We present a mathematical model based on machine learning methodologies to contribute to the design of screening programs in this population. We conducted a retrospective cross-sectional screening program of imported diseases in all immigrant patients who attended the Tropical Medicine Unit between January 2009 and December 2016. We designed a mathematical model based on machine learning methodologies to establish the set of most discriminatory prognostic variables to predict the onset of the: HIV infection, malaria, chronic hepatitis B and C, schistosomiasis, and Chagas in immigrant population. We analyzed 759 patients. HIV was predicted with an accuracy of 84.9% and the number of screenings to detect the first HIV-infected person was 26, as in the case of Chagas disease (with a predictive accuracy of 92.9%). For the other diseases the averages were 12 screenings to detect the first case of chronic hepatitis B (85.4%), or schistosomiasis (86.9%), 23 for hepatitis C (85.6%) or malaria (93.3%), and eight for syphilis (79.4%) and strongyloidiasis (88.4%). The use of machine learning methodologies allowed the prediction of the expected disease burden and made it possible to pinpoint with greater precision those immigrants who are likely to benefit from screening programs, thus contributing effectively to their development and design.

Fernández-Martínez Juan L, Boga José A, de Andrés-Galiana Enrique, Casado Luis, Fernández Jonathan, Menéndez Candela, García-Pérez Alicia, Suarez Noelia Moran, Sela María Martinez, Vázquez Fernando, Rodríguez-Guardado Azucena

2021-Sep-20

General General

Can DXA image-based deep learning model predict the anisotropic elastic behavior of trabecular bone?

In Journal of the mechanical behavior of biomedical materials ; h5-index 50.0

3D image-based finite element (FE) and bone volume fraction (BV/TV)/fabric tensor modeling techniques are currently used to determine the apparent stiffness tensor of trabecular bone for assessing its anisotropic elastic behavior. Inspired by the recent success of deep learning (DL) techniques, we hypothesized that DL modeling techniques could be used to predict the apparent stiffness tensor of trabecular bone directly using dual-energy X-ray absorptiometry (DXA) images. To test the hypothesis, a convolutional neural network (CNN) model was trained and validated to predict the apparent stiffness tensor of trabecular bone cubes using their DXA images. Trabecular bone cubes obtained from human cadaver proximal femurs were used to obtain simulated DXA images as input, and the apparent stiffness tensor of the trabecular cubes determined by using micro-CT based FE simulations was used as output (ground truth) to train the DL model. The prediction accuracy of the DL model was evaluated by comparing it with the micro-CT based FE models, histomorphometric parameter based multiple linear regression models, and BV/TV/fabric tensor based multiple linear regression models. The results showed that DXA image-based DL model achieved high fidelity in predicting the apparent stiffness tensor of trabecular bone cubes (R2 = 0.905-0.973), comparable to or better than the histomorphometric parameter based multiple linear regression and BV/TV/fabric tensor based multiple linear regression models, thus supporting the hypothesis of this study. The outcome of this study could be used to help develop DXA image-based DL techniques for clinical assessment of bone fracture risk.

Xiao Pengwei, Haque Eakeen, Zhang Tinghe, Dong X Neil, Huang Yufei, Wang Xiaodu

2021-Sep-15

DXA, Deep learning, Fabric tensor, Histomorphometric parameters, Stiffness tensor, Trabecular bone

Surgery Surgery

Can TKA outcomes be predicted with computational simulation? Generation of a patient specific planning tool.

In The Knee ; h5-index 38.0

BACKGROUND : Computer simulations of knee movement allow Total Knee Arthroplasty (TKA) dynamic outcomes to be studied. This study aims to build a model predicting patient reported outcome from a simulation of post-operative TKA joint dynamics.

METHODS : Landmark localisation was performed on 239 segmented pre-operative computerized tomography (CT) scans to capture patient specific soft tissue attachments. The pre-operative bones and 3D implant files were registered to post-operative CT scans following TKA. Each post-operative knee was simulated undergoing a deep knee bend with assumed ligament balancing of the extension space. The kinematic results from this simulation were used in a Multivariate Adaptive Regression Spline algorithm, predicting attainment of a Patient Acceptable Symptom State (PASS) score in captured 12 month post-operative Knee Injury and Osteoarthritis Outcome Scores (KOOS). An independent series of 250 patients was evaluated by the predictive model to assess how the predictive model behaved in a pre-operative planning context.

RESULTS : The generated predictive algorithm, called the Dynamic Knee Score (DKS) contained features, in order of significance, related to tibio-femoral force, patello-femoral motion and tibio-femoral motion. Area Under the Curve for predicting attainment of the PASS KOOS Score was 0.64. The predictive model produced a bimodal spread of predictions, reflecting a tendency to either strongly prefer one alignment plan over another or be ambivalent.

CONCLUSION : A predictive algorithm relating patient reported outcome to the outputs of a computational simulation of a deep knee bend has been derived (the DKS). Simulation outcomes related to tibio-femoral balance had the highest correlation with patient reported outcome.

Twiggs Joshua, Miles Brad, Roe Justin, Fritsch Brett, Liu David, Parker David, Dickison David, Shimmin Andrew, BarBo Jonathan, McMahon Stephen, Solomon Michael, Boyle Richard, Walter Len

2021-Sep-17

Computational simulation, Joint dynamics, Kinematics, Machine learning, Outcome, PROMS, Total Knee Arthroplasty (TKA)

Public Health Public Health

Estimating monthly PM2.5 concentrations from satellite remote sensing data, meteorological variables, and land use data using ensemble statistical modeling and a random forest approach.

In Environmental pollution (Barking, Essex : 1987)

Fine particulate matter (PM2.5) is associated with various adverse health outcomes and poses serious concerns for public health. However, ground monitoring stations for PM2.5 measurements are mostly installed in population-dense or urban areas. Thus, satellite retrieved aerosol optical depth (AOD) data, which provide spatial and temporal surrogates of exposure, have become an important tool for PM2.5 estimates in a study area. In this study, we used AOD estimates of surface PM2.5 together with meteorological and land use variables to estimate monthly PM2.5 concentrations at a spatial resolution of 3 km2 over Taiwan Island from 2015 to 2019. An ensemble two-stage estimation procedure was proposed, with a generalized additive model (GAM) for temporal-trend removal in the first stage and a random forest model used to assess residual spatiotemporal variations in the second stage. We obtained a model-fitting R2 of 0.98 with a root mean square error (RMSE) of 1.40 μg/m3. The leave-one-out cross-validation (LOOCV) R2 with seasonal stratification was 0.82, and the RMSE was 3.85 μg/m3, whereas the R2 and RMSE obtained by using the pure random forest approach produced R2 and RMSE values of 0.74 and 4.60 μg/m3, respectively. The results indicated that the ensemble modeling approach had a higher predictive ability than the pure machine learning method and could provide reliable PM2.5 estimates over the entire island, which has complex terrain in terms of land use and topography.

Chen Chu-Chih, Wang Yin-Ru, Yeh Hung-Yi, Lin Tang-Huang, Huang Chun-Sheng, Wu Chang-Fu

2021-Sep-14

Aerosol optical depth, Generalized additive model, Inverse distance weighting, Land use regression, Leave-one-out cross-validation

General General

Effectiveness and safety of rivaroxaban versus warfarin among nonvalvular atrial fibrillation patients with obesity and diabetes.

In Journal of diabetes and its complications ; h5-index 41.0

AIMS : To compare clinical outcomes of rivaroxaban and warfarin in patients with nonvalvular atrial fibrillation (NVAF) and concurrent obesity and diabetes.

METHODS : Patients aged ≥18 years were identified from a healthcare claims database with the following criteria: newly initiating rivaroxaban or warfarin, ≥1 medical claim with a diagnosis of AF, obesity determined by validated machine learning algorithm, and ≥1 claim with a diagnosis of diabetes or for antidiabetic medication. Treatment cohorts were matched using propensity scores and were compared for stroke/systemic embolism (SE) and major bleeding using Cox proportional hazards models.

RESULTS : A total of 9999 matched pairs of NVAF patients with obesity and diabetes who initiated treatment with rivaroxaban or warfarin were included. The composite risk of stroke/SE was significantly lower in the rivaroxaban cohort compared with the warfarin cohort (HR 0.82; 95% CI 0.74-0.90). Risks of ischemic and hemorrhagic strokes were also significantly reduced with rivaroxaban versus warfarin, but not SE. Major bleeding risk was similar between treatment cohorts (HR 0.92; 95% CI 0.78-1.09).

CONCLUSIONS : In NVAF patients with comorbidities of obesity and diabetes, rivaroxaban was associated with lower risks of stroke/SE and similar risk of major bleeding versus warfarin.

Weir Matthew R, Chen Yen-Wen, He Jinghua, Bookhart Brahim, Campbell Alicia, Ashton Veronica

2021-Sep-04

Anticoagulation, Diabetes mellitus, Nonvalvular atrial fibrillation, Obesity, Rivaroxaban, Warfarin

General General

Data-driven operation of the resilient electric grid: A case of COVID-19.

In Journal of engineering (Stevenage, England)

Electrical energy is a vital part of modern life, and expectations for grid resilience to allow a continuous and reliable energy supply has tremendously increased even during adverse events (e.g. Ukraine cyberattack, Hurricane Maria). The global pandemic COVID-19 has raised the electric energy reliability risk due to potential workforce disruptions, supply chain interruptions, and increased possible cybersecurity threats. Additionally, the pandemic introduces a significant degree of uncertainty to the grid operation in the presence of other challenges including aging power grids, high proliferation of distributed generation, market mechanism, and active distribution network. This situation increases the need for measures for the resiliency of power grids to mitigate the impact of the pandemic as well as simultaneous extreme events including cyberattacks and adverse weather events. Solutions to manage such an adverse scenario will be multi-fold: (a) emergency planning and organisational support, (b) following safety protocol, (c) utilising enhanced automation and sensing for situational awareness, and (d) integration of advanced technologies and data points for ML-driven enhanced decision support. Enhanced digitalisation and automation resulted in better network visibility at various levels, including generation, transmission, and distribution. These data or information can be employed to take advantage of advanced machine learning techniques for automation and increased power grid resilience. In this paper, the resilience of power grids in the face of pandemics is explored and various machine learning tools that can be helpful to augment human operators are discused by: (a) reviewing the impact of COVID-19 on power grid operations and actions taken by operators/organisations to minimise the impact of COVID-19, and (b) presenting recently developed tools and concepts of machine learning and artificial intelligence that can be applied to increase the resiliency of power systems in normal and extreme scenarios such as the COVID-19 pandemic.

Noorazar H, Srivastava A, Pannala S, K Sadanandan Sajan

2021-Aug-09

General General

Tracing the origin of honey products based on metagenomics and machine learning.

In Food chemistry

The adulteration of honey is common. Recently, High Throughput Sequencing (HTS)-based metabarcoding method has been applied successfully to pollen/honey identification to determine floral composition that, in turn, can be used to identify the geographical origins of honeys. However, the lack of local references materials posed a serious challenge for HTS-based pollen identification methods. Here, we sampled 28 honey samples from various geographic origins without prior knowledge of local floral information and applied a machine learning method to determine geographical origins. The machine learning method uses a resilient backpropagation algorithm to train a neural network. The results showed that biological components in honey provided characteristic traits that enabled accurate geographic tracing for nearly all honey samples, confidently discriminating honeys to their geographic origin with >99% success rates, including those separated by as little as 39 km.

Liu Shanlin, Lang Dandan, Meng Guanliang, Hu Jiahui, Tang Min, Zhou Xin

2021-Sep-06

Floral composition, Genomics, Honey adulteration, Honeybee, Machine learning, Pollen, Resilient backpropagation

Surgery Surgery

Real-time medical phase recognition using long-term video understanding and progress gate method.

In Medical image analysis

We introduce a real-time system for recognizing five phases of the trauma resuscitation process, the initial management of injured patients in the emergency department. We used depth videos as input to preserve the privacy of the patients and providers. The depth videos were recorded using a Kinect-v2 mounted on the sidewall of the room. Our dataset consisted of 183 depth videos of trauma resuscitations. The model was trained on 150 cases with more than 30 minutes each and tested on the remaining 33 cases. We introduced a reduced long-term operation (RLO) method for extracting features from long segments of video and combined it with the regular model having short-term information only. The model with RLO outperformed the regular short-term model by 5% using the accuracy score. We also introduced a progress gate (PG) method to distinguish visually similar phases using video progress. The final system achieved 91% accuracy and significantly outperformed previous systems for phase recognition in this setting.

Zhang Yanyi, Marsic Ivan, Burd Randall S

2021-Sep-03

Deep learning, Phase recognition, Process gate, Reduced long-term operation, Trauma resuscitation, Video understanding

General General

A deep-learning approach for direct whole-heart mesh reconstruction.

In Medical image analysis

Automated construction of surface geometries of cardiac structures from volumetric medical images is important for a number of clinical applications. While deep-learning-based approaches have demonstrated promising reconstruction precision, these approaches have mostly focused on voxel-wise segmentation followed by surface reconstruction and post-processing techniques. However, such approaches suffer from a number of limitations including disconnected regions or incorrect surface topology due to erroneous segmentation and stair-case artifacts due to limited segmentation resolution. We propose a novel deep-learning-based approach that directly predicts whole heart surface meshes from volumetric CT and MR image data. Our approach leverages a graph convolutional neural network to predict deformation on mesh vertices from a pre-defined mesh template to reconstruct multiple anatomical structures in a 3D image volume. Our method demonstrated promising performance of generating whole heart reconstructions with as good or better accuracy than prior deep-learning-based methods on both CT and MR data. Furthermore, by deforming a template mesh, our method can generate whole heart geometries with better anatomical consistency and produce high-resolution geometries from lower resolution input image data. Our method was also able to produce temporally-consistent surface mesh predictions for heart motion from CT or MR cine sequences, and therefore can potentially be applied for efficiently constructing 4D whole heart dynamics. Our code and pre-trained networks are available at https://github.com/fkong7/MeshDeformNet.

Kong Fanwei, Wilson Nathan, Shadden Shawn

2021-Sep-08

Deep learning, Graph convolutional networks, Surface mesh reconstruction, Whole heart segmentation

Radiology Radiology

Decomposing normal and abnormal features of medical images for content-based image retrieval of glioma imaging.

In Medical image analysis

In medical imaging, the characteristics purely derived from a disease should reflect the extent to which abnormal findings deviate from the normal features. Indeed, physicians often need corresponding images without abnormal findings of interest or, conversely, images that contain similar abnormal findings regardless of normal anatomical context. This is called comparative diagnostic reading of medical images, which is essential for a correct diagnosis. To support comparative diagnostic reading, content-based image retrieval (CBIR) that can selectively utilize normal and abnormal features in medical images as two separable semantic components will be useful. In this study, we propose a neural network architecture to decompose the semantic components of medical images into two latent codes: normal anatomy code and abnormal anatomy code. The normal anatomy code represents counterfactual normal anatomies that should have existed if the sample is healthy, whereas the abnormal anatomy code attributes to abnormal changes that reflect deviation from the normal baseline. By calculating the similarity based on either normal or abnormal anatomy codes or the combination of the two codes, our algorithm can retrieve images according to the selected semantic component from a dataset consisting of brain magnetic resonance images of gliomas. Moreover, it can utilize a synthetic query vector combining normal and abnormal anatomy codes from two different query images. To evaluate whether the retrieved images are acquired according to the targeted semantic component, the overlap of the ground-truth labels is calculated as metrics of the semantic consistency. Our algorithm provides a flexible CBIR framework by handling the decomposed features with qualitatively and quantitatively remarkable results.

Kobayashi Kazuma, Hataya Ryuichiro, Kurose Yusuke, Miyake Mototaka, Takahashi Masamichi, Nakagawa Akiko, Harada Tatsuya, Hamamoto Ryuji

2021-Sep-08

comparative diagnostic reading, content-based image retrieval, deep learning, disentangled representation, feature decomposition

General General

In silico screening of ssDNA aptamer against Escherichia coli O157:H7: A machine learning and the Pseudo K-tuple nucleotide composition based approach.

In Computational biology and chemistry

This study was planned to in silico screening of ssDNA aptamer against Escherichia coli O157:H7 by combination of machine learning and the PseKNC approach. For this, firstly a total numbers of 47 validated ssDNA aptamers as well as 498 random DNA sequences were considered as positive and negative training data respectively. The sequences then converted to numerical vectors using PseKNC method through Pse-in-one 2.0 web server. After that, the numerical vectors were subjected to classification by the SVM, ANN and RF algorithms available in Orange 3.2.0 software. The performances of the tested models were evaluated using cross-validation, random sampling and ROC curve analyzes. The primary results demonstrated that the ANN and RF algorithms have appropriate performances for the data classification. To improve the performances of mentioned classifiers the positive training data was triplicated and re-training process was also performed. The results confirmed that data size improvement had significant effect on the accuracy of data classification especially about RF model. Subsequently, the RF algorithm with accuracy of 98% was selected for aptamer screening. The thermodynamics details of folding process as well as secondary structures of the screened aptamers were also considered as final evaluations. The results confirmed that the selected aptamers by the proposed method had appropriate structure properties and there is no thermodynamics limit for the aptamers folding.

Nosrati Mokhtar, Amani Jafar

2021-Aug-27

Escherichia coli O157:H7, Machine learning, PseKNC, SsDNA aptamer

General General

Integration of various technology-based approaches for enhancing the performance of microbial fuel cell technology: A review.

In Chemosphere

The conflict between climate change and growing global energy demand is an immense sustainability challenge that requires noteworthy scientific and technological developments. Recently the importance of microbial fuel cell (MFC) on this issue has seen profound investigation due to its inherent ability of simultaneous wastewater treatment, and power production. However, the challenges of economy-related manufacturing and operation costs should be lowered to achieve positive field-scale demonstration. Also, a variety of different field deployments will lead to improvisation. Hence, this review article discusses the possibility of integration of MFC technology with various technologies of recent times leading to advanced sustainable MFC technology. Technological innovation in the field of nanotechnology, genetic engineering, additive manufacturing, artificial intelligence, adaptive control, and few other hybrid systems integrated with MFCs is discussed. This comprehensive and state-of-the-art study elaborates hybrid MFCs integrated with various technology and its working principles, modified electrode material, complex and easy to manufacture reactor designs, and the effects of various operating parameters on system performances. Although integrated systems are promising, much future research work is needed to overcome the challenges and commercialize hybrid MFC technology.

Dwivedi Kavya Arun, Huang Song-Jeng, Wang Chin-Tsan

2021-Sep-13

Bio-electricity, Integration, Microbial fuel cell, Technical innovations, Waste water treatment

General General

PrimePatNet87: Prime pattern and tunable q-factor wavelet transform techniques for automated accurate EEG emotion recognition.

In Computers in biology and medicine

Nowadays, many deep models have been presented to recognize emotions using electroencephalogram (EEG) signals. These deep models are computationally intensive, it takes a longer time to train the model. Also, it is difficult to achieve high classification performance using for emotion classification using machine learning techniques. To overcome these limitations, we present a hand-crafted conventional EEG emotion classification network. In this work, we have used novel prime pattern and tunable q-factor wavelet transform (TQWT) techniques to develop an automated model to classify human emotions. Our proposed cognitive model comprises feature extraction, feature selection, and classification steps. We have used TQWT on the EEG signals to obtain the sub-bands. The prime pattern and statistical feature generator are employed on the generated sub-bands and original signal to generate 798 features. 399 (half of them) out of 798 features are selected using minimum redundancy maximum relevance (mRMR) selector, and misclassification rates of each signal are evaluated using support vector machine (SVM) classifier. The proposed network generated 87 feature vectors hence, this model is named PrimePatNet87. In the last step of the feature generation, the best 20 feature vectors which are selected based on the calculated misclassification rates, are concatenated. The generated feature vector is subjected to the feature selection and the most significant 1000 features are selected using the mRMR selector. These selected features are then classified using an SVM classifier. In the last phase, iterative majority voting has been used to generate a general result. We have used three publicly available datasets, namely DEAP, DREAMER, and GAMEEMO, to develop our proposed model. Our presented PrimePatNet87 model reached over 99% classification accuracy on whole datasets with leave one subject out (LOSO) validation. Our results demonstrate that the developed prime pattern network is accurate and ready for real-world applications.

Dogan Abdullah, Akay Merve, Barua Prabal Datta, Baygin Mehmet, Dogan Sengul, Tuncer Turker, Dogru Ali Hikmet, Acharya U Rajendra

2021-Sep-16

EEG signal Classification, Emotion recognition, Hand-crafted method, Prime pattern network, mRMR selector

General General

Penalty weighted glucose prediction models could lead to better clinically usage.

In Computers in biology and medicine

BACKGROUND AND OBJECTIVE : Numerous attempts to predict glucose value from continuous glucose monitors (CGM) have been published. However, there is a lack of proper analysis and modeling of penalty for errors in different glycemic ranges. The aim of this study was to investigate the potential for developing glucose prediction models with focus on the clinical aspects.

METHODS : We developed and compared six different models to test which approach were best suited for predicting glucose levels at different lead times between 10 and 60 min. The models were: last observation carried forward, linear extrapolation, ensemble methods using LSBoost and bagging, neural networks, one without error-weights and one with error-weights. The modeling and test were based on 225 type 1 diabetes patients with 315,000 h of CGM data.

RESULTS : Results show that it is possible to predict glucose levels based on CGM with a reasonable accuracy and precision with a 30-min prediction lead time. A comparison of different methods shows that there are improvements on performance gained from using more advanced machine learning algorithms (MARD 10.26-10.79 @ 30-min lead time) compared to a simple modeling (MARD 10.75-12.97 @ 30-min lead time). Moreover, the proposed use of error weights could lead to better clinical performance of these models, which is an important factor for real usage. E.g., the percentages in the C-zone of the consensus error grid without error-weights (0.57-0.68%) vs including error-weights (0.28%).

CONCLUSIONS : The results point toward that using error weighting in the training of the models could lead to better clinical performance.

Cichosz Simon Lebech, Kronborg Thomas, Jensen Morten Hasselstrøm, Hejlesen Ole

2021-Sep-15

CGM, Continuous glucose monitoring, Ensemble learning, Glucose, Neural network, Prediction, Type 1 diabetes

General General

A low voltage-powered soft electromechanical stimulation patch for haptics feedback in human-machine interfaces.

In Biosensors & bioelectronics

One grand challenge in haptic human-machine interface devices is to electromechanically stimulate sensations on the human skin wirelessly by thin and soft patches under a low driving voltage. Here, we propose a soft haptics-feedback system using highly charged, polymeric electret films with an annulus-shape bump structure to induce mechanical sensations on the fingertip of volunteers under an applied voltage range of 5-20 V. As an application demonstration, a 3 × 3 actuators array is used for transmitting patterned haptic information, such as letters of 'T', 'H', 'U' letters and numbers of '0', '1', '2'. Moreover, together with flexible lithium batteries and a flexible circuit board, an untethered stimulation patch is constructed for operations of 1 h. The analytical model, design principle, and performance characterizations can be applicable for the integration of other wearable electronics toward practical applications in the fields of AR (augmented reality), VR (virtual reality) and robotics.

Qiu Wenying, Zhong Junwen, Jiang Tao, Li Zhaoyang, Yao Mingze, Shao Zhichun, Cheng Qilong, Liang Jiaming, Wang Dongkai, Peng Yande, He Peisheng, Bogy David B, Zhang Min, Wang Xiaohao, Lin Liwei

2021-Sep-08

Haptics feedback, Human-machine interface, Soft actuator, Wearable electronics

General General

Recurrent neural network from adder's perspective: Carry-lookahead RNN.

In Neural networks : the official journal of the International Neural Network Society

The recurrent network architecture is a widely used model in sequence modeling, but its serial dependency hinders the computation parallelization, which makes the operation inefficient. The same problem was encountered in serial adder at the early stage of digital electronics. In this paper, we discuss the similarities between recurrent neural network (RNN) and serial adder. Inspired by carry-lookahead adder, we introduce carry-lookahead module to RNN, which makes it possible for RNN to run in parallel. Then, we design the method of parallel RNN computation, and finally Carry-lookahead RNN (CL-RNN) is proposed. CL-RNN takes advantages in parallelism and flexible receptive field. Through a comprehensive set of tests, we verify that CL-RNN can perform better than existing typical RNNs in sequence modeling tasks which are specially designed for RNNs. Code and models are available at: https://github.com/WinnieJiangHW/Carry-lookahead_RNN.

Jiang Haowei, Qin Feiwei, Cao Jin, Peng Yong, Shao Yanli

2021-Sep-06

Carry-lookahead, Deep learning, Parallel computation, Sequence modeling

Public Health Public Health

New machine learning scoring system for predicting postoperative mortality in gastroduodenal ulcer perforation: A study using a Japanese nationwide inpatient database.

In Surgery ; h5-index 54.0

BACKGROUND : Conventional prediction models for estimating risk of postoperative mortality in gastroduodenal ulcer perforation have suboptimal prediction ability. We aimed to develop and validate new machine learning models and an integer-based score for predicting the postoperative mortality.

METHODS : We retrospectively identified patients with gastroduodenal ulcer perforation who underwent surgical repair, using a nationwide Japanese inpatient database. In a derivation cohort from July 2010 to March 2016, we developed 2 machine learning-based models, Lasso and XGBoost, using 45 candidate predictors, and also developed an integer-based score for clinical use by including important variables in Lasso. In a validation cohort from April 2016 to March 2017, we measured the prediction performances of the models by computing area under the curve and comparing it to the conventional American Society of Anesthesiology risk score.

RESULTS : Of 25,886 patients, 1,176 (4.5%) died after surgical repair. For the validation cohort, Lasso and XGBoost had significantly higher prediction abilities than the American Society of Anesthesiology score (Lasso area under the curve = 0.84; 95% confidence interval 0.81-0.86; American Society of Anesthesiology score area under the curve = 0.70; 95% confidence interval 0.65-0.74, P < .001). The integer-based risk score, which had 13 factors, had a prediction ability similar to those of Lasso and XGBoost (area under the curve = 0.83; 95% confidence interval 0.81-0.86). According to the integer-based score, the mortalities were 0.1%, 2.3%, 9.3%, and 29.0% for the low (score, 0), moderate (1-2), high (3-4), and very high (≥5) score groups, respectively.

CONCLUSION : Machine learning models and the integer-based risk score performed well in predicting risk of postoperative mortality in gastroduodenal ulcer perforation. These models will help in decision making.

Konishi Takaaki, Goto Tadahiro, Fujiogi Michimasa, Michihata Nobuaki, Kumazawa Ryosuke, Matsui Hiroki, Fushimi Kiyohide, Tanabe Masahiko, Seto Yasuyuki, Yasunaga Hideo

2021-Sep-16

oncology Oncology

All around suboptimal health - a joint position paper of the Suboptimal Health Study Consortium and European Association for Predictive, Preventive and Personalised Medicine.

In The EPMA journal

First two decades of the twenty-first century are characterised by epidemics of non-communicable diseases such as many hundreds of millions of patients diagnosed with cardiovascular diseases and the type 2 diabetes mellitus, breast, lung, liver and prostate malignancies, neurological, sleep, mood and eye disorders, amongst others. Consequent socio-economic burden is tremendous. Unprecedented decrease in age of maladaptive individuals has been reported. The absolute majority of expanding non-communicable disorders carry a chronic character, over a couple of years progressing from reversible suboptimal health conditions to irreversible severe pathologies and cascading collateral complications. The time-frame between onset of SHS and clinical manifestation of associated disorders is the operational area for an application of reliable risk assessment tools and predictive diagnostics followed by the cost-effective targeted prevention and treatments tailored to the person. This article demonstrates advanced strategies in bio/medical sciences and healthcare focused on suboptimal health conditions in the frame-work of Predictive, Preventive and Personalised Medicine (3PM/PPPM). Potential benefits in healthcare systems and for society at large include but are not restricted to an improved life-quality of major populations and socio-economical groups, advanced professionalism of healthcare-givers and sustainable healthcare economy. Amongst others, following medical areas are proposed to strongly benefit from PPPM strategies applied to the identification and treatment of suboptimal health conditions:Stress overload associated pathologiesMale and female healthPlanned pregnanciesPeriodontal healthEye disordersInflammatory disorders, wound healing and pain management with associated complicationsMetabolic disorders and suboptimal body weightCardiovascular pathologiesCancersStroke, particularly of unknown aetiology and in young individualsSleep medicineSports medicineImproved individual outcomes under pandemic conditions such as COVID-19.

Wang Wei, Yan Yuxiang, Guo Zheng, Hou Haifeng, Garcia Monique, Tan Xuerui, Anto Enoch Odame, Mahara Gehendra, Zheng Yulu, Li Bo, Kang Timothy, Zhong Zhaohua, Wang Youxin, Guo Xiuhua, Golubnitschaja Olga

2021-Sep-13

Adolescence, Artificial intelligence (AI), Behavioural patterns, Big data management, Body mass index (BMI), COVID-19, Cancers, Cardiovascular disease, Communicable, Dietary habits, Epidemics, Glycan, Health economy, Health policy, Individualised patient profile, Lifestyle, Liquid biopsy, Medical ethics, Microbiome, Modifiable preventable risks, Mood disorders, Multi-level diagnostics, Multi-parametric analysis, Natural substances, Neurologic diseases, Non-communicable diseases, Omics, Pandemics, Periodontal health, Predictive preventive personalised medicine (PPPM/3PM), Risk assessment, Sleep medicine, Stress overload, Suboptimal health status (SHS), Traditional medicine

General General

Deep Tobit networks: A novel machine learning approach to microeconometrics.

In Neural networks : the official journal of the International Neural Network Society

Tobit models (also called as "censored regression models" or classified as "sample selection models" in microeconometrics) have been widely applied to microeconometric problems with censored outcomes. However, due to their linear parametric settings and restrictive normality assumptions, the traditional Tobit models fail to capture the pervading nonlinearities and thus may be inadequate for microeconometric analysis with large-scale datasets. This paper proposes two novel deep neural networks for Tobit problems and explores machine learning approaches in the context of microeconometric modeling. We connect the censored outputs in Tobit models with some deep learning techniques, which are thought to be unrelated to microeconometrics, and use the rectified linear unit activation and a particularly designed network structure to implement the censored output mechanisms and realize the underlying econometric conceptions. The benchmark Tobit-I and Tobit-II models are then reformulated as two carefully designed deep feedforward neural networks named deep Tobit-I network and deep Tobit-II network, respectively. A novel significance testing method is developed based on the proposed networks. Compared with the traditional models, our networks with deep structures can effectively describe the underlying highly nonlinear relationships and achieve considerable improvements in fitting and prediction. With the novel testing method, the proposed networks enable highly accurate and sophisticated econometric analysis with minimal random assumptions. The encouraging numerical experiments on synthetic and realistic datasets demonstrate the utility and advantages of the proposed method.

Zhang Jiaming, Li Zhanfeng, Song Xinyuan, Ning Hanwen

2021-Sep-09

Deep Tobit network, Large dataset, Microeconometrics, Neural network, Significance test

Surgery Surgery

Periprosthetic Joint Infection Prediction via Machine Learning: Comprehensible Personalized Decision Support for Diagnosis.

In The Journal of arthroplasty ; h5-index 65.0

BACKGROUND : The criteria outlined in the International Consensus Meeting (ICM) in 2018, which were prespecified and fixed, have been commonly practiced by clinicians to diagnose periprosthetic joint infection (PJI). We developed a machine learning (ML) system for PJI diagnosis and compared it with the ICM scoring system to verify the feasibility of ML.

METHODS : We designed an ensemble meta-learner, which combined five learning algorithms to achieve superior performance by optimizing their synergy. To increase the comprehensibility of ML, we developed an explanation generator that produces understandable explanations of individual predictions. We performed stratified 5-fold cross-validation on a cohort of 323 patients to compare the ML meta-learner with the ICM scoring system.

RESULTS : Cross-validation demonstrated ML's superior predictive performance to that of the ICM scoring system for various metrics, including accuracy, precision, recall, F1 score, Matthews correlation coefficient, and area under ROC curve. Moreover, the case study showed that ML was capable of identifying personalized important features missing from ICM and providing interpretable decision support for individual diagnosis.

CONCLUSION : Unlike ICM, ML could construct adaptive diagnostic models from the available patient data instead of making diagnoses based on prespecified criteria. The experimental results suggest that ML is feasible and competitive for PJI diagnosis compared with the current widely used ICM scoring criteria. The adaptive ML models can serve as an auxiliary system to ICM for diagnosing PJI.

Kuo Feng-Chih, Hu Wei-Huan, Hu Yuh-Jyh

2021-Sep-17

International Consensus Meeting, decision support, machine learning, periprosthetic joint infection, prediction

General General

Latent Class Analysis Reveals COVID-19-related ARDS Subgroups with Differential Responses to Corticosteroids.

In American journal of respiratory and critical care medicine ; h5-index 108.0

Rationale Two distinct subphenotypes have been identified in acute respiratory distress syndrome (ARDS), but the presence of subgroups in ARDS associated with COVID-19 is unknown. The objective of this study was to identify clinically relevant, novel subgroups in COVID-19-related ARDS, and compare them to previously described ARDS subphenotypes. Methods Eligible participants were adults with COVID-19 and ARDS at Columbia University Irving Medical Center. Latent class analysis (LCA) was used to identify subgroups with baseline clinical, respiratory, and laboratory data serving as partitioning variables. A previously-developed machine learning model was used to classify patients as the hypoinflammatory and hyperinflammatory subphenotypes. Baseline characteristics and clinical outcomes were compared between subgroups. Heterogeneity of treatment effect (HTE) for corticosteroid-use in subgroups was tested. Measurements and Main Results From 3/2-4/30/2020, 483 patients with COVID-19-related ARDS met study criteria. A two-class LCA model best fit the population (p=0.0075). Class 2 (23%) had higher pro-inflammatory markers, troponin, creatinine and lactate, lower bicarbonate and lower blood pressure than Class 1 (77%). 90-day mortality was higher in Class 2 versus Class 1 (75% vs 48%; p<0.0001). Considerable overlap was observed between these subgroups and ARDS subphenotypes. SARS-CoV-2 RT-PCR cycle threshold was associated with mortality in the hypoinflammatory but not the hyperinflammatory phenotype. HTE to corticosteroids was observed (p=0.0295), with improved mortality in the hyperinflammatory phenotype and worse mortality in the hypoinflammatory phenotype, with the caveat that corticosteroid treatment was not randomized. Conclusions We identified two COVID-19-related ARDS subgroups with differential outcomes, similar to previously described ARDS subphenotypes. SARS-CoV-2 PCR cycle threshold had differential value for predicting mortality in the subphenotypes. The subphenotypes had differential treatment responses to corticosteroids. This article is open access and distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives License 4.0 (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Sinha Pratik, Furfaro David, Cummings Matthew J, Abrams Darryl, Delucchi Kevin, Maddali Manoj V, He June, Thompson Alison, Murn Michael, Fountain John, Rosen Amanda, Robbins-Juarez Shelief Y, Adan Matthew A, Satish Tejus, Madhavan Mahesh, Gupta Aakriti, Lyashchenko Alexander K, Agerstrand Cara, Yip Natalie H, Burkart Kristin M, Beitler Jeremy R, Baldwin Matthew R, Calfee Carolyn S, Brodie Daniel, O’Donnell Max R

2021-Sep-20

ARDS, COVID-19, Latent class analysis, Phenotyping

Surgery Surgery

Serial circulating tumor DNA to predict early recurrence in patients with hepatocellular carcinoma: a prospective study.

In Molecular oncology

We studied the value of circulating tumor DNA (ctDNA) in predicting early postoperative tumor recurrence and monitoring tumor burden in patients with hepatocellular carcinoma (HCC). Plasma-free DNA, germline DNA and tissue DNA were isolated from 41 patients with HCC. Serial ctDNAs were analyzed by next-generation sequencing before and after operation. Whole-exome sequencing was used to detect the DNA of HCC and adjacent tissues. In total, 47 gene mutations were identified in the ctDNA of the 41 patients analyzed before surgery. ctDNA was detected in 63.4% and 46% of the patient tissues pre- and post-operation, respectively. The preoperative ctDNA positivity rate was significantly lower in the nonrecurrence group than in the recurrence group. With a median follow-up of 17.7 months, nine patients (22%) experienced tumor recurrence. ctDNA positivity at two timepoints was associated with significantly shorter recurrence-free survival (RFS). Tumors with NRAS, NEF2L2 and MET mutations had significantly shorter times to recurrence than those without mutations and showed high recurrence prediction performance by machine learning. Multivariate analyses showed that the median variant allele frequency (VAF) of mutations in preoperative ctDNA was a strong independent predictor of RFS. ctDNA is a real-time monitoring indicator that can accurately reflect tumor burden. The median VAF of baseline ctDNA is a strong independent predictor of RFS in individuals with HCC.

Zhu Gui-Qi, Liu Wei-Ren, Tang Zheng, Qu Wei-Feng, Fang Yuan, Jiang Xi-Fei, Song Shu-Shu, Wang Han, Tao Chen-Yang, Zhou Pei-Yun, Huang Run, Gao Jun, Sun Hai-Xiang, Ding Zhen-Bin, Peng Yuan-Fei, Dai Zhi, Zhou Jian, Fan Jia, Shi Ying-Hong

2021-Sep-20

Hepatocellular carcinoma, biomarker, ctDNA, tumor recurrence

General General

Assessing direct analysis in real time mass spectrometry for the identification and serotyping of Legionella pneumophila.

In Journal of applied microbiology

AIMS : The efficacy of ambient mass spectrometry to identify and serotype Legionella pneumophila was assessed. To this aim, isolated waterborne colonies were submitted to a rapid extraction method and analysed by direct analysis in real time mass spectrometry (DART-HRMS).

METHODS AND RESULTS : The DART-HRMS profiles, coupled with partial least squares discriminant analysis (PLS-DA), were first evaluated for their ability to differentiate Legionella spp. from other bacteria. The resultant classification model achieved an accuracy of 98.1% on validation. Capitalising on these encouraging results, DART-HRMS profiling was explored as an alternative approach for the identification of L. pneumophila sg. 1, L. pneumophila sg. 2-15 and L. non-pneumophila; therefore, a different PLS-DA classifier was built. When tested on a validation set, this second classifier reached an overall accuracy of 95.93%. It identified the harmful L. pneumophila sg. 1 with an impressive specificity (100%) and slightly lower sensitivity (91.7%), and similar performances were reached in the classification of L. pneumophila sg. 2-15 and L. non-pneumophila.

CONCLUSIONS : The results of this study show the DART-HMRS method has good accuracy and it is an effective method for Legionella serogroup profiling.

SIGNIFICANCE AND IMPACT OF STUDY : These preliminary findings could open a new avenue for the rapid identification and quick epidemiologic tracing of L. pneumophila, with a consequent improvement to risk assessment.

Tata Alessandra, Marzoli Filippo, Massaro Andrea, Passabì Eleonora, Bragolusi Marco, Negro Alessandro, Cristaudo Ilaria, Piro Roberto, Belluco Simone

2021-Sep-20

ambient mass spectrometry, classification model, machine learning, predictions, supervised statistical analysis

Pathology Pathology

The utility of color normalization for AI-based diagnosis of hematoxylin and eosin-stained pathology images.

In The Journal of pathology

The color variation of hematoxylin and eosin (H&E)-stained tissues has presented a challenge for applications of artificial intelligence (AI) in digital pathology. Many color normalization algorithms have been developed in recent years in order to reduce the color variation between H&E images. However, previous efforts in benchmarking these algorithms have produced conflicting results and none have sufficiently assessed the efficacy of the various color normalization methods for improving diagnostic performance of AI systems. In this study, we systematically investigated eight color normalization algorithms for AI-based classification of H&E-stained histopathology slides, in the context of both using images from one center and from multiple centers. Our results show that color normalization does not consistently improve classification performance when both training and testing data are from a single center. However, using four multi-center datasets of two cancer types (ovarian and pleural) and objective functions, we show that color normalization can significantly improve the classification accuracy of images from external datasets (ovarian cancer: 0.25 AUC increase, p = 1.6 e-05, pleural cancer: 0.21 AUC increase, p = 1.4 e-10). Furthermore, we introduce a novel augmentation strategy by mixing color-normalized images using three easily accessible algorithms that consistently improves the diagnosis of test images from external centers, even when the individual normalization methods had varied results. We anticipate our study to be a starting point for reliable use of color normalization to improve AI-based, digital pathology-empowered diagnosis of cancers sourced from multiple centers. This article is protected by copyright. All rights reserved.

Boschman Jeffrey, Farahani Hossein, Darbandsari Amirali, Ahmadvand Pouya, Van Spankeren Ashley, Farnell David, Levine Adrian, Naso Julia R, Churg Andrew, Jones Steven, Yip Stephen, Koebel Martin, Huntsman David, Gilks Blake, Bashashati Ali

2021-Sep-20

Artificial intelligence, Color normalization, Digital image analysis, Digital pathology, Machine learning, Stain normalization

Internal Medicine Internal Medicine

Granulocyte colony-stimulating factor promotes an aggressive phenotype of colon and breast cancer cells with biochemical changes investigated by single-cell Raman microspectroscopy and machine learning analysis.

In The Analyst

Granulocyte colony-stimulating factor (G-CSF) is produced at high levels in several cancers and is directly linked with metastasis in gastrointestinal (GI) cancers. In order to further understand the alteration of molecular compositions and biochemical features triggered by G-CSF treatment at molecular and cell levels, we sought to investigate the long term treatment of G-CSF on colon and breast cancer cells measured by label-free, non-invasive single-cell Raman microspectroscopy. Raman spectrum captures the molecule-specific spectral signatures ("fingerprints") of different biomolecules presented on cells. In this work, mouse breast cancer line 4T1 and mouse colon cancer line CT26 were treated with G-CSF for 7 weeks and subsequently analyzed by machine learning based Raman spectroscopy and gene/cytokine expression. The principal component analysis (PCA) identified the Raman bands that most significantly changed between the control and G-CSF treated cells. Notably, here we proposed the concept of aggressiveness score, which can be derived from the posterior probability of linear discriminant analysis (LDA), for quantitative spectral analysis of tumorigenic cells. The aggressiveness score was effectively applied to analyze and differentiate the overall cell biochemical changes of G-CSF-treated two model cancer cells. All these tumorigenic progressions suggested by Raman analysis were confirmed by pro-tumorigenic cytokine and gene analysis. A high correlation between gene expression data and Raman spectra highlights that the machine learning based non-invasive Raman spectroscopy offers emerging and powerful tools to better understand the regulation mechanism of cytokines in the tumor microenvironment that could lead to the discovery of new targets for cancer therapy.

Zhang Wei, Karagiannidis Ioannis, Van Vliet Eliane De Santana, Yao Ruoxin, Beswick Ellen J, Zhou Anhong

2021-Sep-20

General General

Replicating prediction algorithms for hospitalization and corticosteroid use in patients with inflammatory bowel disease.

In PloS one ; h5-index 176.0

INTRODUCTION : Previous work had shown that machine learning models can predict inflammatory bowel disease (IBD)-related hospitalizations and outpatient corticosteroid use based on patient demographic and laboratory data in a cohort of United States Veterans. This study aimed to replicate this modeling framework in a nationally representative cohort.

METHODS : A retrospective cohort design using Optum Electronic Health Records (EHR) were used to identify IBD patients, with at least 12 months of follow-up between 2007 and 2018. IBD flare was defined as an inpatient/emergency visit with a diagnosis of IBD or an outpatient corticosteroid prescription for IBD. Predictors included demographic and laboratory data. Logistic regression and random forest (RF) models were used to predict IBD flare within 6 months of each visit. A 70% training and 30% validation approach was used.

RESULTS : A total of 95,878 patients across 780,559 visits were identified. Of these, 22,245 (23.2%) patients had at least one IBD flare. Patients were predominantly White (87.7%) and female (57.1%), with a mean age of 48.0 years. The logistic regression model had an area under the receiver operating curve (AuROC) of 0.66 (95% CI: 0.65-0.66), sensitivity of 0.69 (95% CI: 0.68-0.70), and specificity of 0.74 (95% CI: 0.73-0.74) in the validation cohort. The RF model had an AuROC of 0.80 (95% CI: 0.80-0.81), sensitivity of 0.74 (95% CI: 0.73-0.74), and specificity of 0.72 (95% CI: 0.72-0.72) in the validation cohort. Important predictors of IBD flare in the RF model were the number of previous flares, age, potassium, and white blood cell count.

CONCLUSION : The machine learning modeling framework was replicated and results showed a similar predictive accuracy in a nationally representative cohort of IBD patients. This modeling framework could be embedded in routine practice as a tool to distinguish high-risk patients for disease activity.

Gan Ryan W, Sun Diana, Tatro Amanda R, Cohen-Mekelburg Shirley, Wiitala Wyndy L, Zhu Ji, Waljee Akbar K

2021

Public Health Public Health

Identification of high-risk COVID-19 patients using machine learning.

In PloS one ; h5-index 176.0

The current COVID-19 public health crisis, caused by SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2), has produced a devastating toll both in terms of human life loss and economic disruption. In this paper we present a machine-learning algorithm capable of identifying whether a given patient (actually infected or suspected to be infected) is more likely to survive than to die, or vice-versa. We train this algorithm with historical data, including medical history, demographic data, as well as COVID-19-related information. This is extracted from a database of confirmed and suspected COVID-19 infections in Mexico, constituting the official COVID-19 data compiled and made publicly available by the Mexican Federal Government. We demonstrate that the proposed method can detect high-risk patients with high accuracy, in each of four identified clinical stages, thus improving hospital capacity planning and timely treatment. Furthermore, we show that our method can be extended to provide optimal estimators for hypothesis-testing techniques commonly-used in biological and medical statistics. We believe that our work could be of use in the context of the current pandemic in assisting medical professionals with real-time assessments so as to determine health care priorities.

Quiroz-Juárez Mario A, Torres-Gómez Armando, Hoyo-Ulloa Irma, León-Montiel Roberto de J, U’Ren Alfred B

2021

General General

Ethics, Integrity and Retributions of Digital Detection Surveillance Systems on Infectious Diseases: Systematic literature review.

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : The COVID-19 pandemic has raised the importance of the deployment of digital detection surveillance systems to support early warning and monitoring of infectious diseases (ID). These opportunities create a "double-edge sword" as the ethical governance of such approaches often lag behind technological achievements.

OBJECTIVE : The aim was to investigate ethical issues identified from utilizing AI-augmented surveillance or early warning systems to monitor and detect common or novel ID outbreaks.

METHODS : We searched relevant articles in a number of databases that addressed ethical issues of using artificial intelligence, digital surveillance systems, early warning systems and/or big data analytics technology for detecting, monitoring, or tracing ID according to PRISMA guidelines, and further identified and analysed them with a theoretical framework.

RESULTS : This systematic review identified 29 articles presented in six major themes clustered under individual, organizational and societal levels, including: awareness of implementing digital surveillance, digital integrity, trust, privacy and confidentiality, civil rights, and governance. Whilst these measures were understandable during a pandemic, the public were concerned about receiving inadequate information, unclear governance frameworks, and lack of privacy protection, data integrity and autonomy when utilizing ID digital surveillance. The barriers to engagement could widen existing healthcare disparities or digital divides by underrepresenting vulnerable and at-risk populations, and expose patients' highly sensitive data such as their movements and contacts to outside sources, impinging significantly upon basic human and civil rights.

CONCLUSIONS : Our findings inform ethical considerations for service delivery models for medical practitioners and policymakers implicated in the use of digital surveillance for ID spread and a basis for the global governance structure.

CLINICALTRIAL :

Zhao Yan Ivy, Ma Ye Xuan, Yu Man Wai Cecilia, Liu Jia, Dong Wei Nan, Pang Qin, Lu Xiao Qin, Molassiotis Alex, Holroyd Eleanor, Wong Chi Wai William

2021-Sep-14

Surgery Surgery

Road to automating robotic suturing skills assessment: Battling mislabeling of the ground truth.

In Surgery ; h5-index 54.0

OBJECTIVE : To automate surgeon skills evaluation using robotic instrument kinematic data. Additionally, to implement an unsupervised mislabeling detection algorithm to identify potentially mislabeled samples that can be removed to improve model performance.

METHODS : Video recordings and instrument kinematic data were derived from suturing exercises completed on the Mimic FlexVR robotic simulator. A structured human consensus-building process was developed to determine Robotic Anastomosis Competency Evaluation technical scores across 3 human graders. A 2-layer long short-term memory-based classification model used instrument kinematic data to automate suturing skills assessment. An unsupervised label analyzer (NoiseRank) was used to identify potential mislabeling of skills data. Performance of the long short-term memory model's technical skill score prediction was measured by best area under the curve over the training runs. NoiseRank outputted a ranked list of rated skills assessments based on likelihood of mislabeling.

RESULTS : 22 surgeons performed 226 suturing attempts, which were broken down into 1,404 individual skill assessment points. Automation of needle entry angle, needle driving, and needle withdrawal technical skill scores performed better (area under the curve 0.698-0.705) than needle positioning (0.532) at baseline using all available data. Potential mislabels were subsequently identified by NoiseRank and removed, improving model performance across all domains (area under the curve 0.551-0.766).

CONCLUSION : Using ground truth labels from human graders and robotic instrument kinematic data, machine learning models have automated assessment of detailed suturing technical skills with good performance. Further, an unsupervised mislabeling detection algorithm projected mislabeled data, allowing for their removal and subsequent improvement of model performance.

Hung Andrew J, Rambhatla Sirisha, Sanford Daniel I, Pachauri Nilay, Vanstrum Erik, Nguyen Jessica H, Liu Yan

2021-Sep-16

General General

Automated Detection of COVID-19 Cough.

In Biomedical signal processing and control

Easy detection of COVID-19 is a challenge. Quick biological tests do not give enough accuracy. Success in the fight against new outbreaks depends not only on the efficiency of the tests used, but also on the cost, time elapsed and the number of tests that can be done massively. Our proposal provides a solution to this challenge. The main objective is to design a freely available, quick and efficient methodology for the automatic detection of COVID-19 in raw audio files. Our proposal is based on automated extraction of time-frequency cough features and selection of the more significant ones to be used to diagnose COVID-19 using a supervised machine-learning algorithm. Random Forest has performed better than the other models analysed in this study. An accuracy close to 90% was obtained. This study demonstrates the feasibility of the automatic diagnose of COVID-19 from coughs, and its applicability to detecting new outbreaks.

Tena Alberto, Clarià Francesc, Solsona Francesc

2021-Sep-13

COVID-19, automated cough detection, diagnosis, signal processing, time-frequency.

General General

Scale-adaptive Deep Model for Bacterial Raman Spectra Identification.

In IEEE journal of biomedical and health informatics

The combination of Raman spectroscopy and deep learning technology provides an automatic, rapid, and accurate scheme for the clinical diagnosis of pathogenic bacteria. However, the accuracy of existing deep learning methods is still limited because of the single and fixed scales of deep neural networks. We propose a deep neural network that can learn multi-scale features of Raman spectra by using the automatic combination of multi-receptive fields of convolutional layers. This model is based on the expert knowledge that the discrimination information of Raman spectra is composed of multi-scale spectral peaks. We enhance the interpretability of the model by visualizing the activated wavenumbers of the bacterial spectrum that can be used for reference in related work. Compared with existing state-of-the-art methods, the proposed method achieves higher accuracy and efficiency for bacterial identification on isolate-level, empiric-treatment-level, and antibiotic-resistance-level tasks. The clinical bacterial identification task requires significantly fewer patient samples to achieve similar accuracy. Therefore, this method has tremendous potential for the identification of clinical pathogenic bacteria, antibiotic susceptibility testing, and prescription guidance.

Deng Lin, Zhong Yuzhong, Wang Maoning, Zheng Xiujuan, Zhang Jianwei

2021-Sep-20

General General

Regression and Classification of Alzheimers Disease Diagnosis using NMF-TDNet Features from 3D Brain MR Image.

In IEEE journal of biomedical and health informatics

With the development of deep learning and medical imaging technology, many researchers use convolutional neural network(CNN) to obtain deep-level features of medical image in order to better classify Alzheimer's disease (AD) and predict clinical scores. The principal component analysis network (PCANet) is a lightweight deep-learning network that mainly uses principal component analysis (PCA) to generate multilevel filter banks for the centralized learning of samples and then performs binarization and generates blockwise histograms to obtain image features. However, the dimensions of the extracted PCANet features reaching tens of thousands or even hundreds of thousands, and the formation of the multilevel filter banks is sample data dependent, reducing the flexibility of PCANet. In this paper, based on the idea of PCANet, we propose a data-independent network called the nonnegative matrix factorization tensor decomposition network (NMF-TDNet), which improves the computational efficiency and solves the data dependence problem of PCANet. In this network, we use nonnegative matrix factorization (NMF) instead of PCA to create multilevel filter banks for sample learning, then uses the learning results to build a higher-order tensor and perform tensor decomposition (TD) to achieve data dimensionality reduction, producing the final image features. Finally, our method use these features as the input of the support vector machine (SVM) for AD classification diagnosis and clinical score prediction. The performance of our method is extensively evaluated on the ADNI-1, ADNI-2 and OASIS datasets. The experimental results show that NMF-TDNet can achieve data dimensionality reduction (the dimensionality of the extracted features numbers only a few hundred dimensions, far less than the hundreds of thousands required by PCANet) and the NMF-TDNet features as input achieved superior performance than using PCANet features as input.

Lao Huan, Zhang Xuejun

2021-Sep-20

Pathology Pathology

A Cloud Approach for Melanoma Detection based on Deep Learning Networks.

In IEEE journal of biomedical and health informatics

In the era of digitized images, the goal is to be able to extract information from them and create new knowledge thanks to the use of Computer Vision techniques, Machine Learning and Deep Learning. This allows their use for early diagnosis and subsequent determination of the treatment of many pathologies. In the specific case treated here, deep neural networks are used in the dermatological field to distinguish between melanoma and non-melanoma images. In this work we have underlined two essential points of melanoma detection research. The first aspect taken into consideration is how even a simple modification of the parameters in the dataset determines a change of the accuracy of the classifiers, while working on the same original dataset. The second point is the need to have a system architecture that can be more flexible in updating the training datasets for the classification of this pathology. In this context, the proposed architecture reserves the goal of developing and implementing a hybrid architecture based on Cloud, Fog and Edge Computing in order to provide a Melanoma Detection service based on clinical and/or dermoscopic images. At the same time, this architecture must be able to interface with the amount of data to be analyzed by reducing the running time of the necessary computational operations. This has been highlighted with experiments carried out on a single machine and on different distribution systems, highlighting how a distributed approach guarantees the achievement of an output in a much more acceptable time without the need to fully rely on data scientists skills.

Dibiasi Luigi, Risi Michele, Tortora Genoveffa, Auriemma Citarella Alessia

2021-Sep-20

General General

Model-Based Transfer Reinforcement Learning Based on Graphical Model Representations.

In IEEE transactions on neural networks and learning systems

Reinforcement learning (RL) plays an essential role in the field of artificial intelligence but suffers from data inefficiency and model-shift issues. One possible solution to deal with such issues is to exploit transfer learning. However, interpretability problems and negative transfer may occur without explainable models. In this article, we define Relation Transfer as explainable and transferable learning based on graphical model representations, inferring the skeleton and relations among variables in a causal view and generalizing to the target domain. The proposed algorithm consists of the following three steps. First, we leverage a suitable casual discovery method to identify the causal graph based on the augmented source domain data. After that, we make inferences on the target model based on the prior causal knowledge. Finally, offline RL training on the target model is utilized as prior knowledge to improve the policy training in the target domain. The proposed method can answer the question of what to transfer and realize zero-shot transfer across related domains in a principled way. To demonstrate the robustness of the proposed framework, we conduct experiments on four classical control problems as well as one simulation to the real-world application. Experimental results on both continuous and discrete cases demonstrate the efficacy of the proposed method.

Sun Yuewen, Zhang Kun, Sun Changyin

2021-Sep-20

General General

DHNLDA: A novel deep hierarchical network based method for predicting lncRNA-disease associations.

In IEEE/ACM transactions on computational biology and bioinformatics

Recent studies have found that lncRNA (long non-coding RNA) in ncRNA (non-coding RNA) is not only involved in many biological processes, but also abnormally expressed in many complex diseases. Identification of lncRNA-disease associations accurately is of great significance for understanding the function of lncRNA and disease mechanism. In this paper, a deep learning framework consisting of stacked autoencoder(SAE), multi-scale ResNet and stacked ensemble module, named DHNLDA, was constructed to predict lncRNA-disease associations, which integrates multiple biological data sources and constructing feature matrices. Among them, the biological data including the similarity and the interaction of lncRNAs, diseases and miRNAs are integrated. The feature matrices are obtained by node2vec embedding and feature extraction respectively. Then, the SAE and the multi-scale ResNet are used to learn the complementary information between nodes, and the high-level features of node attributes are obtained. Finally, the fusion of high-level feature is input into the stacked ensemble module to obtain the prediction results of lncRNA-disease associations. The experimental results of five-fold cross-validation show that the AUC of DHNLDA reaches 0.975 better than the existing methods. Case studies of stomach cancer, breast cancer and lung cancer have shown the great ability of DHNLDA to discover the potential lncRNA-disease associations.

Xie Fansen, Yang Ziqi, Song Jinmiao, Dai Qiguo, Duan Xiaodong

2021-Sep-20

General General

Semantic-oriented Labeled-to-unlabeled Distribution Translation for Image Segmentation.

In IEEE transactions on medical imaging ; h5-index 74.0

Automatic medical image segmentation plays a crucial role in many medical applications, such as disease diagnosis and treatment planning. Existing deep learning based models usually regarded the segmentation task as pixel-wise classification and neglected the semantic correlations of pixels across different images, leading to vague feature distribution. Moreover, pixel-wise annotated data is rare in medical domain, and the scarce annotated data usually exhibits the biased distribution against the desired one, hindering the performance improvement under the supervised learning setting. In this paper, we propose a novel Labeled-to-unlabeled Distribution Translation (L2uDT) framework with Semantic-oriented Contrastive Learning (SoCL), mainly for addressing the aforementioned issues in medical image segmentation. In SoCL, a semantic grouping module is designed to cluster pixels into a set of semantically coherent groups, and a semantic-oriented contrastive loss is advanced to constrain group-wise prototypes, so as to explicitly learn a feature space with intra-class compactness and inter-class separability. We then establish a L2uDT strategy to approximate the desired data distribution for unbiased optimization, where we translate the labeled data distribution with the guidance of extensive unlabeled data. In particular, a bias estimator is devised to measure the distribution bias, then a gradual-paced shift is derived to progressively translate the labeled data distribution to unlabeled one. Both labeled and translated data are leveraged to optimize the segmentation model simultaneously. We illustrate the effectiveness of the proposed method on two benchmark datasets, EndoScene and PROSTATEx, and our method achieves state-of-the-art performance, which clearly demonstrates its effectiveness for medical image segmentation. The source code is available at https://github.com/CityU-AIM-Group/L2uDT.

Guo Xiaoqing, Liu Jie, Yuan Yixuan

2021-Sep-20

Pathology Pathology

Propagating Uncertainty Across Cascaded Medical Imaging Tasks For Improved Deep Learning Inference.

In IEEE transactions on medical imaging ; h5-index 74.0

Although deep networks have been shown to perform very well on a variety of medical imaging tasks, inference in the presence of pathology presents several challenges to common models. These challenges impede the integration of deep learning models into real clinical workflows, where the customary process of cascading deterministic outputs from a sequence of image-based inference steps (e.g. registration, segmentation) generally leads to an accumulation of errors that impacts the accuracy of downstream inference tasks. In this paper, we propose that by embedding uncertainty estimates across cascaded inference tasks, performance on the downstream inference tasks should be improved. We demonstrate the effectiveness of the proposed approach in three different clinical contexts: (i) We demonstrate that by propagating T2 weighted lesion segmentation results and their associated uncertainties, subsequent T2 lesion detection performance is improved when evaluated on a proprietary large-scale, multi-site, clinical trial dataset acquired from patients with Multiple Sclerosis. (ii) We show an improvement in brain tumour segmentation performance when the uncertainty map associated with a synthesised missing MR volume is provided as an additional input to a follow-up brain tumour segmentation network, when evaluated on the publicly available BraTS-2018 dataset. (iii) We show that by propagating uncertainties from a voxel-level hippocampus segmentation task, the subsequent regression of the Alzheimer's disease clinical score is improved.

Mehta Raghav, Christinck Thomas, Nair Tanya, Bussy Aurelie, Premasiri Swapna, Costantino Manuela, Chakravarty Mallar, Arnold Douglas L, Gal Yarin, Arbel Tal

2021-Sep-20

General General

Commentary: Community Knowledge for Equity in Healthcare.

In Healthcare policy = Politiques de sante

In their insightful article, Sayani et al. (2021) provide five considerations for developing patient partnerships that are meaningful and inclusive. In this brief rejoinder, we outline three points that push the boundaries of the discussion on diverse patient partnerships and represent challenges faced by our own research team as we aim to build and deepen our approach to community engagement. Firstly, we suggest a shift from patient engagement to community engagement; secondly, we propose a shift from engaging various communities together by labelling them as "underserved" or "structurally marginalized" to engaging specific cultural or geographic communities at specific times; and finally, we suggest deferring to community knowledge.

Shaw James, Sky Philina, Chandra Shivani

2021-Aug

General General

A Machine Learning Approach for Predicting Sustained Remission in Rheumatoid Arthritis Patients on Biologic Agents.

In Journal of clinical rheumatology : practical reports on rheumatic & musculoskeletal diseases

Background : Despite several studies having identified factors associated with successful treatment outcomes in rheumatoid arthritis (RA), there is a lack of accurate predictive models for sustained remission in patients on biologic agents. To the best of our knowledge, no machine learning (ML) approaches apart from logistic regression (LR) have ever been tried on this class of problems.

Methods : In this longitudinal study, patients with RA who started a biological disease-modifying antirheumatic drug (bDMARD) in a tertiary care center were analyzed. Demographic and clinical characteristics were collected at treatment baseline, 12-month, and 24-month follow-up. A wrapper feature selection algorithm was used to determine an attribute core set. Four different ML algorithms, namely, LR, random forest, K-nearest neighbors, and extreme gradient boosting, were then trained and validated with 10-fold cross-validation to predict 24-month sustained DAS28 (Disease Activity Score on 28 joints) remission. The performances of the algorithms were then compared assessing accuracy, precision, and recall.

Results : Our analysis included 367 patients (female 323/367, 88%) with mean age +/- SD of 53.7 +/- 12.5 years at bDMARD baseline. Sustained DAS28 remission was achieved by 175 (47.2%) of 367 patients. The attribute core set used to train algorithms included acute phase reactant levels, Clinical Disease Activity Index, Health Assessment Questionnaire-Disability Index, as well as several clinical characteristics. Extreme gradient boosting showed the best performance (accuracy, 72.7%; precision, 73.2%; recall, 68.1%), outperforming random forest (accuracy, 65.9%; precision, 65.6%; recall, 59.3%), LR (accuracy, 64.9%; precision, 62.6%; recall, 61.9%), and K-nearest neighbors (accuracy, 63%; precision, 61.5%; recall, 54.8%).

Conclusions : We showed that ML models can be used to predict sustained remission in RA patients on bDMARDs. Furthermore, our method only relies on a few easy-to-collect patient attributes. Our results are promising but need to be tested on longitudinal cohort studies.

Venerito Vincenzo, Angelini Orazio, Fornaro Marco, Cacciapaglia Fabio, Lopalco Giuseppe, Iannone Florenzo

2021-Feb-10

Surgery Surgery

Self-supervised representation learning for surgical activity recognition.

In International journal of computer assisted radiology and surgery

PURPOSE : Virtual reality-based simulators have the potential to become an essential part of surgical education. To make full use of this potential, they must be able to automatically recognize activities performed by users and assess those. Since annotations of trajectories by human experts are expensive, there is a need for methods that can learn to recognize surgical activities in a data-efficient way.

METHODS : We use self-supervised training of deep encoder-decoder architectures to learn representations of surgical trajectories from video data. These representations allow for semi-automatic extraction of features that capture information about semantically important events in the trajectories. Such features are processed as inputs of an unsupervised surgical activity recognition pipeline.

RESULTS : Our experiments document that the performance of hidden semi-Markov models used for recognizing activities in a simulated myomectomy scenario benefits from using features extracted from representations learned while training a deep encoder-decoder network on the task of predicting the remaining surgery progress.

CONCLUSION : Our work is an important first step in the direction of making efficient use of features obtained from deep representation learning for surgical activity recognition in settings where only a small fraction of the existing data is annotated by human domain experts and where those annotations are potentially incomplete.

Paysan Daniel, Haug Luis, Bajka Michael, Oelhafen Markus, Buhmann Joachim M

2021-Sep-20

Deep Learning, Probabilistic modeling, Representation Learning, Self-supervised Learning, Surgical Activity Recognition, Unsupervised Learning

Pathology Pathology

Deep Anomaly Generation: An Image Translation Approach of Synthesizing Abnormal Banded Chromosome Images

ArXiv Preprint

Advances in deep-learning-based pipelines have led to breakthroughs in a variety of microscopy image diagnostics. However, a sufficiently big training data set is usually difficult to obtain due to high annotation costs. In the case of banded chromosome images, the creation of big enough libraries is difficult for multiple pathologies due to the rarity of certain genetic disorders. Generative Adversarial Networks (GANs) have proven to be effective in generating synthetic images and extending training data sets. In our work, we implement a conditional adversarial network that allows generation of realistic single chromosome images following user-defined banding patterns. To this end, an image-to-image translation approach based on self-generated 2D chromosome segmentation label maps is used. Our validation shows promising results when synthesizing chromosomes with seen as well as unseen banding patterns. We believe that this approach can be exploited for data augmentation of chromosome data sets with structural abnormalities. Therefore, the proposed method could help to tackle medical image analysis problems such as data simulation, segmentation, detection, or classification in the field of cytogenetics.

Lukas Uzolas, Javier Rico, Pierrick Coupé, Juan C. SanMiguel, György Cserey

2021-09-20

Surgery Surgery

Predicting Visual Improvement after Macular Hole Surgery: a Cautionary Tale on Deep Learning with Very Limited Data

ArXiv Preprint

We investigate the potential of machine learning models for the prediction of visual improvement after macular hole surgery from preoperative data (retinal images and clinical features). Collecting our own data for the task, we end up with only 121 total samples, putting our work in the very limited data regime. We explore a variety of deep learning methods for limited data to train deep computer vision models, finding that all tested deep vision models are outperformed by a simple regression model on the clinical features. We believe this is compelling evidence of the extreme difficulty of using deep learning on very limited data.

M. Godbout, A. Lachance, F. Antaki, A. Dirani, A. Durand

2021-09-20

General General

FUTURE-AI: Guiding Principles and Consensus Recommendations for Trustworthy Artificial Intelligence in Future Medical Imaging

ArXiv Preprint

The recent advancements in artificial intelligence (AI) combined with the extensive amount of data generated by today's clinical systems, has led to the development of imaging AI solutions across the whole value chain of medical imaging, including image reconstruction, medical image segmentation, image-based diagnosis and treatment planning. Notwithstanding the successes and future potential of AI in medical imaging, many stakeholders are concerned of the potential risks and ethical implications of imaging AI solutions, which are perceived as complex, opaque, and difficult to comprehend, utilise, and trust in critical clinical applications. Despite these concerns and risks, there are currently no concrete guidelines and best practices for guiding future AI developments in medical imaging towards increased trust, safety and adoption. To bridge this gap, this paper introduces a careful selection of guiding principles drawn from the accumulated experiences, consensus, and best practices from five large European projects on AI in Health Imaging. These guiding principles are named FUTURE-AI and its building blocks consist of (i) Fairness, (ii) Universality, (iii) Traceability, (iv) Usability, (v) Robustness and (vi) Explainability. In a step-by-step approach, these guidelines are further translated into a framework of concrete recommendations for specifying, developing, evaluating, and deploying technically, clinically and ethically trustworthy AI solutions into clinical practice.

Karim Lekadira, Richard Osuala, Catherine Gallin, Noussair Lazrak, Kaisar Kushibar, Gianna Tsakou, Susanna Aussó, Leonor Cerdá Alberich, Konstantinos Marias, Manolis Tskinakis, Sara Colantonio, Nickolas Papanikolaou, Zohaib Salahuddin, Henry C Woodruff, Philippe Lambin, Luis Martí-Bonmatí

2021-09-20

General General

Data Augmentation Through Monte Carlo Arithmetic Leads to More Generalizable Classification in Connectomics

ArXiv Preprint

Machine learning models are commonly applied to human brain imaging datasets in an effort to associate function or structure with behaviour, health, or other individual phenotypes. Such models often rely on low-dimensional maps generated by complex processing pipelines. However, the numerical instabilities inherent to pipelines limit the fidelity of these maps and introduce computational bias. Monte Carlo Arithmetic, a technique for introducing controlled amounts of numerical noise, was used to perturb a structural connectome estimation pipeline, ultimately producing a range of plausible networks for each sample. The variability in the perturbed networks was captured in an augmented dataset, which was then used for an age classification task. We found that resampling brain networks across a series of such numerically perturbed outcomes led to improved performance in all tested classifiers, preprocessing strategies, and dimensionality reduction techniques. Importantly, we find that this benefit does not hinge on a large number of perturbations, suggesting that even minimally perturbing a dataset adds meaningful variance which can be captured in the subsequently designed models.

Gregory Kiar, Yohan Chatelain, Ali Salari, Alan C. Evans, Tristan Glatard

2021-09-20

General General

Incremental Learning Techniques for Online Human Activity Recognition

ArXiv Preprint

Unobtrusive and smart recognition of human activities using smartphones inertial sensors is an interesting topic in the field of artificial intelligence acquired tremendous popularity among researchers, especially in recent years. A considerable challenge that needs more attention is the real-time detection of physical activities, since for many real-world applications such as health monitoring and elderly care, it is required to recognize users' activities immediately to prevent severe damages to individuals' wellness. In this paper, we propose a human activity recognition (HAR) approach for the online prediction of physical movements, benefiting from the capabilities of incremental learning algorithms. We develop a HAR system containing monitoring software and a mobile application that collects accelerometer and gyroscope data and send them to a remote server via the Internet for classification and recognition operations. Six incremental learning algorithms are employed and evaluated in this work and compared with several batch learning algorithms commonly used for developing offline HAR systems. The Final results indicated that considering all performance evaluation metrics, Incremental K-Nearest Neighbors and Incremental Naive Bayesian outperformed other algorithms, exceeding a recognition accuracy of 95% in real-time.

Meysam Vakili, Masoumeh Rezaei

2021-09-20

oncology Oncology

DeepStationing: Thoracic Lymph Node Station Parsing in CT Scans using Anatomical Context Encoding and Key Organ Auto-Search

ArXiv Preprint

Lymph node station (LNS) delineation from computed tomography (CT) scans is an indispensable step in radiation oncology workflow. High inter-user variabilities across oncologists and prohibitive laboring costs motivated the automated approach. Previous works exploit anatomical priors to infer LNS based on predefined ad-hoc margins. However, without voxel-level supervision, the performance is severely limited. LNS is highly context-dependent - LNS boundaries are constrained by anatomical organs - we formulate it as a deep spatial and contextual parsing problem via encoded anatomical organs. This permits the deep network to better learn from both CT appearance and organ context. We develop a stratified referencing organ segmentation protocol that divides the organs into anchor and non-anchor categories and uses the former's predictions to guide the later segmentation. We further develop an auto-search module to identify the key organs that opt for the optimal LNS parsing performance. Extensive four-fold cross-validation experiments on a dataset of 98 esophageal cancer patients (with the most comprehensive set of 12 LNSs + 22 organs in thoracic region to date) are conducted. Our LNS parsing model produces significant performance improvements, with an average Dice score of 81.1% +/- 6.1%, which is 5.0% and 19.2% higher over the pure CT-based deep model and the previous representative approach, respectively.

Dazhou Guo, Xianghua Ye, Jia Ge, Xing Di, Le Lu, Lingyun Huang, Guotong Xie, Jing Xiao, Zhongjie Liu, Ling Peng, Senxiang Yan, Dakai Jin

2021-09-20

Public Health Public Health

Moderate to severe OSA screening based on support vector machine of the Chinese population faciocervical measurements dataset: a cross-sectional study.

In BMJ open

OBJECTIVES : Obstructive sleep apnoea (OSA) has received much attention as a risk factor for perioperative complications and 68.5% of OSA patients remain undiagnosed before surgery. Faciocervical characteristics may screen OSA for Asians due to smaller upper airways compared with Caucasians. Thus, our study aimed to explore a machine-learning model to screen moderate to severe OSA based on faciocervical and anthropometric measurements.

DESIGN : A cross-sectional study.

SETTING : Data were collected from the Shanghai Jiao Tong University School of Medicine affiliated Ruijin Hospital between February 2019 and August 2020.

PARTICIPANTS : A total of 481 Chinese participants were included in the study. PRIMARY AND SECONDARY OUTCOME: (1) Identification of moderate to severe OSA with apnoea-hypopnoea index 15 events/hour and (2) Verification of the machine-learning model.

RESULTS : Sex-Age-Body mass index (BMI)-maximum Interincisal distance-ratio of Height to thyrosternum distance-neck Circumference-waist Circumference (SABIHC2) model was set up. The SABIHC2 model could screen moderate to severe OSA with an area under the curve (AUC)=0.832, the sensitivity of 0.916 and specificity of 0.749, and performed better than the STOP-BANG (snoring, tiredness, observed apnea, high blood pressure, BMI, age, neck circumference, and male gender) questionnaire, which showed AUC=0.631, the sensitivity of 0.487 and specificity of 0.772. Especially for asymptomatic patients (Epworth Sleepiness Scale <10), the SABIHC2 model demonstrated better predictive ability compared with the STOP-BANG questionnaire, with AUC (0.824 vs 0.530), sensitivity (0.892 vs 0.348) and specificity (0.755 vs 0.809).

CONCLUSION : The SABIHC2 machine-learning model provides a simple and accurate assessment of moderate to severe OSA in the Chinese population, especially for those without significant daytime sleepiness.

Zhang Liu, Yan Ya Ru, Li Shi Qi, Li Hong Peng, Lin Ying Ni, Li Ning, Sun Xian Wen, Ding Yong Jie, Li Chuan Xiang, Li Qing Yun

2021-Sep-17

adult anaesthesia, public health, sleep medicine

General General

Generating insights in uncharted territories: real-time learning from data in critically ill patients-an implementer report.

In BMJ health & care informatics

Introduction In the current situation, clinical patient data are often siloed in multiple hospital information systems. Especially in the intensive care unit (ICU), large volumes of clinical data are routinely collected through continuous patient monitoring. Although these data often contain useful information for clinical decision making, they are not frequently used to improve quality of care. During, but also after, pressing times, data-driven methods can be used to mine treatment patterns from clinical data to determine the best treatment options from a hospitals own clinical data.Methods In this implementer report, we describe how we implemented a data infrastructure that enabled us to learn in real time from consecutive COVID-19 ICU admissions. In addition, we explain our step-by-step multidisciplinary approach to establish such a data infrastructure.Conclusion By sharing our steps and approach, we aim to inspire others, in and outside ICU walls, to make more efficient use of data at hand, now and in the future.

van de Sande Davy, Van Genderen Michel E, Huiskens Joost, Veen Robert E R, Meijerink Yvonne, Gommers Diederik, van Bommel Jasper

2021-Sep

COVID-19, artificial intelligence, critical care outcomes, data science, machine learning

Public Health Public Health

A machine learning approach to determine the prognosis of patients with Class III malocclusion.

In American journal of orthodontics and dentofacial orthopedics : official publication of the American Association of Orthodontists, its constituent societies, and the American Board of Orthodontics

INTRODUCTION : The conundrum of determining how to treat a patient with Class III malocclusion is significant, creating a burden on the patient and challenging the orthodontist. The objective of this study was to employ a statistical prediction model derived from our previous cephalometric data on 5 predominant subtypes of skeletal Class III malocclusion to test the hypothesis that Class III subtypes are associated with treatment modalities (eg, surgical vs nonsurgical) and treatment outcome.

METHODS : Pretreatment lateral cephalometric records of 148 patients were digitized for 67 cephalometric variables, and measurements were applied to a mathematical equation to assign a Class III subtype. Subjects were assigned to either a surgical or nonsurgical group depending on the treatment received. Treatment outcome was determined by facial profile and clinical photographs. Log binomial models were used for statistical analysis.

RESULTS : Subtype 1 (mandibular prognathic) patients were 3.5 × more likely to undergo orthognathic surgery than subtypes 2/3 (maxillary deficient) and 5.3 × more likely than 4/5 (combination). Subtype 1 patients were also 1.5 × more likely to experience treatment failure than subtypes 2/3 (maxillary deficient) and 4/5 (combination).

CONCLUSIONS : This assessment of a systematic method to characterize patients with Class III malocclusion into subtypes revealed that subtype 1 (mandibular prognathic) showed a likelihood to undergo orthognathic surgery while subtypes 2/3 experienced significantly lower treatment failure (in response to orthodontics alone). Further refinement of the equation may yield a reliable prediction model for earlier identification of surgical patients and also provide predictive power of Class III treatment outcomes.

Khosravi-Kamrani Pegah, Qiao Xingye, Zanardi Gustavo, Wiesen Christopher A, Slade Gary, Frazier-Bowers Sylvia A

2021-Sep-14

Surgery Surgery

Accurate diagnosis of lung tissues for 2D Raman spectrogram by deep learning based on short-time Fourier transform.

In Analytica chimica acta

Multivariate statistical analysis methods have an important role in spectrochemical analyses to rapidly identify and diagnose cancer and the subtype. However, utilizing these methods to analyze lager amount spectral data is challenging, and poses a major bottleneck toward achieving high accuracy. Here, a new convolutional neural networks (CNN) method based on short-time Fourier transform (STFT) to diagnose lung tissues via Raman spectra readily is proposed. The models yield that the accuracies of the new method are higher than the conventional methods (principal components analysis -linear discriminant analysis and support vector machine) for validation group (95.2% vs 85.5%, 94.4%) and test group (96.5% vs 90.4%, 93.9%) after cross-validation. The results illustrate that the new method which converts one-dimensional Raman data into two-dimensional Raman spectrograms improve the discriminatory ability of lung tissues and can achieve automatically accurate diagnosis of lung tissues.

Qi Yafeng, Yang Lin, Liu Bangxu, Liu Li, Liu Yuhong, Zheng Qingfeng, Liu Dameng, Luo Jianbin

2021-Sep-22

Deep learning, Lung cancer, Raman spectrogram, Short-time Fourier transform

General General

Interpreting convolutional neural network for real-time volatile organic compounds detection and classification using optical emission spectroscopy of plasma.

In Analytica chimica acta

This study presents the investigation of optical emission spectroscopy of plasma using interpretable convolutional neural network (CNN) for real-time volatile organic compounds (VOCs) classification. A microplasma-generation platform was developed to efficiently collect 64 k spectra from various types of VOCs at different concentrations, as training and testing sets for machine learning. A CNN model was trained to classify VOCs with accuracy of 99.9%. To interpret the CNN model and its predictions, the spectral processing mechanism of the CNN was visualized by feature maps and the critical spectral features were identified by gradient-weighted class activation mapping. Such approaches brought insights on how CNN analyzes the spectra and enables the CNN operation to be explainable. Finally, the CNN model was incorporated with the microplasma platform to demonstrate the application of real-time VOC monitoring. The type of VOCs can be identified and reported via messages within 10 s once the microplasma is ignited. We believe that using CNN brings a novel route for plasma spectroscopy analysis for VOC classification and impacts the fields of plasma, spectroscopy, and environmental monitoring.

Wang Ching-Yu, Ko Tsung-Shun, Hsu Cheng-Che

2021-Sep-22

Grad-CAM, Machine learning, Microplasma, Optical emission spectroscopy

General General

Study TPX-100-5: intra-articular TPX-100 significantly delays pathological bone shape change and stabilizes cartilage in moderate to severe bilateral knee OA.

In Arthritis research & therapy ; h5-index 60.0

BACKGROUND : TPX-100, a promotor of osteoblast and chondroblast differentiation, is a potential osteoarthritis (OA) therapy. This retrospective study compared MRI 3D femoral bone shape changes (B-scores) after intra-articular TPX-100 or placebo and analyzed the relationship between cartilage thickness and bone shape change over 12 months.

METHODS : One hundred and four participants with bilateral moderate to severe knee cartilage defects were randomized to receive TPX-100 (200 mg) or placebo. Each subject's contralateral placebo-treated knee served as a paired internal control. After MRI quality control, 78/93 subjects (84%; 156 knees) were analyzed for quantitative femoral B-score and cartilage thickness. All analyses were performed centrally, blind to treatment assignment and clinical data.

RESULTS : TPX-100-treated knees (n = 78) demonstrated a statistically significant decrease in pathologic bone shape change compared with placebo-treated knees at 6 and 12 months: 0.0298 (95% C.I. - 0.037, 0.097) vs 0.1246 (95% C.I. 0.067, 0.182) (P = 0.02), and 0.0856 (95% C.I. 0.013, 0.158) vs. 0.1969 (95% C.I. 0.123, 0.271) (P = 0.01), respectively. The correlation between bone shape change and medial and total tibiofemoral cartilage thickness changes at 12 months was statistically significant in TPX-100-treated knees (P < 0.01).

CONCLUSIONS : This is the first report of a potential therapy demonstrating a significant effect on bone shape measured by B-score in knee OA. These data, in combination with previously reported statistically significant and clinically meaningful improvements in WOMAC physical function versus placebo, support TPX-100 as a candidate for disease modification in knee OA.

TRIAL REGISTRATION : NIH ClinicalTrials.gov, NCT01925261 . Registered 15 August 2013.

McGuire Dawn, Bowes Michael, Brett Alan, Segal Neil A, Miller Meghan, Rosen David, Kumagai Yoshinari

2021-Sep-17

B-score, Bone shape, DMOAD, Machine learning, Osteoarthritis, TPX-100

General General

Improving the estimation of alpine grassland fractional vegetation cover using optimized algorithms and multi-dimensional features.

In Plant methods

BACKGROUND : Fractional vegetation cover (FVC) is an important basic parameter for the quantitative monitoring of the alpine grassland ecosystem on the Qinghai-Tibetan Plateau. Based on unmanned aerial vehicle (UAV) acquisition of measured data and matching it with satellite remote sensing images at the pixel scale, the proper selection of driving data and inversion algorithms can be determined and is crucial for generating high-precision alpine grassland FVC products.

METHODS : This study presents estimations of alpine grassland FVC using optimized algorithms and multi-dimensional features. The multi-dimensional feature set (using original spectral bands, 22 vegetation indices, and topographical factors) was constructed from many sources of information, then the optimal feature subset was determined based on different feature selection algorithms as the driving data for optimized machine learning algorithms. Finally, the inversion accuracy, sensitivity to sample size, and computational efficiency of the four machine learning algorithms were evaluated.

RESULTS : (1) The random forest (RF) algorithm (R2: 0.861, RMSE: 9.5%) performed the best for FVC inversion among the four machine learning algorithms driven by the four typical vegetation indices. (2) Compared with the four typical vegetation indices, using multi-dimensional feature sets as driving data obviously improved the FVC inversion accuracy of the four machine learning algorithms (R2 of the RF algorithm increased to 0.890). (3) Among the three variable selection algorithms (Boruta, sequential forward selection [SFS], and permutation importance-recursive feature elimination [PI-RFE]), the constructed PI-RFE feature selection algorithm had the best dimensionality reduction effect on the multi-dimensional feature set. (4) The hyper-parameter optimization of the machine learning algorithms and feature selection of the multi-dimensional feature set further improved FVC inversion accuracy (R2: 0.917 and RMSE: 7.9% in the optimized RF algorithm).

CONCLUSION : This study provides a highly precise, optimized algorithm with an optimal multi-dimensional feature set for FVC inversion, which is vital for the quantitative monitoring of the ecological environment of alpine grassland.

Lin Xingchen, Chen Jianjun, Lou Peiqing, Yi Shuhua, Qin Yu, You Haotian, Han Xiaowen

2021-Sep-17

Accuracy evaluation, Alpine grassland, Feature selection, Fractional vegetation cover (FVC), Machine learning algorithms, Parameter tuning, Unmanned aerial vehicle (UAV) aerial imagery

General General

Utility and usability of an automated COVID-19 symptom monitoring system (CoSMoS) in primary care during COVID-19 pandemic: A qualitative feasibility study.

In International journal of medical informatics ; h5-index 49.0

BACKGROUND : COVID-19 telemonitoring applications have been developed and used in primary care to monitor patients quarantined at home. There is a lack of evidence on the utility and usability of telemonitoring applications from end-users' perspective.

OBJECTIVES : This study aimed to evaluate the feasibility of a COVID-19 symptom monitoring system (CoSMoS) by exploring its utility and usability with end-users.

METHODS : This was a qualitative study using in-depth interviews. Patients with suspected COVID-19 infection who used CoSMoS Telegram bot to monitor their COVID-19 symptoms and doctors who conducted the telemonitoring via CoSMoS dashboard were recruited. Universal sampling was used in this study. We stopped the recruitment when data saturation was reached. Patients and doctors shared their experiences using CoSMoS, its utility and usability for COVID-19 symptoms monitoring. Data were coded and analysed using thematic analysis.

RESULTS : A total of 11 patients and 4 doctors were recruited into this study. For utility, CoSMoS was useful in providing close monitoring and continuity of care, supporting patients' decision making, ensuring adherence to reporting, and reducing healthcare workers' burden during the pandemic. In terms of usability, patients expressed that CoSMoS was convenient and easy to use. The use of the existing social media application for symptom monitoring was acceptable for the patients. The content in the Telegram bot was easy to understand, although revision was needed to keep the content updated. Doctors preferred to integrate CoSMoS into the electronic medical record.

CONCLUSION : CoSMoS is feasible and useful to patients and doctors in providing remote monitoring and teleconsultation during the COVID-19 pandemic. The utility and usability evaluation enables the refinement of CoSMoS to be a patient-centred monitoring system.

Lim Hooi Min, Abdullah Adina, Ng Chirk Jenn, Teo Chin Hai, Valliyappan Indra Gayatri, Abdul Hadi Haireen, Ng Wei Leik, Noor Azhar Abdul Muhaimin, Chiew Thiam Kian, Liew Chee Sun, Chan Chee Seng

2021-Sep-06

COVID-19, Digital health, Monitoring system, Telemonitoring, Usability, Utility

Cardiology Cardiology

Identifying Ventricular Arrhythmias and Their Predictors by Applying Machine Learning Methods to Electronic Health Records in Patients With Hypertrophic Cardiomyopathy(HCM-VAr-Risk Model)

The American Journal of Cardiology, Volume 123, Issue 10, 15 May 2019, Pages 1681-1689

Clinical risk stratification for sudden cardiac death (SCD) in hypertrophic cardiomyopathy (HC) employs rules derived from American College of Cardiology Foundation/American Heart Association (ACCF/AHA) guidelines or the HCM Risk-SCD model (C-index of 0.69), which utilize a few clinical variables. We assessed whether data-driven machine learning methods that consider a wider range of variables can effectively identify HC patients with ventricular arrhythmias (VAr) that lead to SCD. We scanned the electronic health records of 711 HC patients for sustained ventricular tachycardia or ventricular fibrillation. Patients with ventricular tachycardia or ventricular fibrillation (n = 61) were tagged as VAr cases and the remaining (n = 650) as non-VAr. The 2-sample t test and information gain criterion were used to identify the most informative clinical variables that distinguish VAr from non-VAr; patient records were reduced to include only these variables. Data imbalance stemming from low number of VAr cases was addressed by applying a combination of over- and under-sampling strategies.We trained and tested multiple classifiers under this sampling approach, showing effective classification. We evaluated 93 clinical variables, of which 22 proved predictive of VAr. The ensemble of logistic regression and naive Bayes classifiers, trained based on these 22 variables and corrected for data imbalance, was most effective in separating VAr from non-VAr cases (sensitivity = 0.73, specificity = 0.76, C-index = 0.83). Our method (HCM-VAr-Risk Model) identified 12 new predictors of VAr, in addition to 10 established SCD predictors. In conclusion, this is the first application of machine learning for identifying HC patients with VAr, using clinical attributes.

Moumita Bhattacharya, Dai-Yin Lu, Shibani M Kudchadkar, Gabriela Villarreal Greenland, Prasanth Lingamaneni, Celia P Corona-Villalobos, Yufan Guan, Joseph E Marine, Jeffrey E Olgin, Stefan Zimmerman, Theodore P Abraham, Hagit Shatkay, Maria Roselle Abraham

2021-09-19

Cardiology Cardiology

The smell of lung disease: a review of the current status of electronic nose technology.

In Respiratory research ; h5-index 45.0

There is a need for timely, accurate diagnosis, and personalised management in lung diseases. Exhaled breath reflects inflammatory and metabolic processes in the human body, especially in the lungs. The analysis of exhaled breath using electronic nose (eNose) technology has gained increasing attention in the past years. This technique has great potential to be used in clinical practice as a real-time non-invasive diagnostic tool, and for monitoring disease course and therapeutic effects. To date, multiple eNoses have been developed and evaluated in clinical studies across a wide spectrum of lung diseases, mainly for diagnostic purposes. Heterogeneity in study design, analysis techniques, and differences between eNose devices currently hamper generalization and comparison of study results. Moreover, many pilot studies have been performed, while validation and implementation studies are scarce. These studies are needed before implementation in clinical practice can be realised. This review summarises the technical aspects of available eNose devices and the available evidence for clinical application of eNose technology in different lung diseases. Furthermore, recommendations for future research to pave the way for clinical implementation of eNose technology are provided.

van der Sar I G, Wijbenga N, Nakshbandi G, Aerts J G J V, Manintveld O C, Wijsenbeek M S, Hellemons M E, Moor C C

2021-Sep-17

Breath analysis, Electronic nose, Machine learning, Personalised medicine, Respiratory medicine, Sensor technology

Radiology Radiology

DR-MIL: deep represented multiple instance learning distinguishes COVID-19 from community-acquired pneumonia in CT images.

In Computer methods and programs in biomedicine

BACKGROUND AND OBJECTIVE : Given that the novel coronavirus disease 2019 (COVID-19) has become a pandemic, a method to accurately distinguish COVID-19 from community-acquired pneumonia (CAP) is urgently needed. However, the spatial uncertainty and morphological diversity of COVID-19 lesions in the lungs, and subtle differences with respect to CAP, make differential diagnosis non-trivial.

METHODS : We propose a deep represented multiple instance learning (DR-MIL) method to fulfill this task. A 3D volumetric CT scan of one patient is treated as one bag and ten CT slices are selected as the initial instances. For each instance, deep features are extracted from the pre-trained ResNet-50 with fine-tuning and represented as one deep represented instance score (DRIS). Each bag with a DRIS for each initial instance is then input into a citation k-nearest neighbor search to generate the final prediction. A total of 141 COVID-19 and 100 CAP CT scans were used. The performance of DR-MIL is compared with other potential strategies and state-of-the-art models.

RESULTS : DR-MIL displayed an accuracy of 95% and an area under curve of 0.943, which were superior to those observed for comparable methods. COVID-19 and CAP exhibited significant differences in both the DRIS and the spatial pattern of lesions (p<0.001). As a means of content-based image retrieval, DR-MIL can identify images used as key instances, references, and citers for visual interpretation.

CONCLUSIONS : DR-MIL can effectively represent the deep characteristics of COVID-19 lesions in CT images and accurately distinguish COVID-19 from CAP in a weakly supervised manner. The resulting DRIS is a useful supplement to visual interpretation of the spatial pattern of lesions when screening for COVID-19.

Qi Shouliang, Xu Caiwen, Li Chen, Tian Bin, Xia Shuyue, Ren Jigang, Yang Liming, Wang Hanlin, Yu Hui

2021-Sep-09

COVID-19, Community-acquired pneumonia, Convolutional neural network, Deep learning, Lung CT image, Multiple instance learning

General General

Machine Learning Methods for Identifying Atrial Fibrillation Cases and Their Predictors in Patients With Hypertrophic Cardiomyopathy: The HCM-AF-Risk Model

CJC Open, Volume 3, Issue 6, June 2021, Pages 801-813

Hypertrophic cardiomyopathy (HCM) patients have a high incidence of atrial fibrillation (AF) and increased stroke risk, even with low risk of congestive heart failure, hypertension, age, diabetes, previous stroke/transient ischemic attack scores. Hence, there is a need to understand the pathophysiology of AF and stroke in HCM. In this retrospective study, we develop and apply a data-driven, machine learning based method to identify AF cases, and clinical and imaging features associated with AF, using electronic health record data. HCM patients with documented paroxysmal/persistent/permanent AF (n = 191) were considered AF cases, and the remaining patients in sinus rhythm (n = 640) were tagged as No-AF. We evaluated 93 clinical variables and the most informative variables useful for distinguishing AF from No-AF cases were selected based on the 2-sample t test and the information gain criterion. We identified 18 highly informative variables that are positively (n = 11) and negatively (n = 7) correlated with AF in HCM. Next, patient records were represented via these 18 variables. Data imbalance resulting from the relatively low number of AF cases was addressed via a combination of oversampling and under-sampling strategies. We trained and tested multiple classifiers under this sampling approach, showing effective classification. Specifically, an ensemble of logistic regression and naive Bayes classifiers, trained based on the 18 variables and corrected for data imbalance, proved most effective for separating AF from No-AF cases (sensitivity = 0.74, specificity = 0.70, C-index = 0.80). Our model is the first machine learning based method for identification of AF cases in HCM. This model demonstrates good performance, addresses data imbalance, and suggests that AF is associated with a more severe cardiac HCM phenotype.

Moumita Bhattacharya, Dai-Yin Lu, Ioannis Ventoulis, Gabriela V. Greenland, Hulya Yalcin, Yufan Guan, Joseph E. Marine, Jeffrey E. Olgin, Stefan L. Zimmerman, Theodore P. Abraham, M. Roselle Abraham, Hagit Shatkay

2021-09-19

Radiology Radiology

Artificial intelligence-based hybrid deep learning models for image classification: The first narrative review.

In Computers in biology and medicine

BACKGROUND : Artificial intelligence (AI) has served humanity in many applications since its inception. Currently, it dominates the imaging field-in particular, image classification. The task of image classification became much easier with machine learning (ML) and subsequently got automated and more accurate by using deep learning (DL). By default, DL consists of a single architecture and is termed solo deep learning (SDL). When two or more DL architectures are fused, the result is termed a hybrid deep learning (HDL) model. The use of HDL models is becoming popular in several applications, but no review of these uses has been designed thus far. Therefore, this study provides the first narrative HDL review by considering all facets of image classification using AI.

APPROACH : Our review employs a PRISMA search strategy using Google Scholar, PubMed, IEEE, and Elsevier Science Direct, through which 127 relevant HDL studies were considered. Based on the computer vision evolution, HDLs were subsequently classified into three categories (spatial, temporal, and spatial-temporal). Each study was then analyzed based on several attributes, including continent, publisher, hybridization of two DL or ML, architecture layout, application type, data set type, dataset size, feature extraction methodology, connecting classifier, performance evaluation metrics, and risk-of-bias.

CONCLUSION : The HDL models have shown stable and superior performance by taking the best aspects of two or more solo DL or fusion of DL with ML models. Our findings indicate that HDL is being applied aggressively to several medical and non-medical applications. Furthermore, risk-of-bias is highly debatable for DL and HDL models.

Jena Biswajit, Saxena Sanjay, Nayak Gopal K, Saba Luca, Sharma Neeraj, Suri Jasjit S

2021-Aug-27

Artificial intelligence, Deep learning, Hybrid deep learning, Performance, Risk-of-bias, Spatial, Spatial-temporal, Temporal

General General

Utility and usability of an automated COVID-19 symptom monitoring system (CoSMoS) in primary care during COVID-19 pandemic: A qualitative feasibility study.

In International journal of medical informatics ; h5-index 49.0

BACKGROUND : COVID-19 telemonitoring applications have been developed and used in primary care to monitor patients quarantined at home. There is a lack of evidence on the utility and usability of telemonitoring applications from end-users' perspective.

OBJECTIVES : This study aimed to evaluate the feasibility of a COVID-19 symptom monitoring system (CoSMoS) by exploring its utility and usability with end-users.

METHODS : This was a qualitative study using in-depth interviews. Patients with suspected COVID-19 infection who used CoSMoS Telegram bot to monitor their COVID-19 symptoms and doctors who conducted the telemonitoring via CoSMoS dashboard were recruited. Universal sampling was used in this study. We stopped the recruitment when data saturation was reached. Patients and doctors shared their experiences using CoSMoS, its utility and usability for COVID-19 symptoms monitoring. Data were coded and analysed using thematic analysis.

RESULTS : A total of 11 patients and 4 doctors were recruited into this study. For utility, CoSMoS was useful in providing close monitoring and continuity of care, supporting patients' decision making, ensuring adherence to reporting, and reducing healthcare workers' burden during the pandemic. In terms of usability, patients expressed that CoSMoS was convenient and easy to use. The use of the existing social media application for symptom monitoring was acceptable for the patients. The content in the Telegram bot was easy to understand, although revision was needed to keep the content updated. Doctors preferred to integrate CoSMoS into the electronic medical record.

CONCLUSION : CoSMoS is feasible and useful to patients and doctors in providing remote monitoring and teleconsultation during the COVID-19 pandemic. The utility and usability evaluation enables the refinement of CoSMoS to be a patient-centred monitoring system.

Lim Hooi Min, Abdullah Adina, Ng Chirk Jenn, Teo Chin Hai, Valliyappan Indra Gayatri, Abdul Hadi Haireen, Ng Wei Leik, Noor Azhar Abdul Muhaimin, Chiew Thiam Kian, Liew Chee Sun, Chan Chee Seng

2021-Sep-06

COVID-19, Digital health, Monitoring system, Telemonitoring, Usability, Utility

General General

Reducing variability of breast cancer subtype predictors by grounding deep learning models in prior knowledge.

In Computers in biology and medicine

Deep learning neural networks have improved performance in many cancer informatics problems, including breast cancer subtype classification. However, many networks experience underspecificationwheremultiplecombinationsofparametersachievesimilarperformance, bothin training and validation. Additionally, certain parameter combinations may perform poorly when the test distribution differs from the training distribution. Embedding prior knowledge from the literature may address this issue by boosting predictive models that provide crucial, in-depth information about a given disease. Breast cancer research provides a wealth of such knowledge, particularly in the form of subtype biomarkers and genetic signatures. In this study, we draw on past research on breast cancer subtype biomarkers, label propagation, and neural graph machines to present a novel methodology for embedding knowledge into machine learning systems. We embed prior knowledge into the loss function in the form of inter-subject distances derived from a well-known published breast cancer signature. Our results show that this methodology reduces predictor variability on state-of-the-art deep learning architectures and increases predictor consistency leading to improved interpretation. We find that pathway enrichment analysis is more consistent after embedding knowledge. This novel method applies to a broad range of existing studies and predictive models. Our method moves the traditional synthesis of predictive models from an arbitrary assignment of weights to genes toward a more biologically meaningful approach of incorporating knowledge.

Anderson Paul, Gadgil Richa, Johnson William A, Schwab Ella, Davidson Jean M

2021-Sep-10

Applied computing, Bioinformatics, Genomics, Transcriptomics

General General

In-depth learning of automatic segmentation of shoulder joint magnetic resonance images based on convolutional neural networks.

In Computer methods and programs in biomedicine

OBJECTIVE : Magnetic resonance imaging (MRI) is gradually replacing computed tomography (CT) in the examination of bones and joints. The accurate and automatic segmentation of the bone structure in the MRI of the shoulder joint is essential for the measurement and diagnosis of bone injuries and diseases. The existing bone segmentation algorithms cannot achieve automatic segmentation without any prior knowledge, and their versatility and accuracy are relatively low. For this reason, an automatic segmentation algorithm based on the combination of image blocks and convolutional neural networks is proposed.

METHODS : First, we establish 4 segmentation models, including 3 U-Net-based bone segmentation models (humeral segmentation model, joint bone segmentation model, humeral head and articular bone segmentation model as a whole) and a block-based Alex Net segmentation model; Then we use 4 segmentation models to obtain the candidate bone area, and accurately detect the location area of the humerus and joint bone by voting. Finally, the Alex Net segmentation model is further used in the detected bone area to segment the bone edge with the accuracy of the pixel level.

RESULTS : The experimental data is obtained from 8 groups of patients in the orthopedics department of our hospital. Each group of scan sequence includes about 100 images, which have been segmented and labeled. Five groups of patients were used for training and five-fold cross-validation, and three groups of patients were used to test the actual segmentation effect. The average accuracy of Dice Coefficient, Positive Predicted Value (PPV) and Sensitivity reached 0.91 ± 0.02, respectively. 0.95 ± 0.03 and 0.95 ± 0.02.

CONCLUSIONS : The method in this paper is for a small sample of patient data sets, and only through deep learning on 2D medical images, very accurate shoulder joint segmentation results can be obtained, provide clinical diagnostic guidance to orthopedics. At the same time, the proposed algorithm framework has a certain versatility and is suitable for the precise segmentation of specific organs and tissues in MRI based on a small sample data.

Mu Xinhong, Cui Yi, Bian Rongpeng, Long Long, Zhang Daliang, Wang Huawen, Shen Yidong, Wu Jingjing, Zou Guoyou

2021-Jul-31

Convolutional neural network, Deep learning, Magnetic resonance imaging, Medical image segmentation, Orthopedic diagnosis

Radiology Radiology

DR-MIL: deep represented multiple instance learning distinguishes COVID-19 from community-acquired pneumonia in CT images.

In Computer methods and programs in biomedicine

BACKGROUND AND OBJECTIVE : Given that the novel coronavirus disease 2019 (COVID-19) has become a pandemic, a method to accurately distinguish COVID-19 from community-acquired pneumonia (CAP) is urgently needed. However, the spatial uncertainty and morphological diversity of COVID-19 lesions in the lungs, and subtle differences with respect to CAP, make differential diagnosis non-trivial.

METHODS : We propose a deep represented multiple instance learning (DR-MIL) method to fulfill this task. A 3D volumetric CT scan of one patient is treated as one bag and ten CT slices are selected as the initial instances. For each instance, deep features are extracted from the pre-trained ResNet-50 with fine-tuning and represented as one deep represented instance score (DRIS). Each bag with a DRIS for each initial instance is then input into a citation k-nearest neighbor search to generate the final prediction. A total of 141 COVID-19 and 100 CAP CT scans were used. The performance of DR-MIL is compared with other potential strategies and state-of-the-art models.

RESULTS : DR-MIL displayed an accuracy of 95% and an area under curve of 0.943, which were superior to those observed for comparable methods. COVID-19 and CAP exhibited significant differences in both the DRIS and the spatial pattern of lesions (p<0.001). As a means of content-based image retrieval, DR-MIL can identify images used as key instances, references, and citers for visual interpretation.

CONCLUSIONS : DR-MIL can effectively represent the deep characteristics of COVID-19 lesions in CT images and accurately distinguish COVID-19 from CAP in a weakly supervised manner. The resulting DRIS is a useful supplement to visual interpretation of the spatial pattern of lesions when screening for COVID-19.

Qi Shouliang, Xu Caiwen, Li Chen, Tian Bin, Xia Shuyue, Ren Jigang, Yang Liming, Wang Hanlin, Yu Hui

2021-Sep-09

COVID-19, Community-acquired pneumonia, Convolutional neural network, Deep learning, Lung CT image, Multiple instance learning

General General

The emergence of machine learning in auditory neural impairment: a systematic review.

In Neuroscience letters

Hearing loss is a common neurodegenerative disease that can start at any stage of life. Misalignment of the auditory neural impairment may impose challenges in processing incoming auditory stimulus that can be measured using electroencephalography (EEG). The electrophysiological behaviour response emanated from EEG auditory evoked potential (AEP) requires highly trained professionals for analysis and interpretation. Reliable automated methods using techniques of machine learning would assist the auditory assessment process for informed treatment and practice. It is thus highly required to develop models that are more efficient and precise by considering the characteristics of brain signals. This study aims to provide a comprehensive review of several state-of-the-art techniques of machine learning that adopt EEG evoked response for the auditory assessment within the last 13 years. Out of 161 initially screened articles, 11 were retained for synthesis. The outcome of the review presented that the Support Vector Machine (SVM) classifier outperformed with over 80% accuracy metric and was recognized as the best suited model within the field of auditory research. This paper discussed the comprehensive iterative properties of the proposed computed algorithms and the feasible future direction in hearing impaired rehabilitation.

Rauf A Bakar Abdul, Wee Lai Khin, Azah Hamzaid Nur

2021-Sep-15

Auditory Evoked Potential (AEP), Electroencephalography (EEG), Machine learning, auditory assessment, classification

General General

Automated assignment of cell identity from single-cell multiplexed imaging and proteomic data.

In Cell systems

A major challenge in the analysis of highly multiplexed imaging data is the assignment of cells to a priori known cell types. Existing approaches typically solve this by clustering cells followed by manual annotation. However, these often require several subjective choices and cannot explicitly assign cells to an uncharacterized type. To help address these issues we present Astir, a probabilistic model to assign cells to cell types by integrating prior knowledge of marker proteins. Astir uses deep recognition neural networks for fast inference, allowing for annotations at the million-cell scale in the absence of a previously annotated reference. We apply Astir to over 2.4 million cells from suspension and imaging datasets and demonstrate its scalability, robustness to sample composition, and interpretable uncertainty estimates. We envision deployment of Astir either for a first broad cell type assignment or to accurately annotate cells that may serve as biomarkers in multiple disease contexts. A record of this paper's transparent peer review process is included in the supplemental information.

Geuenich Michael J, Hou Jinyu, Lee Sunyun, Ayub Shanza, Jackson Hartland W, Campbell Kieran R

2021-Sep-13

automated analysis, computational biology, data analysis, highly mutliplexed imaging, machine learning, tumor microenvironment

General General

D-SCRIPT translates genome to phenome with sequence-based, structure-aware, genome-scale predictions of protein-protein interactions.

In Cell systems

We combine advances in neural language modeling and structurally motivated design to develop D-SCRIPT, an interpretable and generalizable deep-learning model, which predicts interaction between two proteins using only their sequence and maintains high accuracy with limited training data and across species. We show that a D-SCRIPT model trained on 38,345 human PPIs enables significantly improved functional characterization of fly proteins compared with the state-of-the-art approach. Evaluating the same D-SCRIPT model on protein complexes with known 3D structure, we find that the inter-protein contact map output by D-SCRIPT has significant overlap with the ground truth. We apply D-SCRIPT to screen for PPIs in cow (Bos taurus) at a genome-wide scale and focusing on rumen physiology, identify functional gene modules related to metabolism and immune response. The predicted interactions can then be leveraged for function prediction at scale, addressing the genome-to-phenome challenge, especially in species where little data are available.

Sledzieski Samuel, Singh Rohit, Cowen Lenore, Berger Bonnie

2021-Sep-14

cow rumen, deep learning, embedding, function prediction, genome to phenome, interpretability, language models, metabolism, module detection, protein-protein interaction

Surgery Surgery

Fully Automated Measurement of Cochlear Duct Length From Clinical Temporal Bone Computed Tomography.

In The Laryngoscope ; h5-index 55.0

OBJECTIVES/HYPOTHESIS : To present and validate a novel fully automated method to measure cochlear dimensions, including cochlear duct length (CDL).

STUDY DESIGN : Cross-sectional study.

METHODS : The computational method combined 1) a deep learning (DL) algorithm to segment the cochlea and otic capsule and 2) geometric analysis to measure anti-modiolar distances from the round window to the apex. The algorithm was trained using 165 manually segmented clinical computed tomography (CT). A Testing group of 159 CTs were then measured for cochlear diameter and width (A- and B-values) and CDL using the automated system and compared against manual measurements. The results were also compared with existing approaches and historical data. In addition, pre- and post-implantation scans from 27 cochlear implant recipients were studied to compare predicted versus actual array insertion depth.

RESULTS : Measurements were successfully obtained in 98.1% of scans. The mean CDL to 900° was 35.52 mm (SD, 2.06; range, [30.91-40.50]), the mean A-value was 8.88 mm (0.47; [7.67-10.49]), and mean B-value was 6.38 mm (0.42; [5.16-7.38]). The R2 fit of the automated to manual measurements was 0.87 for A-value, 0.70 for B-value, and 0.71 for CDL. For anti-modiolar arrays, the distance between the imaged and predicted array tip location was 0.57 mm (1.25; [0.13-5.28]).

CONCLUSION : Our method provides a fully automated means of cochlear analysis from clinical CTs. The distribution of CDL, dimensions, and cochlear quadrant lengths is similar to those from historical data. This approach requires no radiographic experience and is free from user-related variation.

LEVEL OF EVIDENCE : 3 Laryngoscope, 2021.

Neves Caio A, Tran Emma D, Cooperman Shayna P, Blevins Nikolas H

2021-Sep-18

Cochlear duct length, cochlear implantation, deep learning, segmentation

General General

Estimation of postpartum depression risk from electronic health records using machine learning.

In BMC pregnancy and childbirth ; h5-index 58.0

BACKGROUND : Postpartum depression is a widespread disorder, adversely affecting the well-being of mothers and their newborns. We aim to utilize machine learning for predicting risk of postpartum depression (PPD) using primary care electronic health records (EHR) data, and to evaluate the potential value of EHR-based prediction in improving the accuracy of PPD screening and in early identification of women at risk.

METHODS : We analyzed EHR data of 266,544 women from the UK who gave first live birth between 2000 and 2017. We extracted a multitude of socio-demographic and medical variables and constructed a machine learning model that predicts the risk of PPD during the year following childbirth. We evaluated the model's performance using multiple validation methodologies and measured its accuracy as a stand-alone tool and as an adjunct to the standard questionnaire-based screening by Edinburgh postnatal depression scale (EPDS).

RESULTS : The prevalence of PPD in the analyzed cohort was 13.4%. Combing EHR-based prediction with EPDS score increased the area under the receiver operator characteristics curve (AUC) from 0.805 to 0.844 and the sensitivity from 0.72 to 0.76, at specificity of 0.80. The AUC of the EHR-based prediction model alone varied from 0.72 to 0.74 and decreased by only 0.01-0.02 when applied as early as before the beginning of pregnancy.

CONCLUSIONS : PPD risk prediction using EHR data may provide a complementary quantitative and objective tool for PPD screening, allowing earlier (pre-pregnancy) and more accurate identification of women at risk, timely interventions and potentially improved outcomes for the mother and child.

Amit Guy, Girshovitz Irena, Marcus Karni, Zhang Yiye, Pathak Jyotishman, Bar Vered, Akiva Pinchas

2021-Sep-17

Electronic health records, Machine learning, Postpartum depression

General General

Generating insights in uncharted territories: real-time learning from data in critically ill patients-an implementer report.

In BMJ health & care informatics

Introduction In the current situation, clinical patient data are often siloed in multiple hospital information systems. Especially in the intensive care unit (ICU), large volumes of clinical data are routinely collected through continuous patient monitoring. Although these data often contain useful information for clinical decision making, they are not frequently used to improve quality of care. During, but also after, pressing times, data-driven methods can be used to mine treatment patterns from clinical data to determine the best treatment options from a hospitals own clinical data.Methods In this implementer report, we describe how we implemented a data infrastructure that enabled us to learn in real time from consecutive COVID-19 ICU admissions. In addition, we explain our step-by-step multidisciplinary approach to establish such a data infrastructure.Conclusion By sharing our steps and approach, we aim to inspire others, in and outside ICU walls, to make more efficient use of data at hand, now and in the future.

van de Sande Davy, Van Genderen Michel E, Huiskens Joost, Veen Robert E R, Meijerink Yvonne, Gommers Diederik, van Bommel Jasper

2021-Sep

COVID-19, artificial intelligence, critical care outcomes, data science, machine learning

General General

Co-occurrence of medical conditions: Exposing patterns through probabilistic topic modeling of SNOMED codes

Journal of Biomedical Informatics Volume 82, June 2018, Pages 31-40

Patients associated with multiple co-occurring health conditions often face aggravated complications and less favorable outcomes. Co-occurring conditions are especially prevalent among individuals suffering from kidney disease, an increasingly widespread condition affecting 13% of the general population in the US. This study aims to identify and characterize patterns of co-occurring medical conditions in patients employing a probabilistic framework. Specifically, we apply topic modeling in a non-traditional way to find associations across SNOMEDCT codes assigned and recorded in the EHRs of>13,000 patients diagnosed with kidney disease. Unlike most prior work on topic modeling, we apply the method to codes rather than to natural language. Moreover, we quantitatively evaluate the topics, assessing their tightness and distinctiveness, and also assess the medical validity of our results. Our experiments show that each topic is succinctly characterized by a few highly probable and unique disease codes, indicating that the topics are tight. Furthermore, inter-topic distance between each pair of topics is typically high, illustrating distinctiveness. Last, most coded conditions grouped together within a topic, are indeed reported to co-occur in the medical literature. Notably, our results uncover a few indirect associations among conditions that have hitherto not been reported as correlated in the medical literature.

Moumita Bhattacharya, Claudine Jurkovitz, Hagit Shatkay

2021-09-19

Ophthalmology Ophthalmology

OTNet: A CNN Method Based on Hierarchical Attention Maps for Grading Arteriosclerosis of Fundus Images with Small Samples.

In Interdisciplinary sciences, computational life sciences

The severity of fundus arteriosclerosis can be determined and divided into four grades according to fundus images. Automatically grading of the fundus arteriosclerosis is helpful in clinical practices, so this paper proposes a convolutional neural network (CNN) method based on hierarchical attention maps to solve the automatic grading problem. First, we use the retinal vessel segmentation model to separate the important vascular region and the non-vascular background region from the fundus image and obtain two attention maps. The two maps are regarded as inputs to construct a two-stream CNN (TSNet), to focus on feature information through mutual reference between the two regions. In addition, we use convex hull attention maps in the one-stream CNN (OSNet) to learn valuable areas where the retinal vessels are concentrated. Then, we design an integrated OTNet model which is composed of TSNet that learns image feature information and OSNet that learns discriminative areas. After obtaining the representation learning parts of the two networks, we can train the classification layer to achieve better results. Our proposed TSNet reaches the AUC value of 0.796 and the ACC value of 0.592 on the testing set, and the integrated model OTNet reaches the AUC value of 0.806 and the ACC value of 0.606, which are better than the results of other comparable models. As far as we know, this is the first attempt to use deep learning to classify the severity of atherosclerosis in fundus images. The prediction results of our proposed method can be accepted by doctors, which shows that our method has a certain application value.

Bai Hang, Gao Li, Quan Xiongwen, Zhang Han, Gao Shuo, Kang Chuanze, Qi Jiaqiang

2021-Sep-18

Fine-grained visual classification, Fundus arteriosclerosis, Hierarchical attention maps, Multi-stream CNN, Retinal vessel segmentation

General General

Should there be an "AI" in TEAM? Embryologists selection of high implantation potential embryos improves with the aid of an artificial intelligence algorithm.

In Journal of assisted reproduction and genetics ; h5-index 39.0

PURPOSE : A deep learning artificial intelligence (AI) algorithm has been demo