Cardiology

Fully Automatic Atrial Fibrosis Assessment Using a Multilabel Convolutional Neural Network.

In Circulation. Cardiovascular imaging

BACKGROUND : Pathological atrial fibrosis is a major contributor to sustained atrial fibrillation. Currently, late gadolinium enhancement (LGE) scans provide the only noninvasive estimate of atrial fibrosis. However, widespread adoption of atrial LGE has been hindered partly by nonstandardized image processing techniques, which can be operator and algorithm dependent. Minimal validation and limited access to transparent software platforms have also exacerbated the problem. This study aims to estimate atrial fibrosis from cardiac magnetic resonance scans using a reproducible operator-independent fully automatic open-source end-to-end pipeline.

METHODS : A multilabel convolutional neural network was designed to accurately delineate atrial structures including the blood pool, pulmonary veins, and mitral valve. The output from the network removed the operator dependent steps in a reproducible pipeline and allowed for automated estimation of atrial fibrosis from LGE-cardiac magnetic resonance scans. The pipeline results were compared against manual fibrosis burdens, calculated using published thresholds: image intensity ratio 0.97, image intensity ratio 1.61, and mean blood pool signal +3.3 SD.

RESULTS : We validated our methods on a large 3-dimensional LGE-cardiac magnetic resonance data set from 207 labeled scans. Automatic atrial segmentation achieved a 91% Dice score, compared with the mutual agreement of 85% in Dice seen in the interobserver analysis of operators. Intraclass correlation coefficients of the automatic pipeline with manually generated results were excellent and better than or equal to interobserver correlations for all 3 thresholds: 0.94 versus 0.88, 0.99 versus 0.99, 0.99 versus 0.96 for image intensity ratio 0.97, image intensity ratio 1.61, and +3.3 SD thresholds, respectively. Automatic analysis required 3 minutes per case on a standard workstation. The network and the analysis software are publicly available.

CONCLUSIONS : Our pipeline provides a fully automatic estimation of fibrosis burden from LGE-cardiac magnetic resonance scans that is comparable to manual analysis. This removes one key source of variability in the measurement of atrial fibrosis.

Razeghi Orod, Sim Iain, Roney Caroline H, Karim Rashed, Chubb Henry, Whitaker John, O’Neill Louisa, Mukherjee Rahul, Wright Matthew, O’Neill Mark, Williams Steven E, Niederer Steven

2020-Dec

atrial fibrillation, deep learning, fibrosis, magnetic resonance imaging

General

Predicted Cellular Immunity Population Coverage Gaps for SARS-CoV-2 Subunit Vaccines and Their Augmentation by Compact Peptide Sets.

In Cell systems

Subunit vaccines induce immunity to a pathogen by presenting a component of the pathogen and thus inherently limit the representation of pathogen peptides for cellular immunity-based memory. We find that severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) subunit peptides may not be robustly displayed by the major histocompatibility complex (MHC) molecules in certain individuals. We introduce an augmentation strategy for subunit vaccines that adds a small number of SARS-CoV-2 peptides to a vaccine to improve the population coverage of pathogen peptide display. Our population coverage estimates integrate clinical data on peptide immunogenicity in convalescent COVID-19 patients and machine learning predictions. We evaluate the population coverage of 9 different subunits of SARS-CoV-2, including 5 functional domains and 4 full proteins, and augment each of them to fill a predicted coverage gap.

Liu Ge, Carter Brandon, Gifford David K

2020-Nov-27

SARS-CoV-2, combinatorial optimization, haplotype, machine learning, major histocompatibility complex, peptide vaccine, population coverage, subunit, vaccine augmentation, vaccine evaluation

General

Gut Colonization Mechanisms of Lactobacillus and Bifidobacterium: An Argument for Personalized Designs.

In Annual review of food science and technology

Lactobacillus and Bifidobacterium spp. are best understood for their applications as probiotics, which are often transient, but as commensals it is probable that stable colonization in the gut is important for their beneficial roles. Recent research suggests that the establishment and persistence of strains of Lactobacillus and Bifidobacterium in the gut are species- and strain-specific and affected by natural history, genomic adaptability, and metabolic interactions of the bacteria and the microbiome and immune aspects of the host but also regulated by diet. This provides new perspectives on the underlying molecular mechanisms. With an emphasis on host-microbe interaction, this review outlines how the characteristics of individual Lactobacillus and Bifidobacterium bacteria, the host genotype and microbiome structure, diet, and host-microbe coadaptation during bacterial gut transition determine and influence the colonization process. The diet-tuned and personally tailored colonization can be achieved via a machine learning prediction model proposed here. Expected final online publication date for the Annual Review of Food Science and Technology, Volume 12 is March 2021. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Xiao Yue, Zhai Qixiao, Zhang Hao, Chen Wei, Hill Colin

2020-Dec-11

General

Optimised genetic algorithm-extreme learning machine approach for automatic COVID-19 detection.

In PloS one ; h5-index 176.0

The coronavirus disease (COVID-19), is an ongoing global pandemic caused by severe acute respiratory syndrome. Chest Computed Tomography (CT) is an effective method for detecting lung illnesses, including COVID-19. However, the CT scan is expensive and time-consuming. Therefore, this work focus on detecting COVID-19 using chest X-ray images because it is widely available, faster, and cheaper than CT scan. Many machine learning approaches such as Deep Learning, Neural Network, and Support Vector Machine; have used X-ray for detecting the COVID-19. Although the performance of those approaches is acceptable in terms of accuracy, however, they require high computational time and more memory space. Therefore, this work employs an Optimised Genetic Algorithm-Extreme Learning Machine (OGA-ELM) with three selection criteria (i.e., random, K-tournament, and roulette wheel) to detect COVID-19 using X-ray images. The most crucial strength factors of the Extreme Learning Machine (ELM) are: (i) high capability of the ELM in avoiding overfitting; (ii) its usability on binary and multi-type classifiers; and (iii) ELM could work as a kernel-based support vector machine with a structure of a neural network. These advantages make the ELM efficient in achieving an excellent learning performance. ELMs have successfully been applied in many domains, including medical domains such as breast cancer detection, pathological brain detection, and ductal carcinoma in situ detection, but not yet tested on detecting COVID-19. Hence, this work aims to identify the effectiveness of employing OGA-ELM in detecting COVID-19 using chest X-ray images. In order to reduce the dimensionality of a histogram oriented gradient features, we use principal component analysis. The performance of OGA-ELM is evaluated on a benchmark dataset containing 188 chest X-ray images with two classes: a healthy and a COVID-19 infected. The experimental result shows that the OGA-ELM achieves 100.00% accuracy with fast computation time. This demonstrates that OGA-ELM is an efficient method for COVID-19 detecting using chest X-ray images.

Albadr Musatafa Abbas Abbood, Tiun Sabrina, Ayob Masri, Al-Dhief Fahad Taha, Omar Khairuddin, Hamzah Faizal Amri

2020

General

The Hubbard-U correction and optical properties of d0 metal oxide photocatalysts.

In The Journal of chemical physics

We report a systematic investigation of individual and multisite Hubbard-U corrections for the electronic, structural, and optical properties of the metal titanate oxide d⁰ photocatalysts SrTiO₃ and rutile/anatase TiO₂. Accurate bandgaps for these materials can be reproduced with local density approximation and generalized gradient approximation exchange-correlation density functionals via a continuous series of empirically derived U^d and U^p combinations, which are relatively insensitive to the choice of functional. On the other hand, lattice parameters are much more sensitive to the choice of U^d and U^p, but in a systematic way that enables the U^d and U^p corrections to be used to qualitatively gauge the extent of self-interaction error in the electron density. Modest U^d corrections (e.g., 4 eV-5 eV) yield the most reliable dielectric response functions for SrTiO₃ and are comparable to the range of U^d values derived via linear response approaches. For r-TiO₂ and a-TiO₂, however, the U^d,p corrections that yield accurate bandgaps fail to accurately describe both the parallel and perpendicular components of the dielectric response function. Analysis of individual U^d and U^p corrections on the optical properties of SrTiO₃ suggests that the most consequential of the two individual corrections is U^d, as it predominately determines the accuracy of the dominant excitation from O-2p to the Ti-3d t_2g/e_g orbitals. U^p, on the other hand, can be used to shift the entire optical response uniformly to higher frequencies. These results will assist high-throughput and machine learning approaches to screening photoactive materials based on d⁰ photocatalysts.

Brown Joshua J, Page Alister J

2020-Dec-14

General

Study of the Novel AI-Powered Emotional Intelligence and Mindfulness App (Ajivar) for the College population during the COVID-19 Pandemic.

In JMIR formative research

BACKGROUND : Emotional Intelligence (EI) and mindfulness can impact the level of anxiety and depression that an individual experiences. These symptoms have been exacerbated in college students during the COVID-19 pandemic. AjivarTM is an application that utilizes artificial intelligence (AI) and machine learning (ML) to deliver personalized mindfulness and EI training.

OBJECTIVE : The main objective of this research study was to determine the effectiveness of delivering an EI curriculum and mindfulness techniques using an AI conversation platform, AjivarTM to improve symptoms of anxiety and depression during this pandemic.

METHODS : 95 subjects ages 18-29 years were recruited from a second semester freshmen of students. All participants completed the online TestWell inventory at the start and at the end of the 14 week semester. The comparison group (n=45) was given routine mental wellness instruction. The intervention group (n=50) were required to complete AjivarTM activities in addition to routine mental wellness instruction during the semester, which coincided with the onset of the COVID-19 pandemic. This group also completed assessments to evaluate for anxiety (Generalized Anxiety Disorder scale, GAD-7) and depression (Patient Health Questionnaire, PHQ-9).

RESULTS : Study participants were 19.81.9 years old, 28% males (27/95), and 60% Caucasian. No significant demographic differences existed between the comparison and intervention groups. Subjects in the intervention group interacted with AjivarTM for a mean of 14241168 minutes. There was a significant decrease in anxiety as measured by GAD-7 (11.471.85 at the start of the study compared to 6.271.44, P<0.01, at the end). There was a significant reduction in the symptoms of depression measured by PHQ-9 (10.692.04 vs. 6.692.41, P<0.01). Both the intervention and the comparison groups independently had significant improvements in pre-post TestWell inventory. The subgroups in the inventory for social awareness and spirituality showed significant improvement in the intervention group. In a group of participants (n=11) where GAD-7 was available during the onset of the COVID-19 pandemic, it showed an increase in anxiety (11.012.16 at the start to 13.031.34, P=0.23) in mid-March (onset of pandemic) to a significant decrease at the end of the study period (6.31.44, P<0.01).

CONCLUSIONS : It is possible to deliver EI and mindfulness training in a scalable way using the AjivarTM app during the COVID-19 pandemic resulting in improvements in anxiety, depression, and EI in the college population.

CLINICALTRIAL :

Sturgill Ronda, Martinasek Mary, Schmidt Trine, Goyal Raj

2020-Dec-08

Surgery

Machine Learning Based Analysis of Human Serum N-glycome Alterations to Follow up Lung Tumor Surgery.

In Cancers

The human serum N-glycome is a valuable source of biomarkers for malignant diseases, already utilized in multiple studies. In this paper, the N-glycosylation changes in human serum proteins were analyzed after surgical lung tumor resection. Seventeen lung cancer patients were involved in this study and the N-glycosylation pattern of their serum samples was analyzed before and after the surgery using capillary electrophoresis separation with laser-induced fluorescent detection. The relative peak areas of 21 N-glycans were evaluated from the acquired electropherograms using machine learning-based data analysis. Individual glycans as well as their subclasses were taken into account during the course of evaluation. For the data analysis, both discrete (e.g., smoker or not) and continuous (e.g., age of the patient) clinical parameters were compared against the alterations in these 21 N-linked carbohydrate structures. The classification tree analysis resulted in a panel of N-glycans, which could be used to follow up on the effects of lung tumor surgical resection.

Mészáros Brigitta, Járvás Gábor, Kun Renáta, Szabó Miklós, Csánky Eszter, Abonyi János, Guttman András

2020-Dec-09

N-glycans, capillary electrophoresis, lung cancer, machine learning, surgery

Cardiology

Are e-learning Webinars the future of medical education? An exploratory study of a disruptive innovation in the COVID-19 era.

In Cardiology in the young

OBJECTIVE : This study investigated the impact of the Webinar on deep human learning of CHD.

MATERIALS AND METHODS : This cross-sectional survey design study used an open and closed-ended questionnaire to assess the impact of the Webinar on deep learning of topical areas within the management of the post-operative tetralogy of Fallot patients. This was a quantitative research methodology using descriptive statistical analyses with a sequential explanatory design.

RESULTS : One thousand-three-hundred and seventy-four participants from 100 countries on 6 continents joined the Webinar, 557 (40%) of whom completed the questionnaire. Over 70% of participants reported that they "agreed" or "strongly agreed" that the Webinar format promoted deep learning for each of the topics compared to other standard learning methods (textbook and journal learning). Two-thirds expressed a preference for attending a Webinar rather than an international conference. Over 80% of participants highlighted significant barriers to attending conferences including cost (79%), distance to travel (49%), time commitment (51%), and family commitments (35%). Strengths of the Webinar included expertise, concise high-quality presentations often discussing contentious issues, and the platform quality. The main weakness was a limited time for questions. Just over 53% expressed a concern for the carbon footprint involved in attending conferences and preferred to attend a Webinar.

CONCLUSION : E-learning Webinars represent a disruptive innovation, which promotes deep learning, greater multidisciplinary participation, and greater attendee satisfaction with fewer barriers to participation. Although Webinars will never fully replace conferences, a hybrid approach may reduce the need for conferencing, reduce carbon footprint. and promote a "sustainable academia".

McMahon Colin J, Tretter Justin T, Faulkner Theresa, Krishna Kumar R, Redington Andrew N, Windram Jonathan D

2020-Dec-15

COVID-19, Conference, Webinar, e-learning, education, tetralogy of Fallot

General

Using machine learning tools to investigate factors associated with trends in 'no-shows' in outpatient appointments.

In Health & place

Missed appointments are estimated to cost the UK National Health Service (NHS) approximately £1 billion annually. Research that leads to a fuller understanding of the types of factors influencing spatial and temporal patterns of these so-called "Did-Not-Attends" (DNAs) is therefore timely. This research articulates the results of a study that uses machine learning approaches to investigate whether these factors are consistent across a range of medical specialities. A predictive model was used to determine the risk-increasing and risk-mitigating factors associated with missing appointments, which were then used to assign a risk score to patients on an appointment-by-appointment basis for each speciality. Results show that the best predictors of DNAs include the patient's age, appointment history, and the deprivation rank of their area of residence. Findings have been analysed at both a geographical and medical speciality level, and the factors associated with DNAs have been shown to differ in terms of both importance and association. This research has demonstrated how machine learning techniques have real value in informing future intervention policies related to DNAs that can help reduce the burden on the NHS and improve patient care and well-being.

Incze Eduard, Holborn Penny, Higgs Gary, Ware Andrew

2020-Dec-12

Compositional versus contextual, Machine learning, Medical specialities, Missed appointments (‘Did-not-attend’DNA), Outpatients

General

Support vector machine for EELS oxidation state determination.

In Ultramicroscopy

Electron Energy-Loss Spectroscopy (EELS) is a powerful and versatile spectroscopic technique used to study the composition and local optoelectronic properties of nanometric materials. Currently, this technique is generating large amounts of spectra per experiment, producing a huge quantity of data to analyse. Several strategies can be applied in order to classify these data to map physical properties at the nanoscale. In the present study, the Support Vector Machine (SVM) algorithm is applied to EELS, and its effectiveness identifying EEL spectra is assessed. Our results evidence the capacity of SVM to determine the oxidation state of iron and manganese in iron and manganese oxides, based on the ELNES of the white lines of the transition metal. The SVM algorithm is first trained with given datasets and then the resulting models are tested through noisy test data sets. We demonstrate that SVM exhibits a very good performance classifying these EEL spectra, despite the usual level of noise and instrumental energy shifts.

Del-Pozo-Bueno D, Peiró F, Estradé S

2020-Dec-07

Electron Energy-Loss Spectroscopy, Iron Oxides, Machine Learning, Manganese Oxides, Support Vector Machine, Transition Metals

General

Machine learning approach for predicting Fusarium culmorum and F. proliferatum growth and mycotoxin production in treatments with ethylene-vinyl alcohol copolymer films containing pure components of essential oils.

In International journal of food microbiology

Fusarium culmorum and F. proliferatum can grow and produce, respectively, zearalenone (ZEA) and fumonisins (FUM) in different points of the food chain. Application of antifungal chemicals to control these fungi and mycotoxins increases the risk of toxic residues in foods and feeds, and induces fungal resistances. In this study, a new and multidisciplinary approach based on the use of bioactive ethylene-vinyl alcohol copolymer (EVOH) films containing pure components of essential oils (EOCs) and machine learning (ML) methods is evaluated. Bioactive EVOH-EOC films were made incorporating cinnamaldehyde (CINHO), citral (CIT), isoeugenol (IEG) or linalool (LIN). Several ML methods (neural networks, random forests and extreme gradient boosted trees) and multiple linear regression (MLR) were applied and compared for modeling fungal growth and toxin production under different water activity (a_w) (0.96 and 0.99) and temperature (20 and 28 °C) regimes. The effective doses to reduce fungal growth rate (GR) by 50, 90 and 100% (ED₅₀, ED₉₀, and ED₁₀₀) of EOCs in EVOH films were in the ranges 200 to >3330, 450 to >3330, and 660 to >3330 μg/fungal culture (25 g of partly milled maize kernels in Petri dish), respectively, depending on the EOC, a_w and temperature. The type of EVOH-EOC film and EOC doses significantly affected GR in both species and ZEA and FUM production. Temperature also affected GR and a_w only affected GR and FUM production of F. proliferatum. EVOH-CIT was the most effective film against both species and ZEA and FUM production. Usually, when the EOC levels increased, GR and mycotoxin levels in the medium decreased although some treatments in combination with certain a_w and temperature values induced ZEA production. Random forest models predicted the GR of F. culmorum and F. proliferatum and ZEA and FUM production better than neural networks or extreme gradient boosted trees. The MLR mode provided the worst performance. This is the first approach on the ML potential in the study of the impact that bioactive EVOH films containing EOCs and environmental conditions have on F. culmorum and F. proliferatum growth and on ZEA and FUM production. The results suggest that these innovative packaging systems in combination with ML methods can be promising tools in the prediction and control of the risks associated with these toxigenic fungi and mycotoxins in food.

Tarazona Andrea, Mateo Eva M, Gómez José V, Gavara Rafael, Jiménez Misericordia, Mateo Fernando

2020-Dec-03

Bioactive EVOH-films, Fumonisins, Fusarium culmorum, Fusarium proliferatum, Machine learning methods, Zearalenone

Surgery

Weight-bearing CT in foot and ankle pathology.

In Orthopaedics & traumatology, surgery & research : OTSR

Cone-beam scanners (CBCT) enable CT to be performed under weight-bearing - notably for the foot and ankle. The technology is not new: it has been used since 1996 in dental surgery, where it has come to replace panoramic X-ray. What is new is placing the scanner on the ground, so as to have 3D weight-bearing images, initially of the foot and ankle, and later for the knee and pelvis. This saves time, radiation and money. It is now increasingly used, but is unfortunately limited by not having specific national health insurance cover in France, and by the psychological reticence that goes with any technological breakthrough. A review of the topic is indispensable, as it is essential to become properly acquainted with this technique. To this end, we shall be addressing 5 questions. What biases does conventional radiography incur? Projecting a volume onto a plane incurs deformation, precluding true measurement. Conventional CT is therefore often associated with an increased dose of radiation. What is the impact of CBCT on radiation dose, costs and the care pathway? The conical beam turns around the limb (under weight-bearing if so desired) in less than a minute, making the radiation dose no greater than in standard X-ray. What does the literature have to say about CBCT, and what are the indications? CBCT is indicated in all foot and ankle pathologies, and indications now extend to the upper limb and the knee, and will soon include the pelvis. How are angles measured on this 3D technique? The recently developed concept of 3D biometry uses dedicated software to identify anatomic landmarks and automatically segment the bones, thereby enabling every kind of measurement. What further developments are to be expected? CBCT may become indispensable to lower-limb surgical planning. Artificial Intelligence will reveal novel diagnostic, prognostic and therapeutic solutions. Level of evidence: V; expert opinion.

Lintz François, Beaudet Philippe, Richardi Gérard, Brilhault Jean

2020-Dec-12

3D Biometrics, Artificial Intelligence, CT, Cone-Beam, Deep Learning, Weight-Bearing

Public Health

An Aberration Detection-Based Approach for Sentinel Syndromic Surveillance of COVID-19 and Other Novel Influenza-Like Illnesses.

In Journal of biomedical informatics ; h5-index 55.0

Coronavirus Disease 2019 has emerged as a significant global concern, triggering harsh public health restrictions in a successful bid to curb its exponential growth. As discussion shifts towards relaxation of these restrictions, there is significant concern of second-wave resurgence. The key to managing these outbreaks is early detection and intervention, and yet there is a significant lag time associated with usage of laboratory confirmed cases for surveillance purposes. To address this, syndromic surveillance can be considered to provide a timelier alternative for first-line screening. Existing syndromic surveillance solutions are however typically focused around a known disease and have limited capability to distinguish between outbreaks of individual diseases sharing similar syndromes. This poses a challenge for surveillance of COVID-19 as its active periods tend to overlap temporally with other influenza-like illnesses. In this study we explore performing sentinel syndromic surveillance for COVID-19 and other influenza-like illnesses using a deep learning-based approach. Our methods are based on aberration detection utilizing autoencoders that leverages symptom prevalence distributions to distinguish outbreaks of two ongoing diseases that share similar syndromes, even if they occur concurrently. We first demonstrate that this approach works for detection of outbreaks of influenza, which has known temporal boundaries. We then demonstrate that the autoencoder can be trained to not alert on known and well-managed influenza-like illnesses such as the common cold and influenza. Finally, we applied our approach to 2019-2020 data in the context of a COVID-19 syndromic surveillance task to demonstrate how implementation of such a system could have provided early warning of an outbreak of a novel influenza-like illness that did not match the symptom prevalence profile of influenza and other known influenza-like illnesses.

Wen Andrew, Wang Liwei, He Huan, Liu Sijia, Fu Sunyang, Sohn Sunghwan, Kugel Jacob A, Kaggal Vinod C, Huang Ming, Wang Yanshan, Shen Feichen, Fan Jungwei, Liu Hongfang

2020-Dec-12

COVID-19, Deep Learning, Syndromic Surveillance

General

A Memory-Efficient Encoding Method for Processing Mixed-Type Data on Machine Learning.

In Entropy (Basel, Switzerland)

The most common machine-learning methods solve supervised and unsupervised problems based on datasets where the problem's features belong to a numerical space. However, many problems often include data where numerical and categorical data coexist, which represents a challenge to manage them. To transform categorical data into a numeric form, preprocessing tasks are compulsory. Methods such as one-hot and feature-hashing have been the most widely used encoding approaches at the expense of a significant increase in the dimensionality of the dataset. This effect introduces unexpected challenges to deal with the overabundance of variables and/or noisy data. In this regard, in this paper we propose a novel encoding approach that maps mixed-type data into an information space using Shannon's Theory to model the amount of information contained in the original data. We evaluated our proposal with ten mixed-type datasets from the UCI repository and two datasets representing real-world problems obtaining promising results. For demonstrating the performance of our proposal, this was applied for preparing these datasets for classification, regression, and clustering tasks. We demonstrate that our encoding proposal is remarkably superior to one-hot and feature-hashing encoding in terms of memory efficiency. Our proposal can preserve the information conveyed by the original data.

Lopez-Arevalo Ivan, Aldana-Bobadilla Edwin, Molina-Villegas Alejandro, Galeana-Zapién Hiram, Muñiz-Sanchez Victor, Gausin-Valle Saul

2020-Dec-09

categorical data, data preprocessing, machine learning

General

Machine Learning Models for covid-19 future forecasting.

In Materials today. Proceedings

Computational methods for machine learning (ML) have shown their meaning for the projection of potential results for informed decisions. Machine learning algorithms have been applied for a long time in many applications requiring the detection of adverse risk factors. This study shows the ability to predict the number of individuals who are affected by the COVID-19[1] as a potential threat to human beings by ML modelling. In this analysis, the risk factors of COVID-19 were exponential smoothing (ES). The Lower Absolute Reductor and Selection Operator, (LASSo), Vector Assistance (SVM), four normal potential forecasts, such as Linear Regression (LR)). [2] Each of these machine-learning models has three distinct kinds of predictions: the number of newly infected COVID 19 people, mortality rates and the recovered COVID-19 estimates in the next 10 days. These approaches are better used in the latest COVID-19 situation, as shown by the findings of the analysis. The LR, that is effective in predicting new cases of corona, death numbers and recovery.

Mojjada Ramesh Kumar, Yadav Arvind, Prabhu A V, Natarajan Yuvaraj

2020-Dec-09

COVID-19, R2 score adjusted, exponential process of smoothing, future forecasting, machine learning supervised

General

Classification of Parkinson's disease based on Multi-modal Features and Stacking Ensemble Learning.

In Journal of neuroscience methods

BACKGROUND : Early diagnosis of Parkinson's disease (PD) enables timely treatment of patients and helps control the course of the disease. An efficient and reliable approach is therefore needed to develop for improving the clinical ability to diagnose this disease.

NEW METHOD : We proposed a two-layer stacking ensemble learning framework with fusing multi-modal features in this study, for accurately identifying early PD with healthy control (HC). To begin with, we investigated relative importance of multi-modal neuroimaging (T1 weighted image (T1WI), diffusion tensor imaging (DTI)) and early clinical assessment to classify PD and HC. Next, a two-layer stacking ensemble framework was proposed: at the first layer, we evaluated advantages of these four base classifiers: support vector machine (SVM), random forests (RF), K-nearest neighbor (KNN) and artificial neural network (ANN); at the second layer, a logistic regression (LR) classifier was applied to classify PD. The performance of the proposed model was evaluated by comparing with traditional ensemble models.

RESULTS : The proposed method performed an accuracy of 96.88%, a precision of 100%, a recall of 95% and a F₁ score of 97.44% respectively for identifying PD and HC.

COMPARISON WITH EXISTING METHOD : The classification results showed that the proposed model achieved a superior performance in comparison with traditional ensemble models.

CONCLUSION : The stacking ensemble model with efficiently and effectively integrate multiple base classifiers performed higher accuracy than each single traditional model. The method developed in this study provided a novel strategy to enhance the accuracy of diagnosis and early detection of PD.

Yang Yifeng, Wei Long, Hu Ying, Wu Yan, Hu Liangyun, Nie Shengdong

2020-Dec-12

Computer-aided diagnosis, Ensemble Learning, Machine Learning (ML), Magnetic resonance imaging, Parkinson’s disease

General

Predicted Cellular Immunity Population Coverage Gaps for SARS-CoV-2 Subunit Vaccines and Their Augmentation by Compact Peptide Sets.

In Cell systems

Subunit vaccines induce immunity to a pathogen by presenting a component of the pathogen and thus inherently limit the representation of pathogen peptides for cellular immunity-based memory. We find that severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) subunit peptides may not be robustly displayed by the major histocompatibility complex (MHC) molecules in certain individuals. We introduce an augmentation strategy for subunit vaccines that adds a small number of SARS-CoV-2 peptides to a vaccine to improve the population coverage of pathogen peptide display. Our population coverage estimates integrate clinical data on peptide immunogenicity in convalescent COVID-19 patients and machine learning predictions. We evaluate the population coverage of 9 different subunits of SARS-CoV-2, including 5 functional domains and 4 full proteins, and augment each of them to fill a predicted coverage gap.

Liu Ge, Carter Brandon, Gifford David K

2020-Nov-27

SARS-CoV-2, combinatorial optimization, haplotype, machine learning, major histocompatibility complex, peptide vaccine, population coverage, subunit, vaccine augmentation, vaccine evaluation

General

Optimised genetic algorithm-extreme learning machine approach for automatic COVID-19 detection.

In PloS one ; h5-index 176.0

The coronavirus disease (COVID-19), is an ongoing global pandemic caused by severe acute respiratory syndrome. Chest Computed Tomography (CT) is an effective method for detecting lung illnesses, including COVID-19. However, the CT scan is expensive and time-consuming. Therefore, this work focus on detecting COVID-19 using chest X-ray images because it is widely available, faster, and cheaper than CT scan. Many machine learning approaches such as Deep Learning, Neural Network, and Support Vector Machine; have used X-ray for detecting the COVID-19. Although the performance of those approaches is acceptable in terms of accuracy, however, they require high computational time and more memory space. Therefore, this work employs an Optimised Genetic Algorithm-Extreme Learning Machine (OGA-ELM) with three selection criteria (i.e., random, K-tournament, and roulette wheel) to detect COVID-19 using X-ray images. The most crucial strength factors of the Extreme Learning Machine (ELM) are: (i) high capability of the ELM in avoiding overfitting; (ii) its usability on binary and multi-type classifiers; and (iii) ELM could work as a kernel-based support vector machine with a structure of a neural network. These advantages make the ELM efficient in achieving an excellent learning performance. ELMs have successfully been applied in many domains, including medical domains such as breast cancer detection, pathological brain detection, and ductal carcinoma in situ detection, but not yet tested on detecting COVID-19. Hence, this work aims to identify the effectiveness of employing OGA-ELM in detecting COVID-19 using chest X-ray images. In order to reduce the dimensionality of a histogram oriented gradient features, we use principal component analysis. The performance of OGA-ELM is evaluated on a benchmark dataset containing 188 chest X-ray images with two classes: a healthy and a COVID-19 infected. The experimental result shows that the OGA-ELM achieves 100.00% accuracy with fast computation time. This demonstrates that OGA-ELM is an efficient method for COVID-19 detecting using chest X-ray images.

Albadr Musatafa Abbas Abbood, Tiun Sabrina, Ayob Masri, Al-Dhief Fahad Taha, Omar Khairuddin, Hamzah Faizal Amri

2020

General

Fabrication and evaluation of complicated microstructures on cylindrical surface.

In PloS one ; h5-index 176.0

Various items of roll molds are popularly used to fabricate different kinds of optical films for optoelectronic information and other new and high-tech fields, while the fabrication and evaluation of optical microstructures on a cylindrical roller surface is more difficult than ecumenically manufactured products. In this study, the machinability of microstructures on the roll based on a fast tool servo (FTS) system is investigated. First, the flexible hinge holder for a FTS is designed and its structural parameters are optimized with finite-element analysis and fatigue reliability theory. The tool radius compensation algorithm for complicated microstructures is then deduced based on the surface fitting and bilinear interpolation algorithm of discrete data. Meanwhile, the evaluation index and method are proposed by the medium section method. Finally, a machining test of aspheric arrays on a cylindrical aluminum surface is carried out, and the high quality of the microstructure indicates that the proposed method is able to be used to fabricate optical microstructures.

Yong Jiahao, Liu Junfeng, Guan Chaoliang, Dai Yifan, Li Fei, Fan Zhanbin

2020

General

Study of the Novel AI-Powered Emotional Intelligence and Mindfulness App (Ajivar) for the College population during the COVID-19 Pandemic.

In JMIR formative research

BACKGROUND : Emotional Intelligence (EI) and mindfulness can impact the level of anxiety and depression that an individual experiences. These symptoms have been exacerbated in college students during the COVID-19 pandemic. AjivarTM is an application that utilizes artificial intelligence (AI) and machine learning (ML) to deliver personalized mindfulness and EI training.

OBJECTIVE : The main objective of this research study was to determine the effectiveness of delivering an EI curriculum and mindfulness techniques using an AI conversation platform, AjivarTM to improve symptoms of anxiety and depression during this pandemic.

METHODS : 95 subjects ages 18-29 years were recruited from a second semester freshmen of students. All participants completed the online TestWell inventory at the start and at the end of the 14 week semester. The comparison group (n=45) was given routine mental wellness instruction. The intervention group (n=50) were required to complete AjivarTM activities in addition to routine mental wellness instruction during the semester, which coincided with the onset of the COVID-19 pandemic. This group also completed assessments to evaluate for anxiety (Generalized Anxiety Disorder scale, GAD-7) and depression (Patient Health Questionnaire, PHQ-9).

RESULTS : Study participants were 19.81.9 years old, 28% males (27/95), and 60% Caucasian. No significant demographic differences existed between the comparison and intervention groups. Subjects in the intervention group interacted with AjivarTM for a mean of 14241168 minutes. There was a significant decrease in anxiety as measured by GAD-7 (11.471.85 at the start of the study compared to 6.271.44, P<0.01, at the end). There was a significant reduction in the symptoms of depression measured by PHQ-9 (10.692.04 vs. 6.692.41, P<0.01). Both the intervention and the comparison groups independently had significant improvements in pre-post TestWell inventory. The subgroups in the inventory for social awareness and spirituality showed significant improvement in the intervention group. In a group of participants (n=11) where GAD-7 was available during the onset of the COVID-19 pandemic, it showed an increase in anxiety (11.012.16 at the start to 13.031.34, P=0.23) in mid-March (onset of pandemic) to a significant decrease at the end of the study period (6.31.44, P<0.01).

CONCLUSIONS : It is possible to deliver EI and mindfulness training in a scalable way using the AjivarTM app during the COVID-19 pandemic resulting in improvements in anxiety, depression, and EI in the college population.

CLINICALTRIAL :

Sturgill Ronda, Martinasek Mary, Schmidt Trine, Goyal Raj

2020-Dec-08

General

A Review on Deep Learning Techniques for Video Prediction.

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

The ability to predict, anticipate and reason about future outcomes is a key component of intelligent decision-making systems. In light of the success of deep learning in computer vision, deep-learning-based video prediction emerged as a promising research direction. Defined as a self-supervised learning task, video prediction represents a suitable framework for representation learning, as it demonstrated potential capabilities for extracting meaningful representations of the underlying patterns in natural videos. Motivated by the increasing interest in this task, we provide a review on the deep learning methods for prediction in video sequences. We firstly define the video prediction fundamentals, as well as mandatory background concepts and the most used datasets. Next, we carefully analyze existing video prediction models organized according to a proposed taxonomy, highlighting their contributions and their significance in the field. The summary of the datasets and methods is accompanied with experimental results that facilitate the assessment of the state of the art on a quantitative basis. The paper is summarized by drawing some general conclusions, identifying open research challenges and by pointing out future research directions.

Oprea Sergiu, Martinez-Gonzalez Pablo, Garcia-Garcia Alberto, Castro-Vargas John Alejandro, Orts-Escolano Sergio, Garcia-Rodriguez Jose, Argyros Antonis

2020-Dec-15

General

Geometry-Aware Generation of Adversarial Point Clouds.

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

Machine learning models are vulnerable to adversarial examples. While most of the existing adversarial methods are on 2D image, a few recent ones extend the studies to 3D point clouds data. These methods generate point outliers, which are noticeable and easy to defend against using the simple technique of outlier removal. Motivated by the different mechanisms humans perceive by 2D images and 3D shapes, we propose the new design of geometry-aware objectives, whose solutions favor the desired surface properties of smoothness and fairness. To generate adversarial point clouds, we use a misclassification loss that supports continuous pursuit of malicious signals. Regularizing the attack loss with our proposed geometry-aware objectives results in our proposed method, Geometry-Aware Adversarial Attack (GeoA³). The results of GeoA³ tend to be more harmful, harder to defend against, and of the key adversarial characterization of being imperceptible. We also present a simple but effective algorithm termed GeoA₊³-IterNormPro towards surface-level adversarial attacks via generation of adversarial point clouds. We evaluate our methods on both synthetic and physical objects. For a qualitative evaluation, we conduct subjective studies by collecting human preferences from Amazon Mechanical Turk. Comparative results in comprehensive experiments confirm the advantages of our proposed methods. Our source codes are publicly available at https://github.com/Yuxin-Wen/GeoA3.

Wen Yuxin, Lin Jiehong, Chen Ke, Chen C L Philip, Jia Kui

2020-Dec-15

Surgery

AI technology for remote clinical assessment and monitoring.

In Journal of wound care

OBJECTIVE : To report the clinical validation of an innovative, artificial intelligence (AI)-powered, portable and non-invasive medical device called Wound Viewer. The AI medical device uses dedicated sensors and AI algorithms to remotely collect objective and precise clinical data, including three-dimensional (3D) wound measurements, tissue composition and wound classification through the internationally recognised Wound Bed Preparation (WBP) protocol; this data can then be shared through a secure General Data Protection Regulation (GDPR)- and Health Insurance Portability and Accountability Act (HIPAA)-compliant data transfer system. This trial aims to test the reliability and precision of the AI medical device and its ability to aid health professionals in clinically evaluating wounds as efficiently remotely as at the bedside.

METHOD : This non-randomised comparative clinical trial was conducted in the Clinica San Luca (Turin, Italy). Patients were divided into three groups: (i) patients with venous and arterial ulcers in the lower limbs; (ii) patients with diabetes and presenting with diabetic foot syndrome; and (iii) patients with pressure ulcers. Each wound was evaluated for area, depth, volume and WBP wound classification. Each patient was examined once and the results, analysed by the AI medical device, were compared against data obtained following visual evaluation by the physician and research team. The area and depth were compared with a Kruskal-Wallis one-way analysis of variations in the obtained distribution (expected p-value>0.1 for both tests). The WBP classification and tissue segmentation were analysed by directly comparing the classification obtained by the AI medical device against that of the testing physician.

RESULTS : A total of 150 patients took part in the trial. The results demonstrated that the AI medical device's AI algorithm could acquire objective clinical parameters in a completely automated manner. The AI medical device reached 97% accuracy against the WBP classification and tissue segmentation analysis compared with that performed in person by the physician. Moreover, data regarding the measurements of the wounds, as analysed through the Kruskal-Wallis technique, showed that the data distribution proved comparable with the other methods of measurement previously clinically validated in the literature (p=0.9).

CONCLUSION : These findings indicate that remote wound assessment undertaken by physicians is as effective through the AI medical device as bedside examination, and that the device was able to assess wounds and provide a precise WBP wound classification. Furthermore, there was no need for manual data entry, thereby reducing the risk of human error while preserving high-quality clinical diagnostic data.

Zoppo Gianluca, Marrone Francesco, Pittarello Monica, Farina Marco, Uberti Alberto, Demarchi Danilo, Secco Jacopo, Corinto Fernando, Ricci Elia

2020-Dec-02

automatic wound classification, medical device, telemedicine, three-dimensional wound measurement, wound bed preparation score, wound care, wound healing, wound measurement techniques, wounds

Public Health

Incidence of gastrointestinal stromal tumor in Chinese urban population: A national population-based study.

In Cancer medicine

BACKGROUND : Information on incidence of gastrointestinal stromal tumor (GIST), the most common type of mesenchymal tumor in gastrointestinal tract, was limited in China. This study aimed to estimate the incidence of GIST in urban population from mainland China in 2016.

METHODS : Urban Employee Basic Medical Insurance (UEBMI) and Urban Residence Basic Medical Insurance (URBMI) in China were used. The denominator of incidence was the total person-years of insured individuals in 2016 in the database, covering approximately 0.43 billion individuals. The numerator was the number of incident GIST cases in 2016.

RESULTS : The crude incidence in 2016 was 0.40 per 100,000 person-years (95% CI, 0.06-1.03). Male incidence was higher than female incidence (0.44 vs. 0.36, rate ratio: 1.22, p < 0.001). The mean age at diagnosis was 55.20 years (SD = 14.26) and the incidence among those aged 50 years or older was 2.63 times (0.84 vs. 0.32, p < 0.001) higher than those aged under 50. The highest incidence was observed in East China (2.29, 95% CI: 0.46-5.54).

CONCLUSIONS : The incidence of GIST in mainland China was lower than Europe, North America and Korea. The mean age at diagnosis of GIST in China was younger than that of Europe and Canada. This study provides useful information to further research, policy formulating and management of GIST.

Xu Lu, Ma Yanpeng, Wang Shengfeng, Feng Jingnan, Liu Lili, Wang Jinxi, Liu Guozhen, Xiu Dianrong, Fu Wei, Zhan Siyan, Sun Tao, Gao Pei

2020-Dec-15

China, epidemiology, gastrointestinal stromal tumor, incidence

General

Extracting Family History of Patients From Clinical Narratives: Exploring an End-to-End Solution With Deep Learning Models.

In JMIR medical informatics ; h5-index 23.0

BACKGROUND : Patients' family history (FH) is a critical risk factor associated with numerous diseases. However, FH information is not well captured in the structured database but often documented in clinical narratives. Natural language processing (NLP) is the key technology to extract patients' FH from clinical narratives. In 2019, the National NLP Clinical Challenge (n2c2) organized shared tasks to solicit NLP methods for FH information extraction.

OBJECTIVE : This study presents our end-to-end FH extraction system developed during the 2019 n2c2 open shared task as well as the new transformer-based models that we developed after the challenge. We seek to develop a machine learning-based solution for FH information extraction without task-specific rules created by hand.

METHODS : We developed deep learning-based systems for FH concept extraction and relation identification. We explored deep learning models including long short-term memory-conditional random fields and bidirectional encoder representations from transformers (BERT) as well as developed ensemble models using a majority voting strategy. To further optimize performance, we systematically compared 3 different strategies to use BERT output representations for relation identification.

RESULTS : Our system was among the top-ranked systems (3 out of 21) in the challenge. Our best system achieved micro-averaged F1 scores of 0.7944 and 0.6544 for concept extraction and relation identification, respectively. After challenge, we further explored new transformer-based models and improved the performances of both subtasks to 0.8249 and 0.6775, respectively. For relation identification, our system achieved a performance comparable to the best system (0.6810) reported in the challenge.

CONCLUSIONS : This study demonstrated the feasibility of utilizing deep learning methods to extract FH information from clinical narratives.

Yang Xi, Zhang Hansi, He Xing, Bian Jiang, Wu Yonghui

2020-Dec-15

deep learning, family history, information extraction, natural language processing

General

MirLocPredictor: A ConvNet-Based Multi-Label MicroRNA Subcellular Localization Predictor by Incorporating k-Mer Positional Information.

In Genes

MicroRNAs (miRNA) are small noncoding RNA sequences consisting of about 22 nucleotides that are involved in the regulation of almost 60% of mammalian genes. Presently, there are very limited approaches for the visualization of miRNA locations present inside cells to support the elucidation of pathways and mechanisms behind miRNA function, transport, and biogenesis. MIRLocator, a state-of-the-art tool for the prediction of subcellular localization of miRNAs makes use of a sequence-to-sequence model along with pretrained k-mer embeddings. Existing pretrained k-mer embedding generation methodologies focus on the extraction of semantics of k-mers. However, in RNA sequences, positional information of nucleotides is more important because distinct positions of the four nucleotides define the function of an RNA molecule. Considering the importance of the nucleotide position, we propose a novel approach (kmerPR2vec) which is a fusion of positional information of k-mers with randomly initialized neural k-mer embeddings. In contrast to existing k-mer-based representation, the proposed kmerPR2vec representation is much more rich in terms of semantic information and has more discriminative power. Using novel kmerPR2vec representation, we further present an end-to-end system (MirLocPredictor) which couples the discriminative power of kmerPR2vec with Convolutional Neural Networks (CNNs) for miRNA subcellular location prediction. The effectiveness of the proposed kmerPR2vec approach is evaluated with deep learning-based topologies (i.e., Convolutional Neural Networks (CNN) and Recurrent Neural Network (RNN)) and by using 9 different evaluation measures. Analysis of the results reveals that MirLocPredictor outperform state-of-the-art methods with a significant margin of 18% and 19% in terms of precision and recall.

Asim Muhammad Nabeel, Malik Muhammad Imran, Zehe Christoph, Trygg Johan, Dengel Andreas, Ahmed Sheraz

2020-Dec-09

convolutional neural network, k-mer positional encoding, microRNA location predictor, microRNA multi-label classification, microRNA subcellular localization

General

Modeling and analysis of different scenarios for the spread of COVID-19 by using the modified multi-agent systems - Evidence from the selected countries.

In Results in physics

Currently, there is a global pandemic of COVID-19. To assess its prevalence, it is necessary to have adequate models that allow real-time modeling of the impact of various quarantine measures by the state. The SIR model, which is implemented using a multi-agent system based on mobile cellular automata, was improved. The paper suggests ways to improve the rules of the interaction and behavior of agents. Methods of comparing the parameters of the SIR model with real geographical, social and medical indicators have been developed. That allows the modeling of the spatial distribution of COVID-19 as a single location and as the whole country consisting of individual regions that interact with each other by transport, taking into account factors such as public transport, supermarkets, schools, universities, gyms, churches, parks. The developed model also allows us to assess the impact of quarantine, restrictions on transport connections between regions, to take into account such factors as the incubation period, the mask regime, maintaining a safe distance between people, and so on. A number of experiments were conducted in the work, which made it possible to assess both the impact of individual measures to stop the pandemic and their comprehensive application. A method of comparing computer-time and dynamic parameters of the model with real data is proposed, which allowed assessing the effectiveness of the government in stopping the pandemic in the Chernivtsi region, Ukraine. A simulation of the pandemic spread in countries such as Slovakia, Turkey and Serbia was also conducted. The calculations showed the high-accuracy matching of the forecast model with real data.

Vyklyuk Yaroslav, Manylich Mykhailo, Škoda Miroslav, Radovanović Milan M, Petrović Marko D

2020-Dec-09

COVID-19, forecasting, modified multi-agent systems, public activities, simulations

Cardiology

Are e-learning Webinars the future of medical education? An exploratory study of a disruptive innovation in the COVID-19 era.

In Cardiology in the young

OBJECTIVE : This study investigated the impact of the Webinar on deep human learning of CHD.

MATERIALS AND METHODS : This cross-sectional survey design study used an open and closed-ended questionnaire to assess the impact of the Webinar on deep learning of topical areas within the management of the post-operative tetralogy of Fallot patients. This was a quantitative research methodology using descriptive statistical analyses with a sequential explanatory design.

RESULTS : One thousand-three-hundred and seventy-four participants from 100 countries on 6 continents joined the Webinar, 557 (40%) of whom completed the questionnaire. Over 70% of participants reported that they "agreed" or "strongly agreed" that the Webinar format promoted deep learning for each of the topics compared to other standard learning methods (textbook and journal learning). Two-thirds expressed a preference for attending a Webinar rather than an international conference. Over 80% of participants highlighted significant barriers to attending conferences including cost (79%), distance to travel (49%), time commitment (51%), and family commitments (35%). Strengths of the Webinar included expertise, concise high-quality presentations often discussing contentious issues, and the platform quality. The main weakness was a limited time for questions. Just over 53% expressed a concern for the carbon footprint involved in attending conferences and preferred to attend a Webinar.

CONCLUSION : E-learning Webinars represent a disruptive innovation, which promotes deep learning, greater multidisciplinary participation, and greater attendee satisfaction with fewer barriers to participation. Although Webinars will never fully replace conferences, a hybrid approach may reduce the need for conferencing, reduce carbon footprint. and promote a "sustainable academia".

McMahon Colin J, Tretter Justin T, Faulkner Theresa, Krishna Kumar R, Redington Andrew N, Windram Jonathan D

2020-Dec-15

COVID-19, Conference, Webinar, e-learning, education, tetralogy of Fallot

General

Non-linear association of efficiency of practice of adult elite athletes with their youth multi-sport practice.

In Journal of sports sciences ; h5-index 52.0

We explored associations of elite athletes' multi-year efficiency of practice and improvement of performance with their current and earlier participation patterns. Participants were 80 adult German track-and-field national-squad athletes. Performance improvement was measured as development of athletes' highest track-and-field championship level and placing from 19 to 25 years (t₁-t₂). Practice efficiency was defined as performance improvement per amount of coach-led athletics practice from t₁ to t₂. Participation variables included amounts of coach-led practice and peer-led play in athletics and other sports through t₁ and t₁-t₂. Analyses involved an advanced machine learning procedure, XGBoost, allowing non-linear, multivariate exploration. We computed two models, one for performance improvement ("good" discriminative performance, AUC = 0.82) and one for practice efficiency ("fair", AUC = 0.73). Four central findings emerged: 1. Childhood/adolescent coach-led multi-sport practice was a critical discriminator of adult practice efficiency and performance improvement. 2. Associations were non-linear, displaying a saturation pattern. 3. The likelihood of achieving high adult practice efficiency was greatest when combining ~1,000-2,500 track-and-field practice hours until t₁ with ~1,250 other-sports practice hours until t₁. 4. Peer-led engagement in any sport had negligible effects. Childhood/adolescent multi-sport coach-led practice apparently facilitated long-term sustainability of athletes' development of adult practice efficiency and performance improvement in athletics.

Barth Michael, Güllich Arne

2020-Dec-15

Elite sport, early specialization, efficiency of practice, machine learning, sustainability, talent development

General

Ambiguity in medical concept normalization: An analysis of types and coverage in electronic health record datasets.

In Journal of the American Medical Informatics Association : JAMIA

OBJECTIVES : Normalizing mentions of medical concepts to standardized vocabularies is a fundamental component of clinical text analysis. Ambiguity-words or phrases that may refer to different concepts-has been extensively researched as part of information extraction from biomedical literature, but less is known about the types and frequency of ambiguity in clinical text. This study characterizes the distribution and distinct types of ambiguity exhibited by benchmark clinical concept normalization datasets, in order to identify directions for advancing medical concept normalization research.

MATERIALS AND METHODS : We identified ambiguous strings in datasets derived from the 2 available clinical corpora for concept normalization and categorized the distinct types of ambiguity they exhibited. We then compared observed string ambiguity in the datasets with potential ambiguity in the Unified Medical Language System (UMLS) to assess how representative available datasets are of ambiguity in clinical language.

RESULTS : We found that <15% of strings were ambiguous within the datasets, while over 50% were ambiguous in the UMLS, indicating only partial coverage of clinical ambiguity. The percentage of strings in common between any pair of datasets ranged from 2% to only 36%; of these, 40% were annotated with different sets of concepts, severely limiting generalization. Finally, we observed 12 distinct types of ambiguity, distributed unequally across the available datasets, reflecting diverse linguistic and medical phenomena.

DISCUSSION : Existing datasets are not sufficient to cover the diversity of clinical concept ambiguity, limiting both training and evaluation of normalization methods for clinical text. Additionally, the UMLS offers important semantic information for building and evaluating normalization methods.

CONCLUSIONS : Our findings identify 3 opportunities for concept normalization research, including a need for ambiguity-specific clinical datasets and leveraging the rich semantics of the UMLS in new methods and evaluation measures for normalization.

Newman-Griffis Denis, Divita Guy, Desmet Bart, Zirikly Ayah, Rosé Carolyn P, Fosler-Lussier Eric

2020-Dec-15

Unified Medical Language System, controlled, machine learning, natural language processing, semantics, vocabulary

General

Learning fine-grained estimation of physiological states from coarse-grained labels by distribution restoration.

In Scientific reports ; h5-index 158.0

Due to its importance in clinical science, the estimation of physiological states (e.g., the severity of pathological tremor) has aroused growing interest in machine learning community. While the physiological state is a continuous variable, its continuity is lost when the physiological state is quantized into a few discrete classes during recording and labeling. The discreteness introduces misalignment between the true value and its label, meaning that these labels are unfortunately imprecise and coarse-grained. Most previous work did not consider the inaccuracy and directly utilized the coarse labels to train the machine learning algorithms, whose predictions are also coarse-grained. In this work, we propose to learn a precise, fine-grained estimation of physiological states using these coarse-grained ground truths. Established on mathematical rigorous proof, we utilize imprecise labels to restore the probabilistic distribution of precise labels in an approximate order-preserving fashion, then the deep neural network learns from this distribution and offers fine-grained estimation. We demonstrate the effectiveness of our approach in assessing the pathological tremor in Parkinson's Disease and estimating the systolic blood pressure from bioelectrical signals.

Qin Zengyi, Chen Jiansheng, Jiang Zhenyu, Yu Xumin, Hu Chunhua, Ma Yu, Miao Suhua, Zhou Rongsong

2020-Dec-15

General

Biodegradable carboxymethyl cellulose based material for sustainable packaging application.

In Scientific reports ; h5-index 158.0

The main goal of the present work was to develop a value-added product of biodegradable material for sustainable packaging. The use of agriculture waste-derived carboxymethyl cellulose (CMC) mainly is to reduce the cost involved in the development of the film, at present commercially available CMS is costly. The main focus of the research is to translate the agricultural waste-derived CMC to useful biodegradable polymer suitable for packaging material. During this process CMC was extracted from the agricultural waste mainly sugar cane bagasse and the blends were prepared using CMC (waste derived), gelatin, agar and varied concentrations of glycerol; 1.5% (sample A), 2% (sample B), and 2.5% (sample C) was added. Thus, the film derived from the sample C (gelatin + CMC + agar) with 2.0% glycerol as a plasticizer exhibited excellent properties than other samples A and B. The physiochemical properties of each developed biodegradable plastics (sample A, B, C) were characterized using Fourier Transform Infra-Red (FTIR) spectroscopy and Differential Scanning Calorimetry (DSC), Thermogravimetric analysis (TGA). The swelling test, solubility in different solvents, oil permeability coefficient, water permeability (WP), mechanical strength of the produced material was claimed to be a good material for packaging and meanwhile its biodegradability (soil burial method) indicated their environmental compatibility nature and commercial properties. The reflected work is a novel approach, and which is vital in the conversion of organic waste to value-added product development. There is also another way to utilize commercial CMC in preparation of polymeric blends for the packaging material, which can save considerable time involved in the recovery of CMC from sugarcane bagasse.

Yaradoddi Jayachandra S, Banapurmath Nagaraj R, Ganachari Sharanabasava V, Soudagar Manzoore Elahi M, Mubarak N M, Hallad Shankar, Hugar Shoba, Fayaz H

2020-Dec-15

General

The MoCA dataset, kinematic and multi-view visual streams of fine-grained cooking actions.

In Scientific data

MoCA is a bi-modal dataset in which we collect Motion Capture data and video sequences acquired from multiple views, including an ego-like viewpoint, of upper body actions in a cooking scenario. It has been collected with the specific purpose of investigating view-invariant action properties in both biological and artificial systems. Besides that, it represents an ideal test bed for research in a number of fields - including cognitive science and artificial vision - and application domains - as motor control and robotics. Compared to other benchmarks available, MoCA provides a unique compromise for research communities leveraging very different approaches to data gathering: from one extreme of action recognition in the wild - the standard practice nowadays in the fields of Computer Vision and Machine Learning - to motion analysis in very controlled scenarios - as for motor control in biomedical applications. In this work we introduce the dataset and its peculiarities, and discuss a baseline analysis as well as examples of applications for which the dataset is well suited.

Nicora Elena, Goyal Gaurvi, Noceti Nicoletta, Vignolo Alessia, Sciutti Alessandra, Odone Francesca

2020-Dec-15

General

The default network of the human brain is associated with perceived social isolation.

In Nature communications ; h5-index 260.0

Humans survive and thrive through social exchange. Yet, social dependency also comes at a cost. Perceived social isolation, or loneliness, affects physical and mental health, cognitive performance, overall life expectancy, and increases vulnerability to Alzheimer's disease-related dementias. Despite severe consequences on behavior and health, the neural basis of loneliness remains elusive. Using the UK Biobank population imaging-genetics cohort (n = ~40,000, aged 40-69 years when recruited, mean age = 54.9), we test for signatures of loneliness in grey matter morphology, intrinsic functional coupling, and fiber tract microstructure. The loneliness-linked neurobiological profiles converge on a collection of brain regions known as the 'default network'. This higher associative network shows more consistent loneliness associations in grey matter volume than other cortical brain networks. Lonely individuals display stronger functional communication in the default network, and greater microstructural integrity of its fornix pathway. The findings fit with the possibility that the up-regulation of these neural circuits supports mentalizing, reminiscence and imagination to fill the social void.

Spreng R Nathan, Dimas Emile, Mwilambwe-Tshilobo Laetitia, Dagher Alain, Koellinger Philipp, Nave Gideon, Ong Anthony, Kernbach Julius M, Wiecki Thomas V, Ge Tian, Li Yue, Holmes Avram J, Yeo B T Thomas, Turner Gary R, Dunbar Robin I M, Bzdok Danilo

2020-12-15

General

Comprehension of computer code relies primarily on domain-general executive brain regions.

In eLife

Computer programming is a novel cognitive tool that has transformed modern society. What cognitive and neural mechanisms support this skill? Here, we used functional magnetic resonance imaging to investigate two candidate brain systems: the multiple demand (MD) system, typically recruited during math, logic, problem solving, and executive tasks, and the language system, typically recruited during linguistic processing. We examined MD and language system responses to code written in Python, a text-based programming language (Experiment 1) and in ScratchJr, a graphical programming language (Experiment 2); for both, we contrasted responses to code problems with responses to content-matched sentence problems. We found that the MD system exhibited strong bilateral responses to code in both experiments, whereas the language system responded strongly to sentence problems, but weakly or not at all to code problems. Thus, the MD system supports the use of novel cognitive tools even when the input is structurally similar to natural language.

Ivanova Anna A, Srikant Shashank, Sueoka Yotaro, Kean Hope H, Dhamala Riva, O’Reilly Una-May, Bers Marina U, Fedorenko Evelina

2020-Dec-15

computer code, fMRI, human, language, multiple demand, neuroscience, programming

General

Emerging Materials for Neuromorphic Devices and Systems.

In iScience

Neuromorphic devices and systems have attracted attention as next-generation computing due to their high efficiency in processing complex data. So far, they have been demonstrated using both machine-learning software and complementary metal-oxide-semiconductor-based hardware. However, these approaches have drawbacks in power consumption and learning speed. An energy-efficient neuromorphic computing system requires hardware that can mimic the functions of a brain. Therefore, various materials have been introduced for the development of neuromorphic devices. Here, recent advances in neuromorphic devices are reviewed. First, the functions of biological synapses and neurons are discussed. Also, deep neural networks and spiking neural networks are described. Then, the operation mechanism and the neuromorphic functions of emerging devices are reviewed. Finally, the challenges and prospects for developing neuromorphic devices that use emerging materials are discussed.

Kim Min-Kyu, Park Youngjun, Kim Ik-Jyae, Lee Jang-Sik

2020-Dec-18

Devices, Electronic Materials, Materials Design, Memory Structure

General

Revealing Epigenetic Factors of circRNA Expression by Machine Learning in Various Cellular Contexts.

In iScience

Circular RNAs (circRNAs) have been identified as naturally occurring RNAs that are highly represented in the eukaryotic transcriptome. Although a large number of circRNAs have been reported, the underlying regulatory mechanism of circRNAs biogenesis remains largely unknown. Here, we integrated in-depth multi-omics data including epigenome, transcriptome, and non-coding RNA and identified candidate circRNAs in six cellular contexts. Next, circRNAs were divided into two classes (high versus low) with different expression levels. Machine learning models were constructed that predicted circRNA expression levels based on 11 different histone modifications and host gene expression. We found that the models achieve great accuracy in predicting high versus low expressed circRNAs. Furthermore, the expression levels of host genes of circRNAs, H3k36me3, H3k79me2, and H4k20me1 contributed greatly to the classification models in six cellular contexts. In summary, all these results suggest that epigenetic modifications, particularly histone modifications, can effectively predict expression levels of circRNAs.

Zhang Mengying, Xu Kang, Fu Limei, Wang Qi, Chang Zhenghong, Zou Haozhe, Zhang Yan, Li Yongsheng

2020-Dec-18

Bioinformatics, Omics, Transcriptomics

General

Machine Learning-Assisted High-Throughput Molecular Dynamics Simulation of High-Mechanical Performance Carbon Nanotube Structure.

In Nanomaterials (Basel, Switzerland)

Carbon nanotubes (CNTs) are novel materials with extraordinary mechanical properties. To gain insight on the design of high-mechanical-performance CNT-reinforced composites, the optimal structure of CNTs with high nominal tensile strength was determined in this study, where the nominal values correspond to the cross-sectional area of the entire specimen, including the hollow core. By using machine learning-assisted high-throughput molecular dynamics (HTMD) simulation, the relationship among the following structural parameters/properties was investigated: diameter, number of walls, chirality, and crosslink density. A database, comprising the various tensile test simulation results, was analyzed using a self-organizing map (SOM). It was observed that the influence of crosslink density on the nominal tensile strength tends to gradually decrease from the outside to the inside; generally, the crosslink density between the outermost wall and its adjacent wall is highly significant. In particular, based on our calculation conditions, five-walled, armchair-type CNTs with an outer diameter of 43.39 Å and crosslink densities (between the inner wall and outer wall) of 1.38 ± 1.16%, 1.13 ± 0.69%, 1.54 ± 0.57%, and 1.36 ± 0.35% were believed to be the optimal structure, with the nominal tensile strength and nominal Young's modulus reaching approximately 58-64 GPa and 677-698 GPa.

Xiang Yi, Shimoyama Koji, Shirasu Keiichi, Yamamoto Go

2020-Dec-09

Frenkel-pair crosslink, carbon nanotube, machine learning, mechanical properties, molecular dynamics simulations

General

How artificial intelligence and machine learning can help healthcare systems respond to COVID-19.

In Machine learning

The COVID-19 global pandemic is a threat not only to the health of millions of individuals, but also to the stability of infrastructure and economies around the world. The disease will inevitably place an overwhelming burden on healthcare systems that cannot be effectively dealt with by existing facilities or responses based on conventional approaches. We believe that a rigorous clinical and societal response can only be mounted by using intelligence derived from a variety of data sources to better utilize scarce healthcare resources, provide personalized patient management plans, inform policy, and expedite clinical trials. In this paper, we introduce five of the most important challenges in responding to COVID-19 and show how each of them can be addressed by recent developments in machine learning (ML) and artificial intelligence (AI). We argue that the integration of these techniques into local, national, and international healthcare systems will save lives, and propose specific methods by which implementation can happen swiftly and efficiently. We offer to extend these resources and knowledge to assist policymakers seeking to implement these techniques.

van der Schaar Mihaela, Alaa Ahmed M, Floto Andres, Gimson Alexander, Scholtes Stefan, Wood Angela, McKinney Eoin, Jarrett Daniel, Lio Pietro, Ercole Ari

2020-Dec-09

COVID-19, Clinical decision support, Healthcare

Internal Medicine

Machine Learning Analysis of the Bleomycin Mouse Model Reveals the Compartmental and Temporal Inflammatory Pulmonary Fingerprint.

In iScience

The bleomycin mouse model is the extensively used model to study pulmonary fibrosis; however, the inflammatory cell kinetics and their compartmentalization is still incompletely understood. Here we assembled historical flow cytometry data, totaling 303 samples and 16 inflammatory-cell populations, and applied advanced data modeling and machine learning methods to conclusively detail these kinetics. Three days post-bleomycin, the inflammatory profile was typified by acute innate inflammation, pronounced neutrophilia, especially of SiglecF⁺ neutrophils, and alveolar macrophage loss. Between 14 and 21 days, rapid responders were increasingly replaced by T and B cells and monocyte-derived alveolar macrophages. Multicolour imaging revealed the spatial-temporal cell distribution and the close association of T cells with deposited collagen. Unbiased immunophenotyping and data modeling exposed the dynamic shifts in immune-cell composition over the course of bleomycin-triggered lung injury. These results and workflow provide a reference point for future investigations and can easily be applied in the analysis of other datasets.

Bordag Natalie, Biasin Valentina, Schnoegl Diana, Valzano Francesco, Jandl Katharina, Nagy Bence M, Sharma Neha, Wygrecka Malgorzata, Kwapiszewska Grazyna, Marsh Leigh M

2020-Dec-18

Artificial Intelligence, Immune Response, Immunology

Radiology

Artificial intelligence in gastrointestinal endoscopy.

In VideoGIE : an official video journal of the American Society for Gastrointestinal Endoscopy

Background and Aims : Artificial intelligence (AI)-based applications have transformed several industries and are widely used in various consumer products and services. In medicine, AI is primarily being used for image classification and natural language processing and has great potential to affect image-based specialties such as radiology, pathology, and gastroenterology (GE). This document reviews the reported applications of AI in GE, focusing on endoscopic image analysis.

Methods : The MEDLINE database was searched through May 2020 for relevant articles by using key words such as machine learning, deep learning, artificial intelligence, computer-aided diagnosis, convolutional neural networks, GI endoscopy, and endoscopic image analysis. References and citations of the retrieved articles were also evaluated to identify pertinent studies. The manuscript was drafted by 2 authors and reviewed in person by members of the American Society for Gastrointestinal Endoscopy Technology Committee and subsequently by the American Society for Gastrointestinal Endoscopy Governing Board.

Results : Deep learning techniques such as convolutional neural networks have been used in several areas of GI endoscopy, including colorectal polyp detection and classification, analysis of endoscopic images for diagnosis of Helicobacter pylori infection, detection and depth assessment of early gastric cancer, dysplasia in Barrett's esophagus, and detection of various abnormalities in wireless capsule endoscopy images.

Conclusions : The implementation of AI technologies across multiple GI endoscopic applications has the potential to transform clinical practice favorably and improve the efficiency and accuracy of current diagnostic methods.

Pannala Rahul, Krishnan Kumar, Melson Joshua, Parsi Mansour A, Schulman Allison R, Sullivan Shelby, Trikudanathan Guru, Trindade Arvind J, Watson Rabindra R, Maple John T, Lichtenstein David R

2020-Dec

ADR, adenoma detection rate, AI, artificial intelligence, AMR, adenoma miss rate, ANN, artificial neural network, BE, Barrett’s esophagus, CAD, computer-aided diagnosis, CADe, CAD studies for colon polyp detection, CADx, CAD studies for colon polyp classification, CI, confidence interval, CNN, convolutional neural network, CRC, colorectal cancer, DL, deep learning, GI, gastroenterology, HD-WLE, high-definition white light endoscopy, HDWL, high-definition white light, ML, machine learning, NBI, narrow-band imaging, NPV, negative predictive value, PIVI, preservation and Incorporation of Valuable Endoscopic Innovations, SVM, support vector machine, VLE, volumetric laser endomicroscopy, WCE, wireless capsule endoscopy, WL, white light

General

Identifying patterns in urban housing density in developing countries using convolutional networks and satellite imagery.

In Heliyon

The use of Deep Neural Networks for remote sensing scene image analysis is growing fast. Despite this, data sets on developing countries are conspicuously absent in the public domain for benchmarking machine learning algorithms, rendering existing data sets unrepresentative. Secondly, current literature uses low-level semantic scene image class definitions, which may not have many relevant applications in certain domains. To examine these problems, we applied Convolutional Neural Networks (CNN) to high-level scene image classification for identifying patterns in urban housing density in a developing country setting. An end-to-end model training workflow is proposed for this purpose. A method for quantifying spatial extent of urban housing classes which gives insight into settlement patterns is also proposed. The method consists of computing the ratio between area covered by a given housing class and total area occupied by all classes. In the current work this method is implemented based on grid count, whereby the number of predicted grids for one housing class is divided by the total grid count for all classes. Results from the proposed method were validated against building density data computed on OpenStreetMap data. Our results for scene image classification are comparable to current state-of-the-art, despite focusing only on most difficult classes in those works. We also contribute a new satellite scene image data set that captures some general characteristics of urban housing in developing countries. The data set has similar but also some distinct attributes to existing data sets.

Sanya Rahman, Mwebaze Ernest

2020-Dec

Computer science, Convolutional neural networks, Developing countries, Housing classification, Satellite imagery, Urban areas

General

A dataset for automatic violence detection in videos.

In Data in brief

The automatic detection of violence and crimes in videos is gaining attention, specifically as a tool to unburden security officers and authorities from the need to watch hours of footages to identify event lasting few seconds. So far, most of the available datasets was composed of few clips, in low resolution, often built on too specific cases (e.g. hockey fight). While high resolution datasets are emerging, there is still the need of datasets to test the robustness of violence detection techniques to false positives, due to behaviours which might resemble violent actions. To this end, we propose a dataset composed of 350 clips (MP4 video files, 1920 × 1080 pixels, 30 fps), labelled as non-violent (120 clips) when representing non-violent behaviours, and violent (230 clips) when representing violent behaviours. In particular, the non-violent clips include behaviours (hugs, claps, exulting, etc.) that can cause false positives in the violence detection task, due to fast movements and the similarity with violent behaviours. The clips were performed by non-professional actors, varying from 2 to 4 per clip.

Bianculli Miriana, Falcionelli Nicola, Sernani Paolo, Tomassini Selene, Contardo Paolo, Lombardi Mara, Dragoni Aldo Franco

2020-Dec

Computer vision, Crime detection, Deep learning, Violence detection

General

Helminth Egg Automatic Detector (HEAD): Improvements in development for digital identification and quantification of Helminth eggs and its application online.

In MethodsX

Conventional analytical techniques for evaluating Helminth eggs are based on different steps to concentrate them in a pellet for direct observation and quantification under a light microscope, which can generate under-counts or over-counts and be time consuming. To enhance this process, a new approach via automatic identification was implemented in which various image processing detectors were developed and incorporated into a Helminth Egg Automatic Detector (HEAD) system. This allowed the identification and quantification of pathogenic eggs of global medical importance. More than 2.6 billion people are currently affected and infected, and this results in approximately 80,000 child deaths each year. As a result, since 1980 the World Health Organization (WHO) has implemented guidelines, regulations and criteria for the control of the health risk. After the initial release of the analytical technique, two improvements were developed in the detector: first, a texture verification process that reduced the number of false positive results; and second, the establishment of the optimal thresholds for each species. In addition, the software was made available on a free platform. After performing an internal statistical verification of the system, testing with internationally recognized parasitology laboratories was carried out, Subsequently, the HEAD System is capable of identifying and quantifying different species of Helminth eggs in different environmental samples: wastewater, sludge, biosolids, excreta and soil, with in-service sensitivity and specificity values for the open library for machine learning TensorFlow (TF) model of 96.82% and 97.96% respectively. The current iteration uses AutoML Vision (a computer platform for the automatization of machine learning models, making it easier to train, optimize and export results to cloud applications or devices). It represents a useful and cheap tool that could be utilized by environmental monitoring facilities and laboratories around the world.•The HEAD Software will significantly reduce the costs associated with the detection and quantification of helminth eggs to a high level of accuracy.•It represents a tool, not only for microbiologists and researchers, but also for various agencies involved in sanitation, such as environmental regulation agencies, which currently require highly trained technicians.•The simplicity of the device contributes to the control the contamination of water, soil, and crops, even in poor and isolated communities.

Jiménez Blanca, Maya Catalina, Velásquez Gustavo, Barrios José Antonio, Pérez Mónica, Román Angélica

2020

AutoML vision, Automatic identification, Environmental samples, Helminth eggs, Object characterization, Sensitivity, Specificity, TensorFlow

General

Machine Learning Models for covid-19 future forecasting.

In Materials today. Proceedings

Computational methods for machine learning (ML) have shown their meaning for the projection of potential results for informed decisions. Machine learning algorithms have been applied for a long time in many applications requiring the detection of adverse risk factors. This study shows the ability to predict the number of individuals who are affected by the COVID-19[1] as a potential threat to human beings by ML modelling. In this analysis, the risk factors of COVID-19 were exponential smoothing (ES). The Lower Absolute Reductor and Selection Operator, (LASSo), Vector Assistance (SVM), four normal potential forecasts, such as Linear Regression (LR)). [2] Each of these machine-learning models has three distinct kinds of predictions: the number of newly infected COVID 19 people, mortality rates and the recovered COVID-19 estimates in the next 10 days. These approaches are better used in the latest COVID-19 situation, as shown by the findings of the analysis. The LR, that is effective in predicting new cases of corona, death numbers and recovery.

Mojjada Ramesh Kumar, Yadav Arvind, Prabhu A V, Natarajan Yuvaraj

2020-Dec-09

COVID-19, R2 score adjusted, exponential process of smoothing, future forecasting, machine learning supervised

General

Modeling and analysis of different scenarios for the spread of COVID-19 by using the modified multi-agent systems - Evidence from the selected countries.

In Results in physics

Currently, there is a global pandemic of COVID-19. To assess its prevalence, it is necessary to have adequate models that allow real-time modeling of the impact of various quarantine measures by the state. The SIR model, which is implemented using a multi-agent system based on mobile cellular automata, was improved. The paper suggests ways to improve the rules of the interaction and behavior of agents. Methods of comparing the parameters of the SIR model with real geographical, social and medical indicators have been developed. That allows the modeling of the spatial distribution of COVID-19 as a single location and as the whole country consisting of individual regions that interact with each other by transport, taking into account factors such as public transport, supermarkets, schools, universities, gyms, churches, parks. The developed model also allows us to assess the impact of quarantine, restrictions on transport connections between regions, to take into account such factors as the incubation period, the mask regime, maintaining a safe distance between people, and so on. A number of experiments were conducted in the work, which made it possible to assess both the impact of individual measures to stop the pandemic and their comprehensive application. A method of comparing computer-time and dynamic parameters of the model with real data is proposed, which allowed assessing the effectiveness of the government in stopping the pandemic in the Chernivtsi region, Ukraine. A simulation of the pandemic spread in countries such as Slovakia, Turkey and Serbia was also conducted. The calculations showed the high-accuracy matching of the forecast model with real data.

Vyklyuk Yaroslav, Manylich Mykhailo, Škoda Miroslav, Radovanović Milan M, Petrović Marko D

2020-Dec-09

COVID-19, forecasting, modified multi-agent systems, public activities, simulations

General

How artificial intelligence and machine learning can help healthcare systems respond to COVID-19.

In Machine learning

The COVID-19 global pandemic is a threat not only to the health of millions of individuals, but also to the stability of infrastructure and economies around the world. The disease will inevitably place an overwhelming burden on healthcare systems that cannot be effectively dealt with by existing facilities or responses based on conventional approaches. We believe that a rigorous clinical and societal response can only be mounted by using intelligence derived from a variety of data sources to better utilize scarce healthcare resources, provide personalized patient management plans, inform policy, and expedite clinical trials. In this paper, we introduce five of the most important challenges in responding to COVID-19 and show how each of them can be addressed by recent developments in machine learning (ML) and artificial intelligence (AI). We argue that the integration of these techniques into local, national, and international healthcare systems will save lives, and propose specific methods by which implementation can happen swiftly and efficiently. We offer to extend these resources and knowledge to assist policymakers seeking to implement these techniques.

van der Schaar Mihaela, Alaa Ahmed M, Floto Andres, Gimson Alexander, Scholtes Stefan, Wood Angela, McKinney Eoin, Jarrett Daniel, Lio Pietro, Ercole Ari

2020-Dec-09

COVID-19, Clinical decision support, Healthcare

General

Information Technology Solutions, Challenges, and Suggestions for Tackling the COVID-19 Pandemic.

In International journal of information management

Various technology innovations and applications have been developed to fight the coronavirus pandemic. The pandemic also has implications for the design, development, and use of technologies. There is an urgent need for a greater understanding of what roles information systems and technology researchers can play in this global pandemic. This paper examines emerging technologies used to mitigate the threats of COVID-19 and relevant challenges related to technology design, development, and use. It also provides insights and suggestions into how information systems and technology scholars can help fight the COVID-19 pandemic. This paper helps promote future research and technology development to produce better solutions for tackling the COVID-19 pandemic and future pandemics.

He Wu, Zhang Justin, Li Wenzhuo

2020-Dec-09

Artificial Intelligence, Big Data, Blockchain, COVID-19, Digital Divide, Human Behavior, Information Systems, System Integration

General

Patients' perceptions of teleconsultation during COVID-19: A cross-national study.

In Technological forecasting and social change

In recent months, humanity has had to deal with a worldwide pandemic called COVID-19, which has caused the death of hundreds of thousands of people and paralyzed the global economy. Struggling to cure infected patients while continuing to care for patients with other pathologies, health authorities have faced the lack of medical staff and infrastructure. This study aimed to investigate the acceptance of teleconsultation solutions by patients, which help to avoid the spread of the disease during this pandemic period. The model was built using some constructs of the technology acceptance model UTAUT2, Personal traits, Availability, and Perceived Risks. A new scale on Contamination Avoidance was developed by the authors. The questionnaire was disseminated in several countries in Europe and Asia and a total sample of 386 respondents was collected. The results emphasize the huge impact of Performance Expectancy, the negative influence of Perceived Risk, and the positive influence of Contamination Avoidance on the adoption of teleconsultation solutions. The findings highlight the moderating effects of Age, Gender, and Country.

Baudier Patricia, Kondrateva Galina, Ammi Chantal, Chang Victor, Schiavone Francesco

2020-Dec-07

Acceptance, COVID-19, Pandemic, Teleconsultation, Telemedicine

General

Literature on Applied Machine Learning in Metagenomic Classification: A Scoping Review.

In Biology

Applied machine learning in bioinformatics is growing as computer science slowly invades all research spheres. With the arrival of modern next-generation DNA sequencing algorithms, metagenomics is becoming an increasingly interesting research field as it finds countless practical applications exploiting the vast amounts of generated data. This study aims to scope the scientific literature in the field of metagenomic classification in the time interval 2008-2019 and provide an evolutionary timeline of data processing and machine learning in this field. This study follows the scoping review methodology and PRISMA guidelines to identify and process the available literature. Natural Language Processing (NLP) is deployed to ensure efficient and exhaustive search of the literary corpus of three large digital libraries: IEEE, PubMed, and Springer. The search is based on keywords and properties looked up using the digital libraries' search engines. The scoping review results reveal an increasing number of research papers related to metagenomic classification over the past decade. The research is mainly focused on metagenomic classifiers, identifying scope specific metrics for model evaluation, data set sanitization, and dimensionality reduction. Out of all of these subproblems, data preprocessing is the least researched with considerable potential for improvement.

Tonkovic Petar, Kalajdziski Slobodan, Zdravevski Eftim, Lameski Petre, Corizzo Roberto, Pires Ivan Miguel, Garcia Nuno M, Loncar-Turukalo Tatjana, Trajkovik Vladimir

2020-Dec-09

classification, data preprocessing, metagenomics, scoping review

General

Information Technology Solutions, Challenges, and Suggestions for Tackling the COVID-19 Pandemic.

In International journal of information management

Various technology innovations and applications have been developed to fight the coronavirus pandemic. The pandemic also has implications for the design, development, and use of technologies. There is an urgent need for a greater understanding of what roles information systems and technology researchers can play in this global pandemic. This paper examines emerging technologies used to mitigate the threats of COVID-19 and relevant challenges related to technology design, development, and use. It also provides insights and suggestions into how information systems and technology scholars can help fight the COVID-19 pandemic. This paper helps promote future research and technology development to produce better solutions for tackling the COVID-19 pandemic and future pandemics.

He Wu, Zhang Justin, Li Wenzhuo

2020-Dec-09

Artificial Intelligence, Big Data, Blockchain, COVID-19, Digital Divide, Human Behavior, Information Systems, System Integration

General

Cellpose: a generalist algorithm for cellular segmentation.

In Nature methods ; h5-index 152.0

Many biological applications require the segmentation of cell bodies, membranes and nuclei from microscopy images. Deep learning has enabled great progress on this problem, but current methods are specialized for images that have large training datasets. Here we introduce a generalist, deep learning-based segmentation method called Cellpose, which can precisely segment cells from a wide range of image types and does not require model retraining or parameter adjustments. Cellpose was trained on a new dataset of highly varied images of cells, containing over 70,000 segmented objects. We also demonstrate a three-dimensional (3D) extension of Cellpose that reuses the two-dimensional (2D) model and does not require 3D-labeled data. To support community contributions to the training data, we developed software for manual labeling and for curation of the automated results. Periodically retraining the model on the community-contributed data will ensure that Cellpose improves constantly.

Stringer Carsen, Wang Tim, Michaelos Michalis, Pachitariu Marius

2020-Dec-14

Oncology

Distinction of surgically resected gastrointestinal stromal tumor by near-infrared hyperspectral imaging.

In Scientific reports ; h5-index 158.0

The diagnosis of gastrointestinal stromal tumor (GIST) using conventional endoscopy is difficult because submucosal tumor (SMT) lesions like GIST are covered by a mucosal layer. Near-infrared hyperspectral imaging (NIR-HSI) can obtain optical information from deep inside tissues. However, far less progress has been made in the development of techniques for distinguishing deep lesions like GIST. This study aimed to investigate whether NIR-HSI is suitable for distinguishing deep SMT lesions. In this study, 12 gastric GIST lesions were surgically resected and imaged with an NIR hyperspectral camera from the aspect of the mucosal surface. Thus, the images were obtained ex-vivo. The site of the GIST was defined by a pathologist using the NIR image to prepare training data for normal and GIST regions. A machine learning algorithm, support vector machine, was then used to predict normal and GIST regions. Results were displayed using color-coded regions. Although 7 specimens had a mucosal layer (thickness 0.4-2.5 mm) covering the GIST lesion, NIR-HSI analysis by machine learning showed normal and GIST regions as color-coded areas. The specificity, sensitivity, and accuracy of the results were 73.0%, 91.3%, and 86.1%, respectively. The study suggests that NIR-HSI analysis may potentially help distinguish deep lesions.

Sato Daiki, Takamatsu Toshihiro, Umezawa Masakazu, Kitagawa Yuichi, Maeda Kosuke, Hosokawa Naoki, Okubo Kyohei, Kamimura Masao, Kadota Tomohiro, Akimoto Tetsuo, Kinoshita Takahiro, Yano Tomonori, Kuwata Takeshi, Ikematsu Hiroaki, Takemura Hiroshi, Yokota Hideo, Soga Kohei

2020-Dec-14

General

Liquid temperature prediction in bubbly flow using ant colony optimization algorithm in the fuzzy inference system as a trainer.

In Scientific reports ; h5-index 158.0

In the current research paper a novel hybrid model combining first-principle and artificial intelligence (AI) was developed for simulation of a chemical reactor. We study a 2-dimensional reactor with heating sources inside it by using computational fluid dynamics (CFD). The type of considered reactor is bubble column reactor (BCR) in which a two-phase system is created. Results from CFD were analyzed in two different stages. The first stage, which is the learning stage, takes advantage of the swarm intelligence of the ant colony. The second stage results from the first stage, and in this stage, the predictions are according to the previous stage. This stage is related to the fuzzy logic system, and the ant colony optimization learning framework is build-up this part of the model. Ants movements or swarm intelligence of ants lead to the optimization of physical, chemical, or any kind of processes in nature. From point to point optimization, we can access a kind of group optimization, meaning that a group of data is studied and optimized. In the current study, the swarm intelligence of ants was used to learn the data from CFD in different parts of the BCR. The learning was also used to map the input and output data and find out the complex connection between the parameters. The results from mapping the input and output data show the full learning framework. By using the AI framework, the learning process was transferred into the fuzzy logic process through membership function specifications; therefore, the fuzzy logic system could predict a group of data. The results from the swarm intelligence of ants and fuzzy logic suitably adapt to CFD results. Also, the ant colony optimization fuzzy inference system (ACOFIS) model is employed to predict the temperature distribution in the reactor based on the CFD results. The results indicated that instead of solving Navier-Stokes equations and complex solving procedures, the swarm intelligence could be used to predict a process. For better comparisons and assessment of the ACOFIS model, this model is compared with the genetic algorithm fuzzy inference system (GAFIS) and Particle swarm optimization fuzzy inference system (PSOFIS) method with regards to model accuracy, pattern recognition, and prediction capability. All models are at a similar level of accuracy and prediction ability, and the prediction time for all models is less than one second. The results show that the model's accuracy with low computational learning time can be achieved with the high number of CIR (0.5) when the number of inputs ≥ 4. However, this finding is vice versa, when the number of inputs < 4. In this case, the CIR number should be 0.2 to achieve the best accuracy of the model. This finding could also highlight the importance of sensitivity analysis of tuning parameters to achieve an accurate model with a cost-effective computational run.

Babanezhad Meisam, Behroyan Iman, Nakhjiri Ali Taghvaie, Marjani Azam, Heydarinasab Amir, Shirazian Saeed

2020-Dec-14

General

Fully automatic wound segmentation with deep convolutional neural networks.

In Scientific reports ; h5-index 158.0

Acute and chronic wounds have varying etiologies and are an economic burden to healthcare systems around the world. The advanced wound care market is expected to exceed $22 billion by 2024. Wound care professionals rely heavily on images and image documentation for proper diagnosis and treatment. Unfortunately lack of expertise can lead to improper diagnosis of wound etiology and inaccurate wound management and documentation. Fully automatic segmentation of wound areas in natural images is an important part of the diagnosis and care protocol since it is crucial to measure the area of the wound and provide quantitative parameters in the treatment. Various deep learning models have gained success in image analysis including semantic segmentation. This manuscript proposes a novel convolutional framework based on MobileNetV2 and connected component labelling to segment wound regions from natural images. The advantage of this model is its lightweight and less compute-intensive architecture. The performance is not compromised and is comparable to deeper neural networks. We build an annotated wound image dataset consisting of 1109 foot ulcer images from 889 patients to train and test the deep learning models. We demonstrate the effectiveness and mobility of our method by conducting comprehensive experiments and analyses on various segmentation neural networks. The full implementation is available at https://github.com/uwm-bigdata/wound-segmentation .

Wang Chuanbo, Anisuzzaman D M, Williamson Victor, Dhar Mrinal Kanti, Rostami Behrouz, Niezgoda Jeffrey, Gopalakrishnan Sandeep, Yu Zeyun

2020-Dec-14

Pathology

Effectiveness of transfer learning for enhancing tumor classification with a convolutional neural network on frozen sections.

In Scientific reports ; h5-index 158.0

Fast and accurate confirmation of metastasis on the frozen tissue section of intraoperative sentinel lymph node biopsy is an essential tool for critical surgical decisions. However, accurate diagnosis by pathologists is difficult within the time limitations. Training a robust and accurate deep learning model is also difficult owing to the limited number of frozen datasets with high quality labels. To overcome these issues, we validated the effectiveness of transfer learning from CAMELYON16 to improve performance of the convolutional neural network (CNN)-based classification model on our frozen dataset (N = 297) from Asan Medical Center (AMC). Among the 297 whole slide images (WSIs), 157 and 40 WSIs were used to train deep learning models with different dataset ratios at 2, 4, 8, 20, 40, and 100%. The remaining, i.e., 100 WSIs, were used to validate model performance in terms of patch- and slide-level classification. An additional 228 WSIs from Seoul National University Bundang Hospital (SNUBH) were used as an external validation. Three initial weights, i.e., scratch-based (random initialization), ImageNet-based, and CAMELYON16-based models were used to validate their effectiveness in external validation. In the patch-level classification results on the AMC dataset, CAMELYON16-based models trained with a small dataset (up to 40%, i.e., 62 WSIs) showed a significantly higher area under the curve (AUC) of 0.929 than those of the scratch- and ImageNet-based models at 0.897 and 0.919, respectively, while CAMELYON16-based and ImageNet-based models trained with 100% of the training dataset showed comparable AUCs at 0.944 and 0.943, respectively. For the external validation, CAMELYON16-based models showed higher AUCs than those of the scratch- and ImageNet-based models. Model performance for slide feasibility of the transfer learning to enhance model performance was validated in the case of frozen section datasets with limited numbers.

Kim Young-Gon, Kim Sungchul, Cho Cristina Eunbee, Song In Hye, Lee Hee Jin, Ahn Soomin, Park So Yeon, Gong Gyungyub, Kim Namkug

2020-Dec-14

General

Large-scale mapping of live corals to guide reef conservation.

In Proceedings of the National Academy of Sciences of the United States of America

Coral is the life-form that underpins the habitat of most tropical reef ecosystems, thereby supporting biological diversity throughout the marine realm. Coral reefs are undergoing rapid change from ocean warming and nearshore human activities, compromising a myriad of services provided to societies including coastal protection, fishing, and cultural practices. In the face of these challenges, large-scale operational mapping of live coral cover within and across reef ecosystems could provide more opportunities to address reef protection, resilience, and restoration at broad management- and policy-relevant scales. We developed an airborne mapping approach combining laser-guided imaging spectroscopy and deep learning models to quantify, at a large archipelago scale, the geographic distribution of live corals to 16-m water depth throughout the main Hawaiian islands. Airborne estimates of live coral cover were highly correlated with field-based estimates of live coral cover (R² = 0.94). Our maps were used to assess the relative condition of reefs based on live coral, and to identify potential coral refugia in the face of human-driven stressors, including marine heat waves. Geospatial modeling revealed that water depth, wave power, and nearshore development accounted for the majority (>60%) of live coral cover variation, but other human-driven factors were also important. Mapped interisland and intraisland variation in live coral location improves our understanding of reef geography and its human impacts, thereby guiding environmental management for reef resiliency.

Asner Gregory P, Vaughn Nicholas R, Heckler Joseph, Knapp David E, Balzotti Christopher, Shafron Ethan, Martin Roberta E, Neilson Brian J, Gove Jamison M

2020-Dec-14

Hawaiian Islands, coral mapping, coral reef, coral refugia, reef restoration

General

Predicting long-term dynamics of soil salinity and sodicity on a global scale.

In Proceedings of the National Academy of Sciences of the United States of America

Knowledge of spatiotemporal distribution and likelihood of (re)occurrence of salt-affected soils is crucial to our understanding of land degradation and for planning effective remediation strategies in face of future climatic uncertainties. However, conventional methods used for tracking the variability of soil salinity/sodicity are extensively localized, making predictions on a global scale difficult. Here, we employ machine-learning techniques and a comprehensive set of climatic, topographic, soil, and remote sensing data to develop models capable of making predictions of soil salinity (expressed as electrical conductivity of saturated soil extract) and sodicity (measured as soil exchangeable sodium percentage) at different longitudes, latitudes, soil depths, and time periods. Using these predictive models, we provide a global-scale quantitative and gridded dataset characterizing different spatiotemporal facets of soil salinity and sodicity variability over the past four decades at a ∼1-km resolution. Analysis of this dataset reveals that a soil area of 11.73 Mkm² located in nonfrigid zones has been salt-affected with a frequency of reoccurrence in at least three-fourths of the years between 1980 and 2018, with 0.16 Mkm² of this area being croplands. Although the net changes in soil salinity/sodicity and the total area of salt-affected soils have been geographically highly variable, the continents with the highest salt-affected areas are Asia (particularly China, Kazakhstan, and Iran), Africa, and Australia. The proposed method can also be applied for quantifying the spatiotemporal variability of other dynamic soil properties, such as soil nutrients, organic carbon content, and pH.

Hassani Amirhossein, Azapagic Adisa, Shokri Nima

2020-Dec-14

global scale modeling, machine learning, soil salinity, soil salinization, soil sodicity

General

Machine-learning-driven biomarker discovery for the discrimination between allergic and irritant contact dermatitis.

In Proceedings of the National Academy of Sciences of the United States of America

Contact dermatitis tremendously impacts the quality of life of suffering patients. Currently, diagnostic regimes rely on allergy testing, exposure specification, and follow-up visits; however, distinguishing the clinical phenotype of irritant and allergic contact dermatitis remains challenging. Employing integrative transcriptomic analysis and machine-learning approaches, we aimed to decipher disease-related signature genes to find suitable sets of biomarkers. A total of 89 positive patch-test reaction biopsies against four contact allergens and two irritants were analyzed via microarray. Coexpression network analysis and Random Forest classification were used to discover potential biomarkers and selected biomarker models were validated in an independent patient group. Differential gene-expression analysis identified major gene-expression changes depending on the stimulus. Random Forest classification identified CD47, BATF, FASLG, RGS16, SYNPO, SELE, PTPN7, WARS, PRC1, EXO1, RRM2, PBK, RAD54L, KIFC1, SPC25, PKMYT, HISTH1A, TPX2, DLGAP5, TPX2, CH25H, and IL37 as potential biomarkers to distinguish allergic and irritant contact dermatitis in human skin. Validation experiments and prediction performances on external testing datasets demonstrated potential applicability of the identified biomarker models in the clinic. Capitalizing on this knowledge, novel diagnostic tools can be developed to guide clinical diagnosis of contact allergies.

Fortino Vittorio, Wisgrill Lukas, Werner Paulina, Suomela Sari, Linder Nina, Jalonen Erja, Suomalainen Alina, Marwah Veer, Kero Mia, Pesonen Maria, Lundin Johan, Lauerma Antti, Aalto-Korte Kristiina, Greco Dario, Alenius Harri, Fyhrquist Nanna

2020-Dec-14

allergic contact dermatitis, artificial intelligence, biomarker, irritant contact dermatitis, machine learning

Oncology

Next-Generation Liquid Biopsies: Embracing Data Science in Oncology.

In Trends in cancer

Deeper and broader sequencing of circulating tumor DNA (ctDNA) has identified a wealth of cancer markers in the circulation, resulting in a paradigm shift towards data science-driven liquid biopsies in oncology. Although panel sequencing for actionable mutations in plasma is moving towards the clinic, the next generation of liquid biopsies is increasingly shifting from analyzing digital mutation signals towards analog signals, requiring a greater role for machine learning. Concomitantly, there is an increasing acceptance that these cancer signals do not have to arise from the tumor itself. In this Opinion, we discuss the opportunities and challenges arising from increasingly complex cancer liquid biopsy data.

Im Y R, Tsui D W Y, Diaz L A, Wan J C M

2020-Dec-11

cancer, cell-free DNA, circulating tumor DNA, liquid biopsy, oncology

Radiology

Automatic Detection and Segmentation of Breast Cancer on MRI Using Mask R-CNN Trained on Non-Fat-Sat Images and Tested on Fat-Sat Images.

In Academic radiology

RATIONALE AND OBJECTIVES : Computer-aided methods have been widely applied to diagnose lesions on breast magnetic resonance imaging (MRI). The first step was to identify abnormal areas. A deep learning Mask Regional Convolutional Neural Network (R-CNN) was implemented to search the entire set of images and detect suspicious lesions.

MATERIALS AND METHODS : Two DCE-MRI datasets were used, 241 patients acquired using non-fat-sat sequence for training, and 98 patients acquired using fat-sat sequence for testing. All patients have confirmed unilateral mass cancers. The tumor was segmented using fuzzy c-means clustering algorithm to serve as the ground truth. Mask R-CNN was implemented with ResNet-101 as the backbone. The neural network output the bounding boxes and the segmented tumor for evaluation using the Dice Similarity Coefficient (DSC). The detection performance, and the trade-off between sensitivity and specificity, was analyzed using free response receiver operating characteristic.

RESULTS : When the precontrast and subtraction image of both breasts were used as input, the false positive from the heart and normal parenchymal enhancements could be minimized. The training set had 1469 positive slices (containing lesion) and 9135 negative slices. In 10-fold cross-validation, the mean accuracy = 0.86 and DSC = 0.82. The testing dataset had 1568 positive and 7264 negative slices, with accuracy = 0.75 and DSC = 0.79. When the obtained per-slice results were combined, 240 of 241 (99.5%) lesions in the training and 98 of 98 (100%) lesions in the testing datasets were identified.

CONCLUSION : Deep learning using Mask R-CNN provided a feasible method to search breast MRI, localize, and segment lesions. This may be integrated with other artificial intelligence algorithms to develop a fully automatic breast MRI diagnostic system.

Zhang Yang, Chan Siwa, Park Vivian Youngjean, Chang Kai-Ting, Mehta Siddharth, Kim Min Jung, Combs Freddie J, Chang Peter, Chow Daniel, Parajuli Ritesh, Mehta Rita S, Lin Chin-Yao, Chien Sou-Hsin, Chen Jeon-Hor, Su Min-Ying

2020-Dec-11

Breast MRI, Deep learning, Fully-automatic detection, Mask R-CNN

Pathology

Deep Learning Predicts Underlying Features on Pathology Images with Therapeutic Relevance for Breast and Gastric Cancer.

In Cancers

DNA repair deficiency (DRD) is an important driver of carcinogenesis and an efficient target for anti-tumor therapies to improve patient survival. Thus, detection of DRD in tumors is paramount. Currently, determination of DRD in tumors is dependent on wet-lab assays. Here we describe an efficient machine learning algorithm which can predict DRD from histopathological images. The utility of this algorithm is demonstrated with data obtained from 1445 cancer patients. Our method performs rather well when trained on breast cancer specimens with homologous recombination deficiency (HRD), AUC (area under curve) = 0.80. Results for an independent breast cancer cohort achieved an AUC = 0.70. The utility of our method was further shown by considering the detection of mismatch repair deficiency (MMRD) in gastric cancer, yielding an AUC = 0.81. Our results demonstrate the capacity of our learning-base system as a low-cost tool for DRD detection.

Valieris Renan, Amaro Lucas, Osório Cynthia Aparecida Bueno de Toledo, Bueno Adriana Passos, Rosales Mitrowsky Rafael Andres, Carraro Dirce Maria, Nunes Diana Noronha, Dias-Neto Emmanuel, Silva Israel Tojal da

2020-Dec-09

DNA repair deficiency, biomarker, deep learning, digital pathology, mutational signature

General

Patients' perceptions of teleconsultation during COVID-19: A cross-national study.

In Technological forecasting and social change

In recent months, humanity has had to deal with a worldwide pandemic called COVID-19, which has caused the death of hundreds of thousands of people and paralyzed the global economy. Struggling to cure infected patients while continuing to care for patients with other pathologies, health authorities have faced the lack of medical staff and infrastructure. This study aimed to investigate the acceptance of teleconsultation solutions by patients, which help to avoid the spread of the disease during this pandemic period. The model was built using some constructs of the technology acceptance model UTAUT2, Personal traits, Availability, and Perceived Risks. A new scale on Contamination Avoidance was developed by the authors. The questionnaire was disseminated in several countries in Europe and Asia and a total sample of 386 respondents was collected. The results emphasize the huge impact of Performance Expectancy, the negative influence of Perceived Risk, and the positive influence of Contamination Avoidance on the adoption of teleconsultation solutions. The findings highlight the moderating effects of Age, Gender, and Country.

Baudier Patricia, Kondrateva Galina, Ammi Chantal, Chang Victor, Schiavone Francesco

2020-Dec-07

Acceptance, COVID-19, Pandemic, Teleconsultation, Telemedicine

Surgery

Coronary angiography image segmentation based on PSPNet.

In Computer methods and programs in biomedicine

PURPOSE : Coronary artery disease (CAD) is known to have high prevalence, high disability and mortality. The incidence and mortality of cardiovascular disease are also gradually increasing worldwide. Therefore, our paper proposes to use a more efficient image processing method to extract accurate vascular structures from vascular images by combining computer vision and deep learning.

METHOD : Our proposed segmentation of coronary angiography images based on PSPNet network was compared with FCN, and analyzed and discussed the experimental results using three evaluation indicators of precision, recall and Fl-score. Aiming at the complex and changeable structure of coronary angiography images and over-fitting or parameter structure destruction, we implemented the parallel multi-scale convolutional neural network model using PSPNet, using small sample transfer learning that limits parameter learning method.

RESULTS : The accuracy of our technique proposed in this paper is 0.957. The accuracy of PSPNet is 26.75% higher than the traditional algorithm and 4.59% higher than U-Net. The average segmentation accuracy of the PSPNet model using transfer learning on the test set increased from 0.926 to 0.936, the sensitivity increased from 0.846 to 0.865, and the specificity increased from 0.921 to 0.949. The segmentation effect in this paper is closest to the segmentation result of the human expert, and is smoother than that of U-Net segmentation.

CONCLUSION : The PSPNet network reduces manual interaction in diagnosis, reduces dependence on medical personnel, improves the efficiency of disease diagnosis, and provides auxiliary strategies for subsequent medical diagnosis systems based on cardiac coronary angiography.

Zhu Xiliang, Cheng Zhaoyun, Wang Sheng, Chen Xianjie, Lu Guoqing

2020-Dec-04

Coronary angiography images, blood vessel segmentation, deep learning, multi-scale convolutional neural network, transfer learning

General

A novel extended Kalman filter with support vector machine based method for the automatic diagnosis and segmentation of brain tumors.

In Computer methods and programs in biomedicine

BACKGROUND : Brain tumors are life-threatening, and their early detection is crucial for improving survival rates. Conventionally, brain tumors are detected by radiologists based on their clinical experience. However, this process is inefficient. This paper proposes a machine learning-based method to 1) determine the presence of a tumor, 2) automatically segment the tumor, and 3) classify it as benign or malignant.

METHODS : We implemented an Extended Kalman Filter with Support Vector Machine (EKF-SVM), an image analysis platform based on an SVM for automated brain tumor detection. A development dataset of 120 patients which supported by Tiantan Hospital was used for algorithm training. Our machine learning algorithm has 5 components as follows. Firstly, image standardization is applied to all the images. This is followed by noise removal with a non-local means filter, and contrast enhancement with improved dynamic histogram equalization. Secondly, a gray-level co-occurrence matrix is utilized for feature extraction to get the image features. Thirdly, the extracted features are fed into a SVM for classify the MRI initially, and an EKF is used to classify brain tumors in the brain MRIs. Fourthly, cross-validation is used to verify the accuracy of the classifier. Finally, an automatic segmentation method based on the combination of k-means clustering and region growth is used for detecting brain tumors.

RESULTS : With regard to the diagnostic performance, the EKF-SVM had a 96.05% accuracy for automatically classifying brain tumors. Segmentation based on k-means clustering was capable of identifying the tumor boundaries and extracting the whole tumor.

CONCLUSION : The proposed EKF-SVM based method has better classification performance for positive brain tumor images, which was mainly due to the dearth of negative examples in our dataset. Therefore, future work should obtain more negative examples and investigate the performance of deep learning algorithms such as the convolutional neural networks for automatic diagnosis and segmentation of brain tumors.

Chen Baoshi, Zhang Lingling, Chen Hongyan, Liang Kewei, Chen Xuzhu

2020-Oct-31

Automatic segmentation, Brain MRI diagnosis, Brain tumor segmentation, EKF-SVM, Image standardization

General

High-resolution analyses of human sperm dynamic methylome reveal thousands of novel age-related epigenetic alterations.

In Clinical epigenetics

BACKGROUND : Children of aged fathers are at a higher risk of developing mental disorders. Alterations in sperm DNA methylation have been implicated as a potential cause. However, age-dependent modifications of the germ cells' epigenome remain poorly understood. Our objective was to assess the DNA methylation profile of human spermatozoa during aging.

RESULTS : We used a high throughput, customized methylC-capture sequencing (MCC-seq) approach to characterize the dynamic DNA methylation in spermatozoa from 94 fertile and infertile men, who were categorized as young, 48 men between 18-38 years or old 46 men between 46-71 years. We identified more than 150,000 age-related CpG sites that are significantly differentially methylated among 2.65 million CpG sites covered. We conducted machine learning using our dataset to predict the methylation age of subjects; the age prediction accuracy based on our assay provided a more accurate prediction than that using the 450 K chip approach. In addition, we found that there are more hypermethylated (62%) than hypomethylated (38%) CpG sites in sperm of aged men, corresponding to 798 of total differential methylated regions (DMRs), of which 483 are hypermethylated regions (HyperDMR), and 315 hypomethylated regions (HypoDMR). Moreover, the distribution of age-related hyper- and hypomethylated CpGs in sperm is not random; the CpG sites that were hypermethylated with advanced age were frequently located in the distal region to genes, whereas hypomethylated sites were near to gene transcription start sites (TSS). We identified a high density of age-associated CpG changes in chromosomes 4 and 16, particularly HyperDMRs with localized clusters, the chr4 DMR cluster overlaps PGC1α locus, a protein involved in metabolic aging and the chr16 DMR cluster overlaps RBFOX1 locus, a gene implicated in neurodevelopmental disease. Gene ontology analysis revealed that the most affected genes by age were associated with development, neuron projection, differentiation and recognition, and behaviour, suggesting a potential link to the higher risk of neurodevelopmental disorders in children of aged fathers.

CONCLUSION : We identified thousands of age-related and sperm-specific epigenetic alterations. These findings provide novel insight in understanding human sperm DNA methylation dynamics during paternal aging, and the subsequently affected genes potentially related to diseases in offspring.

Cao Mingju, Shao Xiaojian, Chan Peter, Cheung Warren, Kwan Tony, Pastinen Tomi, Robaire Bernard

2020-Dec-14

Advanced paternal age, DNA methylation, Fertility, MCC-seq, Spermatozoa

Cardiology

A stacking-based model for predicting 30-day all-cause hospital readmissions of patients with acute myocardial infarction.

In BMC medical informatics and decision making ; h5-index 38.0

BACKGROUND : Acute myocardial infarction (AMI) is a serious cardiovascular disease, followed by a high readmission rate within 30-days of discharge. Accurate prediction of AMI readmission is a crucial way to identify the high-risk group and optimize the distribution of medical resources.

METHODS : In this study, we propose a stacking-based model to predict the risk of 30-day unplanned all-cause hospital readmissions for AMI patients based on clinical data. Firstly, we conducted an under-sampling method of neighborhood cleaning rule (NCR) to alleviate the class imbalance and then utilized a feature selection method of SelectFromModel (SFM) to select effective features. Secondly, we adopted a self-adaptive approach to select base classifiers from eight candidate models according to their performances in datasets. Finally, we constructed a three-layer stacking model in which layer 1 and layer 2 were base-layer and level 3 was meta-layer. The predictions of the base-layer were used to train the meta-layer in order to make the final forecast.

RESULTS : The results show that the proposed model exhibits the highest AUC (0.720), which is higher than that of decision tree (0.681), support vector machine (0.707), random forest (0.701), extra trees (0.709), adaBoost (0.702), bootstrap aggregating (0.704), gradient boosting decision tree (0.710) and extreme gradient enhancement (0.713).

CONCLUSION : It is evident that our model could effectively predict the risk of 30-day all cause hospital readmissions for AMI patients and provide decision support for the administration.

Zhang Zhen, Qiu Hang, Li Weihao, Chen Yucheng

2020-Dec-14

Acute myocardial infarction, Clinical data, Hospital readmission, Machine learning, Self-adaptive, Stacking-based model learning

Surgery

Prediction of in-hospital mortality in patients on mechanical ventilation post traumatic brain injury: machine learning approach.

In BMC medical informatics and decision making ; h5-index 38.0

BACKGROUND : The study aimed to introduce a machine learning model that predicts in-hospital mortality in patients on mechanical ventilation (MV) following moderate to severe traumatic brain injury (TBI).

METHODS : A retrospective analysis was conducted for all adult patients who sustained TBI and were hospitalized at the trauma center from January 2014 to February 2019 with an abbreviated injury severity score for head region (HAIS) ≥ 3. We used the demographic characteristics, injuries and CT findings as predictors. Logistic regression (LR) and Artificial neural networks (ANN) were used to predict the in-hospital mortality. Accuracy, area under the receiver operating characteristics curve (AUROC), precision, negative predictive value (NPV), sensitivity, specificity and F-score were used to compare the models` performance.

RESULTS : Across the study duration; 785 patients met the inclusion criteria (581 survived and 204 deceased). The two models (LR and ANN) achieved good performance with an accuracy over 80% and AUROC over 87%. However, when taking the other performance measures into account, LR achieved higher overall performance than the ANN with an accuracy and AUROC of 87% and 90.5%, respectively compared to 80.9% and 87.5%, respectively. Venous thromboembolism prophylaxis, severity of TBI as measured by abbreviated injury score, TBI diagnosis, the need for blood transfusion, heart rate upon admission to the emergency room and patient age were found to be the significant predictors of in-hospital mortality for TBI patients on MV.

CONCLUSIONS : Machine learning based LR achieved good predictive performance for the prognosis in mechanically ventilated TBI patients. This study presents an opportunity to integrate machine learning methods in the trauma registry to provide instant clinical decision-making support.

Abujaber Ahmad, Fadlalla Adam, Gammoh Diala, Abdelrahman Husham, Mollazehi Monira, El-Menyar Ayman

2020-Dec-14

Machine learning predictive model, Mechanical ventilation, Mortality, Traumatic brain injury

General

Selected articles from the Fourth International Workshop on Semantics-Powered Data Mining and Analytics (SEPDA 2019).

In BMC medical informatics and decision making ; h5-index 38.0

In this introduction, we first summarize the Fourth International Workshop on Semantics-Powered Data Mining and Analytics (SEPDA 2019) held on October 26, 2019 in conjunction with the 18th International Semantic Web Conference (ISWC 2019) in Auckland, New Zealand, and then briefly introduce seven research articles included in this supplement issue, covering the topics on Knowledge Graph, Ontology-Powered Analytics, and Deep Learning.

He Zhe, Tao Cui, Bian Jiang, Zhang Rui

2020-Dec-14

General

A deep learning approach for identifying cancer survivors living with post-traumatic stress disorder on Twitter.

In BMC medical informatics and decision making ; h5-index 38.0

BACKGROUND : Emotions after surviving cancer can be complicated. The survivors may have gained new strength to continue life, but some of them may begin to deal with complicated feelings and emotional stress due to trauma and fear of cancer recurrence. The widespread use of Twitter for socializing has been the alternative medium for data collection compared to traditional studies of mental health, which primarily depend on information taken from medical staff with their consent. These social media data, to a certain extent, reflect the users' psychological state. However, Twitter also contains a mix of noisy and genuine tweets. The process of manually identifying genuine tweets is expensive and time-consuming.

METHODS : We stream the data using cancer as a keyword to filter the tweets with cancer-free and use post-traumatic stress disorder (PTSD) related keywords to reduce the time spent on the annotation task. Convolutional Neural Network (CNN) learns the representations of the input to identify cancer survivors with PTSD.

RESULTS : The results present that the proposed CNN can effectively identify cancer survivors with PTSD. The experiments on real-world datasets show that our model outperforms the baselines and correctly classifies the new tweets.

CONCLUSIONS : PTSD is one of the severe anxiety disorders that could affect individuals who are exposed to traumatic events, including cancer. Cancer survivors are at risk of short-term or long-term effects on physical and psycho-social well-being. Therefore, the evaluation and treatment of PTSD are essential parts of cancer survivorship care. It will act as an alarming system by detecting the PTSD presence based on users' postings on Twitter.

Ismail Nur Hafieza, Liu Ninghao, Du Mengnan, He Zhe, Hu Xia

2020-Dec-14

Cancer survivor, Deep learning, PTSD, Social media

General

MINDWALC: mining interpretable, discriminative walks for classification of nodes in a knowledge graph.

In BMC medical informatics and decision making ; h5-index 38.0

BACKGROUND : Leveraging graphs for machine learning tasks can result in more expressive power as extra information is added to the data by explicitly encoding relations between entities. Knowledge graphs are multi-relational, directed graph representations of domain knowledge. Recently, deep learning-based techniques have been gaining a lot of popularity. They can directly process these type of graphs or learn a low-dimensional numerical representation. While it has been shown empirically that these techniques achieve excellent predictive performances, they lack interpretability. This is of vital importance in applications situated in critical domains, such as health care.

METHODS : We present a technique that mines interpretable walks from knowledge graphs that are very informative for a certain classification problem. The walks themselves are of a specific format to allow for the creation of data structures that result in very efficient mining. We combine this mining algorithm with three different approaches in order to classify nodes within a graph. Each of these approaches excels on different dimensions such as explainability, predictive performance and computational runtime.

RESULTS : We compare our techniques to well-known state-of-the-art black-box alternatives on four benchmark knowledge graph data sets. Results show that our three presented approaches in combination with the proposed mining algorithm are at least competitive to the black-box alternatives, even often outperforming them, while being interpretable.

CONCLUSIONS : The mining of walks is an interesting alternative for node classification in knowledge graphs. Opposed to the current state-of-the-art that uses deep learning techniques, it results in inherently interpretable or transparent models without a sacrifice in terms of predictive performance.

Vandewiele Gilles, Steenwinckel Bram, Turck Filip De, Ongenae Femke

2020-Dec-14

Data mining, Decision tree, Explainable AI, Feature extraction, Knowledge graphs, Random forest

Cardiology

Treatment effect prediction with adversarial deep learning using electronic health records.

In BMC medical informatics and decision making ; h5-index 38.0

BACKGROUND : Treatment effect prediction (TEP) plays an important role in disease management by ensuring that the expected clinical outcomes are obtained after performing specialized and sophisticated treatments on patients given their personalized clinical status. In recent years, the wide adoption of electronic health records (EHRs) has provided a comprehensive data source for intelligent clinical applications including the TEP investigated in this study.

METHOD : We examined the problem of using a large volume of heterogeneous EHR data to predict treatment effects and developed an adversarial deep treatment effect prediction model to address the problem. Our model employed two auto-encoders for learning the representative and discriminative features of both patient characteristics and treatments from EHR data. The discriminative power of the learned features was further enhanced by decoding the correlational information between the patient characteristics and subsequent treatments by means of a generated adversarial learning strategy. Thereafter, a logistic regression layer was appended on the top of the resulting feature representation layer for TEP.

RESULT : The proposed model was evaluated on two real clinical datasets collected from the cardiology department of a Chinese hospital. In particular, on acute coronary syndrome (ACS) dataset, the proposed adversarial deep treatment effect prediction (ADTEP) (0.662) exhibited 1.4, 2.2, and 6.3% performance gains in terms of the area under the ROC curve (AUC) over deep treatment effect prediction (DTEP) (0.653), logistic regression (LR) (0.648), and support vector machine (SVM) (0.621), respectively. As for heart failure (HF) case study, the proposed ADTEP also outperformed all benchmarks. The experimental results demonstrated that our proposed model achieved competitive performance compared to state-of-the-art models in tackling the TEP problem.

CONCLUSION : In this work, we propose a novel model to address the TEP problem by utilizing a large volume of observational data from EHR. With adversarial learning strategy, our proposed model can further explore the correlational information between patient statuses and treatments to extract more robust and discriminative representation of patient samples from their EHR data. Such representation finally benefits the model on TEP. The experimental results of two case studies demonstrate the superiority of our proposed method compared to state-of-the-art methods.

Chu Jiebin, Dong Wei, Wang Jinliang, He Kunlun, Huang Zhengxing

2020-Dec-14

Adversarial learning, Deep learning, Electronic health records, Treatment effect prediction

General

Flora Capture: a citizen science application for collecting structured plant observations.

In BMC bioinformatics

BACKGROUND : Digital plant images are becoming increasingly important. First, given a large number of images deep learning algorithms can be trained to automatically identify plants. Second, structured image-based observations provide information about plant morphological characteristics. Finally in the course of digitalization, digital plant collections receive more and more interest in schools and universities.

RESULTS : We developed a freely available mobile application called Flora Capture allowing users to collect series of plant images from predefined perspectives. These images, together with accompanying metadata, are transferred to a central project server where each observation is reviewed and validated by a team of botanical experts. Currently, more than 4800 plant species, naturally occurring in the Central European region, are covered by the application. More than 200,000 images, depicting more than 1700 plant species, have been collected by thousands of users since the initial app release in 2016.

CONCLUSION : Flora Capture allows experts, laymen and citizen scientists to collect a digital herbarium and share structured multi-modal observations of plants. Collected images contribute, e.g., to the training of plant identification algorithms, but also suit educational purposes. Additionally, presence records collected with each observation allow contribute to verifiable records of plant occurrences across the world.

Boho David, Rzanny Michael, Wäldchen Jana, Nitsche Fabian, Deggelmann Alice, Wittich Hans Christian, Seeland Marco, Mäder Patrick

2020-Dec-14

Citizen science, Digital herbariumn, Digital plant collection, Mobile app, Multi-organ plant identification, Structured plant observations

Public Health

An Aberration Detection-Based Approach for Sentinel Syndromic Surveillance of COVID-19 and Other Novel Influenza-Like Illnesses.

In Journal of biomedical informatics ; h5-index 55.0

Coronavirus Disease 2019 has emerged as a significant global concern, triggering harsh public health restrictions in a successful bid to curb its exponential growth. As discussion shifts towards relaxation of these restrictions, there is significant concern of second-wave resurgence. The key to managing these outbreaks is early detection and intervention, and yet there is a significant lag time associated with usage of laboratory confirmed cases for surveillance purposes. To address this, syndromic surveillance can be considered to provide a timelier alternative for first-line screening. Existing syndromic surveillance solutions are however typically focused around a known disease and have limited capability to distinguish between outbreaks of individual diseases sharing similar syndromes. This poses a challenge for surveillance of COVID-19 as its active periods tend to overlap temporally with other influenza-like illnesses. In this study we explore performing sentinel syndromic surveillance for COVID-19 and other influenza-like illnesses using a deep learning-based approach. Our methods are based on aberration detection utilizing autoencoders that leverages symptom prevalence distributions to distinguish outbreaks of two ongoing diseases that share similar syndromes, even if they occur concurrently. We first demonstrate that this approach works for detection of outbreaks of influenza, which has known temporal boundaries. We then demonstrate that the autoencoder can be trained to not alert on known and well-managed influenza-like illnesses such as the common cold and influenza. Finally, we applied our approach to 2019-2020 data in the context of a COVID-19 syndromic surveillance task to demonstrate how implementation of such a system could have provided early warning of an outbreak of a novel influenza-like illness that did not match the symptom prevalence profile of influenza and other known influenza-like illnesses.

Wen Andrew, Wang Liwei, He Huan, Liu Sijia, Fu Sunyang, Sohn Sunghwan, Kugel Jacob A, Kaggal Vinod C, Huang Ming, Wang Yanshan, Shen Feichen, Fan Jungwei, Liu Hongfang

2020-Dec-12

COVID-19, Deep Learning, Syndromic Surveillance

General

A dynamic general type-2 fuzzy system with optimized secondary membership for online frequency regulation.

In ISA transactions

This study suggests a new control system for frequency regulation in AC microgrids. Unlike to the most studies, challenging conditions such as variation of wind speed, multiple load disturbance, unknown dynamics and variable solar radiation are taken to account. To cope with uncertainties, a novel dynamic general type-2 (GT2) fuzzy logic system (FLS) by an optimized secondary membership is suggested. The secondary membership and rule parameters of dynamic GT2-FLS are online tuned through the adaptive optimization rules. The optimization rules are determined such that the robustness and stability to be guaranteed. Also, a new compensator is presented to tackle with estimation error and perturbations. The simulations verify that schemed controller outperforms than conventional methods.

Mohammadzadeh Ardashir, Sabzalian Mohammad Hosein, Ahmadian Ali, Nabipour Narjes

2020-Dec-07

AC microgrids, Frequency regulation, Fuzzy systems, Learning methods, Machine learning

General

RNN-VirSeeker: a deep learning method for identification of short viral sequences from metagenomes.

In IEEE/ACM transactions on computational biology and bioinformatics

Viruses are the most abundant biological entities on earth, and play vital roles in many aspects of microbial communities. As major human pathogens, viruses have caused huge mortality and morbidity to human society in history. Metagenomic sequencing methods could capture all microorganisms from microbiota, with sequences of viruses mixed with these of other species. Therefore, it is necessary to identify viral sequences from metagenomes. However, existing methods perform poorly on identifying short viral sequences. To solve this problem, a deep learning based method, RNN-VirSeeker, is proposed in this paper. RNN-VirSeeker was trained by sequences of 500bp sampled from known Virus and Host RefSeq genomes. Experimental results on the testing set have shown that RNN-VirSeeker exhibited AUROC of 0.9175, recall of 0.8640 and precision of 0.9211 for sequences of 500bp, and outperformed three widely used methods, VirSorter, VirFinder, and DeepVirFinder, on identifying short viral sequences. RNN-VirSeeker was also used to identify viral sequences from a CAMI dataset and a human gut metagenome. Compared with DeepVirFinder, RNN-VirSeeker identified more viral sequences from these metagenomes and achieved greater values of AUPRC and AUROC. RNN-VirSeeker is freely available at https://github.com/crazyinter/RNN-VirSeeker.

Liu Fu, Miao Yan, Liu Yun, Hou Tao

2020-Dec-14

General

Chartem: Reviving Chart Images with Data Embedding.

In IEEE transactions on visualization and computer graphics

In practice, charts are widely stored as bitmap images. Although easily consumed by humans, they are not convenient for other uses. For example, changing the chart style or type or a data value in a chart image practically requires creating a completely new chart, which is often a time-consuming and error-prone process. To assist these tasks, many approaches have been proposed to automatically extract information from chart images with computer vision and machine learning techniques. Although they have achieved promising preliminary results, there are still a lot of challenges to overcome in terms of robustness and accuracy. In this paper, we propose a novel alternative approach called Chartem to address this issue directly from the root. Specifically, we design a data-embedding schema to encode a significant amount of information into the background of a chart image without interfering human perception of the chart. The embedded information, when extracted from the image, can enable a variety of visualization applications to reuse or repurpose chart images. To evaluate the effectiveness of Chartem, we conduct a user study and performance experiments on Chartem embedding and extraction algorithms. We further present several prototype applications to demonstrate the utility of Chartem.

Fu Jiayun, Zhu Bin, Cui Weiwei, Ge Song, Wang Yun, Zhang Haidong, Huang He, Tang Yuanyuan, Zhang Dongmei, Ma Xiaojing

2020-Dec-14

General

Low Complexity Error Resilient HEVC Video Coding: A Deep Learning Approach.

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

Intra/inter switching-based error resilient video coding effectively enhances the robustness of video streaming when transmitting over error-prone networks. But it has a high computation complexity, due to the detailed end-to-end distortion prediction and brute-force search for rate-distortion optimization. In this paper, a Low Complexity Mode Switching based Error Resilient Encoding (LC-MSERE) method is proposed to reduce the complexity of the encoder through a deep learning approach. By designing and training multi-scale information fusion-based convolutional neural networks (CNN), intra and inter mode coding unit (CU) partitions can be predicted by the networks rapidly and accurately, instead of using brute-force search and a large number of end-to-end distortion estimations. In the intra CU partition prediction, we propose a spatial multi-scale information fusion based CNN (SMIF-Intra). In this network a shortcut convolution architecture is designed to learn the multi-scale and multi-grained image information, which is correlated with the CU partition. In the inter CU partition, we propose a spatial-temporal multi-scale information fusion-based CNN (STMIF-Inter), in which a two-stream convolution architecture is designed to learn the spatial-temporal image texture and the distortion propagation among frames. With information from the image, and coding and transmission parameters, the networks are able to accurately predict CU partitions for both intra and inter coding tree units (CTUs). Experiments show that our approach significantly reduces computation time for error resilient video encoding with acceptable quality decrement.

Wang Taiyu, Li Fan, Qiao Xiaoya, Cosman Pamela C

2020-Dec-14

General

Learning Spatial Attention for Face Super-Resolution.

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

General image super-resolution techniques have difficulties in recovering detailed face structures when applying to low resolution face images. Recent deep learning based methods tailored for face images have achieved improved performance by jointly trained with additional task such as face parsing and landmark prediction. However, multi-task learning requires extra manually labeled data. Besides, most of the existing works can only generate relatively low resolution face images (e.g., 128×128), and their applications are therefore limited. In this paper, we introduce a novel SPatial Attention Residual Network (SPARNet) built on our newly proposed Face Attention Units (FAUs) for face super-resolution. Specifically, we introduce a spatial attention mechanism to the vanilla residual blocks. This enables the convolutional layers to adaptively bootstrap features related to the key face structures and pay less attention to those less feature-rich regions. This makes the training more effective and efficient as the key face structures only account for a very small portion of the face image. Visualization of the attention maps shows that our spatial attention network can capture the key face structures well even for very low resolution faces (e.g., 16×16). Quantitative comparisons on various kinds of metrics (including PSNR, SSIM, identity similarity, and landmark detection) demonstrate the superiority of our method over current state-of-the-arts. We further extend SPARNet with multi-scale discriminators, named as SPARNetHD, to produce high resolution results (i.e., 512×512). We show that SPARNetHD trained with synthetic data cannot only produce high quality and high resolution outputs for synthetically degraded face images, but also show good generalization ability to real world low quality face images.

Chen Chaofeng, Gong Dihong, Wang Hao, Li Zhifeng, Wong Kwan-Yee K

2020-Dec-14

General

Insights into Algorithms for Separable Nonlinear Least Squares Problems.

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

Separable nonlinear least squares (SNLLS) problems have attracted interest in a wide range of research fields such as machine learning, computer vision, and signal processing. During the past few decades, several algorithms, including the joint optimization algorithm, alternated least squares (ALS) algorithm, embedded point iterations (EPI) algorithm, and variable projection (VP) algorithms, have been employed for solving SNLLS problems in the literature. The VP approach has been proven to be quite valuable for SNLLS problems and the EPI method has been successful in solving many computer vision tasks. However, no clear explanations about the intrinsic relationships of these algorithms have been provided in the literature. In this paper, we give some insights into these algorithms for SNLLS problems. We derive the relationships among different forms of the VP algorithms, EPI algorithm and ALS algorithm. In addition, the convergence and robustness of some algorithms are investigated. Moreover, the analysis of the VP algorithm generates a negative answer to Kaufman's conjecture. Numerical experiments on the image restoration task, fitting the time series data using the radial basis function network based autoregressive (RBF-AR) model, and bundle adjustment are given to compare the performance of different algorithms.

Chen Guang-Yong, Gan Min, Wang Shu-Qiang, Philip Chena C L

2020-Dec-14

General

Predictive Models on the Rise, But Do They Work for Health Care?

In IEEE pulse

Predictive models are designed to remove some of the subjectivity inherent in medical decision-making and to automate certain health-related services with the idea of improving the accuracy of diagnosis, providing personalized treatment options, and streamlining the health care industry overall. More and more of these models using approaches including machine learning are showing up for use in doctor's offices and hospitals, as well as in telemedicine applications, which have become prevalent with the growing demand for online alternatives to office visits.

Mertz Leslie

General

Identifying Patient Phenotype Cohorts Using Prehospital Electronic Health Record Data.

In Prehospital emergency care : official journal of the National Association of EMS Physicians and the National Association of State EMS Directors

Objective: Emergency medical services (EMS) provide critical interventions for patients with acute illness and injury and are important in implementing prehospital emergency care research. Retrospective, manual patient record review, the current reference-standard for identifying patient cohorts, requires significant time and financial investment. We developed automated classification models to identify eligible patients for prehospital clinical trials using EMS clinical notes and compared model performance to manual review. Methods: With eligibility criteria for an ongoing prehospital study of chest pain patients, we used EMS clinical notes (n = 1208) to manually classify patients as eligible, ineligible, and indeterminate. We randomly split these same records into training and test sets to develop and evaluate machine-learning (ML) algorithms using natural language processing (NLP) for feature (variable) selection. We compared models to the manual classification to calculate sensitivity, specificity, accuracy, positive predictive value, and F1 measure. We measured clinical expert time to perform review for manual and automated methods. Results: ML models' sensitivity, specificity, accuracy, positive predictive value, and F1 measure ranged from 0.93 to 0.98. Compared to manual classification (N = 363 records), the automated method excluded 90.9% of records as ineligible and leaving only 33 records for manual review. Conclusions: Our ML derived approach demonstrates the feasibility of developing a high-performing, automated classification system using EMS clinical notes to streamline the identification of a specific cardiac patient cohort. This efficient approach can be leveraged to facilitate prehospital patient-trial matching, patient phenotyping (i.e. influenza-like illness), and create prehospital patient registries.

Stemerman Rachel, Bunning Thomas, Grover Joseph, Kitzmiller Rebecca, Patel Mehul D

2020-Dec-14

machine learning, natural language processing, patient phenotype, prehospital

General

Disentangling temporal dynamics in attention bias from measurement error: A state-space modeling approach.

In Journal of abnormal psychology

Temporal dynamics in attention bias (AB) have gained increasing attention in recent years. It has been proposed that AB is variable over trials within a single test session of the dot-probe task, and that the variability in AB is more predictive of psychopathology than the traditional mean AB score. More important, one of the dynamics indices has shown better reliability than the traditional mean AB score. However, it has been also suggested that the dynamics indices are unable to uncouple random measurement error from true variability in AB, which questions the estimation precision of the dynamics indices. To clarify and overcome this issue, the current article introduces a state-space modeling (SSM) approach to estimate trial-level AB more accurately by filtering random measurement error. The estimation error of the extant dynamics indices versus SSM were evaluated by computer simulations with different parameter settings for the temporal variability and between-person variance in AB. Throughout the simulations, SSM showed robustly lower estimation error than the extant dynamics indices. We also applied these indices to real data sets, which revealed that the dynamics indices overestimate within-person variability relative to SSM. Here SSM indicated less temporal dynamics in AB than previously proposed. These findings suggest that SSM might be a better alternative to estimate trial level AB than the extant dynamics indices. However, it is still unclear whether AB has meaningful in-session variability that is predictive of psychopathology. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

Takano Keisuke, Taylor Charles T, Wittekind Charlotte E, Sakamoto Jiro, Ehring Thomas

2020-Dec-14

General

The Emerging Hazard of AI-Related Health Care Discrimination.

In The Hastings Center report

Artificial intelligence holds great promise for improved health-care outcomes. But it also poses substantial new hazards, including algorithmic discrimination. For example, an algorithm used to identify candidates for beneficial "high risk care management" programs routinely failed to select racial minorities. Furthermore, some algorithms deliberately adjust for race in ways that divert resources away from minority patients. To illustrate, algorithms have underestimated African Americans' risks of kidney stones and death from heart failure. Algorithmic discrimination can violate Title VI of the Civil Rights Act and Section 1557 of the Affordable Care Act when it unjustifiably disadvantages underserved populations. This article urges that both legal and technical tools be deployed to promote AI fairness. Plaintiffs should be able to assert disparate impact claims in health-care litigation, and Congress should enact an Algorithmic Accountability Act. In addition, fairness should be a key element in designing, implementing, validating, and employing AI.

Hoffman Sharona

2020-Dec-14

algorithmic fairness, artificial intelligence, civil rights, discrimination, disparate impact

General

Frontal electroencephalogram based drug, sex, and age independent sedation level prediction using non-linear machine learning algorithms.

In Journal of clinical monitoring and computing

Brain monitors which track quantitative electroencephalogram (EEG) signatures to monitor sedation levels are drug and patient specific. There is a need for robust sedation level monitoring systems to accurately track sedation levels across all drug classes, sex and age groups. Forty-four quantitative features estimated from a pooled dataset of 204 EEG recordings from 66 healthy adult volunteers who received either propofol, dexmedetomidine, or sevoflurane (all with and without remifentanil) were used in a machine learning based automated system to estimate the depth of sedation. Model training and evaluation were performed using leave-one-out cross validation methodology. We trained four machine learning models to predict sedation levels and evaluated the influence of remifentanil, age, and sex on the prediction performance. The area under the receiver-operator characteristic curve (AUC) was used to assess the performance of the prediction model. The ensemble tree with bagging outperformed other machine learning models and predicted sedation levels with an AUC = 0.88 (0.81-0.90). There were significant differences in the prediction probability of the automated systems when trained and tested across different age groups and sex. The performance of the EEG based sedation level prediction system is drug, sex, and age specific. Nonlinear machine-learning models using quantitative EEG features can accurately predict sedation levels. The results obtained in this study may provide a useful reference for developing next generation EEG based sedation level prediction systems using advanced machine learning algorithms. Clinical trial registration: NCT02043938 and NCT03143972.

Ramaswamy S M, Kuizenga M H, Weerink M A S, Vereecke H E M, Struys M M R F, Belur Nagaraj S

2020-Dec-14

Anaesthesia, Consciousness Monitors, Electroencephalogram, Machine learning, Medical informatics

Public Health

Paradigm shift in efforts to end TB by 2025.

In The Indian journal of tuberculosis

TB is a deadly infectious disease, in existence since time immemorial. This article traces the journey of TB developments in the last few decades and the path breaking moments that have accelerated the efforts towards Ending TB from National Tuberculosis Control Program (NTCP 1962-1992) to Revised National Tuberculosis Control Program (RNTCP - 1992-2019) and to National Tuberculosis Elimination Program (NTEP) as per the vision of Honorable Prime Minister of India. From increased funding for TB, the discovery of newer drugs and diagnostics, increased access to health facilities, greater investment in research and expanded reach of public health education, seasoned with TB activism and media's proactive role, private sector participation to political advocacy and community engagement, coupled with vaccine trials has renewed the hope of finding the elusive and miraculous breakthrough to END TB and it seems the goal is within the realms of the possibility. The recent paradigm shift in the policy and the drive of several states & UTs to move towards TB free status through rigorous population-based vulnerability mapping and screening coupled with active case finding is expected to act as the driving force to lead the country towards Ending TB by 2025. Continued investments in research, innovations and availability of more effective drugs and the vaccines will add to existing armamentarium towards Ending TB.

Sachdeva K S, Parmar Malik, Rao Raghuram, Chauhan Sandeep, Shah Vaibhav, Pirabu Ra, Balasubramaniam Deepak, Vadera Bhavin, Anand S, Mathew Manu, Solanki Hardik, Sundar V V

2020-Dec

NTEP, Paradigm, Policy, Tuberculosis

General

[Network medicine and health services research in urology].

In Der Urologe. Ausg. A

It is the aim of the Medical Informatics Funding Scheme and other national and local projects for digital networking in healthcare to facilitate the exchange and use of patient data between institutions in compliance with data protection regulations. This requires the integration of data from various sources-such as digital workplace systems, laboratory systems, picture archiving and communication (PAC) systems or tumor boards-into a data warehouse or research databases. Digital networking of service providers and research institutions will open access to high-performance and precision medicine (e.g., virtual molecular tumor boards) for even more patients, thereby providing data for basic and care research. Network medicine will establish the translational link between basic research (e.g., genome research) and patient care. Digitally integrated "real world" patient data will also facilitate a detailed analysis of health care and the quality of treatments.

Schönthaler M, Schlomm T

2020-Dec-14

Artificial intelligence, Genome, Medical informatics, Quality control, Register

Radiology

Hippocampal Subfields Alterations in Paediatric Patients with Post-Traumatic Stress Disorder.

In Social cognitive and affective neuroscience ; h5-index 61.0

The hippocampus, a key structure with distinct subfield functions, is strongly implicated in the pathophysiology of post-traumatic stress disorder (PTSD); however, few studies of hippocampus subfields in PTSD have focused on paediatric patients. We therefore investigated hippocampal subfield volume using an automated segmentation method, and explored subfield-centered functional connectivity aberrations related to the anatomical changes, in a homogenous population of traumatized children with and without PTSD. To investigate potential diagnostic value in individual patients, we used a machine learning approach to identify features with significant discriminative power for diagnosis of PTSD using random forest classifiers. Compared to controls, we found significant mean volume reductions of 8.4% and 9.7% in the right presubiculum and hippocampal tail in patients, respectively. These two subfields' volumes were the most significant contributors to group discrimination, with mean classification accuracy 69% and specificity 81%. These anatomical alterations, along with the altered functional connectivity between (pre)subiculum and inferior frontal gyrus, may underlie deficits in fear circuitry leading to dysfunction of fear extinction and episodic memory, causally important in post-traumatic symptoms such as hypervigilance and re-experience. For the first time, we suggest that hippocampal subfield volumes might be useful in discriminating traumatized children with and without PTSD.

Li Lei, Pan Nanfang, Zhang Lianqing, Lui Su, Huang Xiaoqi, Xu Xin, Wang Song, Lei Du, Li Lingjiang, Kemp Graham J, Gong Qiyong

2020-Dec-14

hippocampus, magnetic resonance imaging, post-traumatic stress disorder, psychoradiology, stress

General

A structural approach to vibrational properties ranging from crystals to disordered systems.

In Soft matter

Many scientists generally attribute the vibrational anomalies of disordered solids to the structural disorder, which, however, is still under intense debate. Here we conduct simulations on two-dimensional packings with a finite temperature, whose structure is tuned from a crystalline configuration to an amorphous one, then the amorphous from very dense state to a relatively loose state. By measuring the vibrational density of states and the reduced density of states, we clearly observe the evolution of the boson peak with the change of the disorder and volume fractions. Meanwhile, to understand the structural origin of this anomaly, we identify the soft regimes of all systems with a novel machine-learning method, where the "softness", a local structural quantity, is defined. Interestingly, we find a strong monotonic relationship between the shape of the boson peak and the softness as well as its spatial heterogeneity, suggesting that the softness of a system may be a new structural approach to the anomalous vibrational properties of amorphous solids.

Tan Xin, Guo Ying, Huang Duan, Zhang Ling

2020-Dec-14

General

[Assessing soil pH in Anhui Province based on different features mining methods combined with generalized boosted regression models].

In Ying yong sheng tai xue bao = The journal of applied ecology

We explored the application of different feature mining methods combined with genera-lized boosted regression models in digital soil mapping. Environmental covariates were selected by two feature selection methods i.e., recursive feature elimination and selection by filtering. Using the original environmental covariates and the selected optimal variable combination as independent varia-bles, soil pH prediction model of Anhui Province was established and mapped based on the genera-lized boosted regression model and random forest model. The results showed that both kinds of feature mining methods could effectively improve the accuracy of soil pH prediction by generalized boosted regression models and random forest model, and could reduce dimensionality. Compared with the random forest model, the prediction accuracy of the validation set of the generalized boosted regression model was slightly lower. In the training set, the accuracy of the generalized boosted regression models was much higher than that of the random forest model, with higher interpretation and better overall effect. The main parameters of the random forest model, ntree and mtry, had limi-ted effect on the model. Different parameters and their combination could affect the prediction accuracy of the generalized boosted regression models, and thus should be tuned before modeling. The results of spatial mapping showed that soil pH in Anhui Province showed a pattern of "south acid and north alkali".

Wang Shi-Hang, Lu Hong-Liang, Zhao Ming-Song, Zhou Ling-Mei

2020-Oct

Anhui Province, feature mining, generalized boosted regression models, machine learning, random forest, soil pH

General

The Pre-treatment Gut Microbiome is Associated with Lack of Response to Methotrexate in New Onset Rheumatoid Arthritis.

In Arthritis & rheumatology (Hoboken, N.J.)

OBJECTIVES : Although oral methotrexate (MTX) remains the anchor drug for RA, up to 50% of patients do not achieve a clinically adequate outcome. Concomitantly, there is a lack of prognostic tools for treatment response prior to drug initiation. Here we study whether inter-individual differences in the human gut microbiome can aid in the prediction of MTX efficacy in new-onset RA (NORA).

METHODS : 16S rRNA gene and shotgun metagenomic sequencing were performed on the baseline gut microbiomes of drug-naïve, NORA patients (n=26). Results were validated in an additional independent cohort (n=21). To gain insight into potential microbial mechanisms, ex vivo experiments coupled with metabolomics analysis evaluated the association between microbiome-driven MTX depletion and clinical response.

RESULTS : Our analysis revealed significant associations between the abundance of gut bacterial taxa and their genes with future clinical response, including orthologs related to purine and methotrexate metabolism. Machine learning techniques were applied to the metagenomic data, resulting in a microbiome-based model that predicts lack of response to MTX in an independent group of patients. Finally, MTX levels remaining after ex vivo incubation with distal gut samples from pre-treatment RA patients significantly correlated with the magnitude of future clinical response, suggesting a possible direct effect of the gut microbiome on MTX metabolism and treatment outcomes.

CONCLUSIONS : Together, these results provide the first step towards predicting lack of response to oral MTX in NORA patients and support the value of the gut microbiome as a possible prognostic tool and as a potential target in RA therapeutics.

Artacho Alejandro, Isaac Sandrine, Nayak Renuka, Flor-Duro Alejandra, Alexander Margaret, Koo Imhoi, Manasson Julia, Smith Philip B, Rosenthal Pamela, Homsi Yamen, Gulko Percio, Pons Javier, Puchades-Carrasco Leonor, Izmirly Peter, Patterson Andrew, Abramson Steven B, Pineda-Lucena Antonio, Turnbaugh Peter J, Ubeda Carles, Scher Jose U

2020-Dec-13

Oncology

Quantitative ultrasound radiomics in predicting recurrence for patients with node-positive head-neck squamous cell carcinoma treated with radical radiotherapy.

In Cancer medicine

This prospective study was conducted to investigate the role of quantitative ultrasound (QUS) radiomics in predicting recurrence for patients with node-positive head-neck squamous cell carcinoma (HNSCC) treated with radical radiotherapy (RT). The most prominent cervical lymph node (LN) was scanned with a clinical ultrasound device having central frequency of 6.5 MHz. Ultrasound radiofrequency data were processed to obtain 7 QUS parameters. Color-coded parametric maps were generated based on individual QUS spectral features corresponding to each of the smaller units. A total of 31 (7 primary QUS and 24 texture) features were obtained before treatment. All patients were treated with radical RT and followed according to standard institutional practice. Recurrence (local, regional, or distant) served as an endpoint. Three different machine learning classifiers with a set of maximally three features were used for model development and tested with leave-one-out cross-validation for nonrecurrence and recurrence groups. Fifty-one patients were included, with a median follow up of 38 months (range 7-64 months). Recurrence was observed in 17 patients. The best results were obtained using a k-nearest neighbor (KNN) classifier with a sensitivity, specificity, accuracy, and an area under curve of 76%, 71%, 75%, and 0.74, respectively. All the three features selected for the KNN model were texture features. The KNN-model-predicted 3-year recurrence-free survival was 81% and 40% in the predicted no-recurrence and predicted-recurrence groups, respectively. (p = 0.001). The pilot study demonstrates pretreatment QUS-radiomics can predict the recurrence group with an accuracy of 75% in patients with node-positive HNSCC. Clinical trial registration: clinicaltrials.gov.in identifier NCT03908684.

Dasgupta Archya, Fatima Kashuf, DiCenzo Daniel, Bhardwaj Divya, Quiaoit Karina, Saifuddin Murtuza, Karam Irene, Poon Ian, Husain Zain, Tran William T, Sannachi Lakshmanan, Czarnota Gregory J

2020-Dec-13

head-neck squamous cell carcinoma, machine learning, quantitative ultrasound, radiomics, radiotherapy, recurrence, texture analysis

General

High fidelity modeling of aerosol pathogen propagation in built environments with moving pedestrians.

In International journal for numerical methods in biomedical engineering

A high fidelity model for the propagation of pathogens via aerosols in the presence of moving pedestrians is proposed. The key idea is the tight coupling of computational fluid dynamics and computational crowd dynamics in order to capture the emission, transport and inhalation of pathogen loads in space and time. An example simulating pathogen propagation in a narrow corridor with moving pedestrians clearly shows the considerable effect that pedestrian motion has on airflow, and hence on pathogen propagation and potential infectivity. This article is protected by copyright. All rights reserved.

Löhner Rainald, Antil Harbir

2020-Dec-13

Aerosol Transmission, Computational Crowd Dynamics, Computational Fluid Dynamics, Viral Infection

Public Health

Prediction of brain age and cognitive age: Quantifying brain and cognitive maintenance in aging.

In Human brain mapping

The concept of brain maintenance refers to the preservation of brain integrity in older age, while cognitive reserve refers to the capacity to maintain cognition in the presence of neurodegeneration or aging-related brain changes. While both mechanisms are thought to contribute to individual differences in cognitive function among older adults, there is currently no "gold standard" for measuring these constructs. Using machine-learning methods, we estimated brain and cognitive age based on deviations from normative aging patterns in the Whitehall II MRI substudy cohort (N = 537, age range = 60.34-82.76), and tested the degree of correspondence between these constructs, as well as their associations with premorbid IQ, education, and lifestyle trajectories. In line with established literature highlighting IQ as a proxy for cognitive reserve, higher premorbid IQ was linked to lower cognitive age independent of brain age. No strong evidence was found for associations between brain or cognitive age and lifestyle trajectories from midlife to late life based on latent class growth analyses. However, post hoc analyses revealed a relationship between cumulative lifestyle measures and brain age independent of cognitive age. In conclusion, we present a novel approach to characterizing brain and cognitive maintenance in aging, which may be useful for future studies seeking to identify factors that contribute to brain preservation and cognitive reserve mechanisms in older age.

Anatürk Melis, Kaufmann Tobias, Cole James H, Suri Sana, Griffanti Ludovica, Zsoldos Enikő, Filippini Nicola, Singh-Manoux Archana, Kivimäki Mika, Westlye Lars T, Ebmeier Klaus P, de Lange Ann-Marie G

2020-Dec-14

aging, brain maintenance, cognitive reserve, lifestyle, machine learning, neuroimaging, trajectories

General

Text Mining Approaches for Postmarket Food Safety Surveillance Using Online Media.

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

Food contamination and food poisoning pose enormous risks to consumers across the world. As discussions of consumer experiences have spread through online media, we propose the use of text mining to rapidly screen online media for mentions of food safety hazards. We compile a large data set of labeled consumer posts spanning two major websites. Utilizing text mining and supervised machine learning, we identify unique words and phrases in online posts that identify consumers' interactions with hazardous food products. We compare our methods to traditional sentiment-based text mining. We assess performance in a high-volume setting, utilizing a data set of over 4 million online reviews. Our methods were 77-90% accurate in top-ranking reviews, while sentiment analysis was just 11-26% accurate. Moreover, we aggregate review-level results to make product-level risk assessments. A panel of 21 food safety experts assessed our model's hazard-flagged products to exhibit substantially higher risk than baseline products. We suggest the use of these tools to profile food items and assess risk, building a postmarket decision support system to identify hazardous food products. Our research contributes to the literature and practice by providing practical and inexpensive means for rapidly monitoring food safety in real time.

Goldberg David M, Khan Samee, Zaman Nohel, Gruss Richard J, Abrahams Alan S

2020-Dec-12

Food safety, online reviews, text mining

General

Machine learning predictions of pH in the glacial aquifer system, northern USA.

In Ground water

A boosted regression tree model was developed to predict pH conditions in three-dimensions throughout the glacial aquifer system of the contiguous United States using pH measurements in samples from 18,386 wells and predictor variables that represent aspects of the hydrogeologic setting. Model results indicate that the carbonate content of soils and aquifer materials strongly controls pH and, when coupled with long flow paths, results in the most alkaline conditions. Conversely, in areas where glacial sediments are thin and carbonate-poor, pH conditions remain acidic. At depths typical of drinking-water supplies, predicted pH > 7.5 - which is associated with arsenic mobilization - occurs more frequently than predicted pH < 6 - which is associated with water corrosivity and the mobilization of other trace elements. A novel aspect of this model was the inclusion of numerically based estimates of groundwater flow characteristics (age and flow path length) as predictor variables. The sensitivity of pH predictions to these variables was consistent with hydrologic understanding of groundwater flow systems and the geochemical evolution of groundwater quality. The model was not developed to provide precise estimates of pH at any given location. Rather, it can be used to more generally identify areas where contaminants may be mobilized into groundwater and where corrosivity issues may be of concern to prioritize areas for future groundwater monitoring. This article is protected by copyright. All rights reserved.

Stackelberg Paul E, Belitz Kenneth, Brown Craig J, Erickson Melinda L, Elliott Sarah M, Kauffman Leon J, Ransom Katherine M, Reddy James E

2020-Dec-11

General

Recent Advances of Artificial Intelligence in Cardiovascular Disease.

In Journal of biomedical nanotechnology

Cardiovascular disease (CVD) is one of the most serious health disorders with increasing prevalence and high morbidity and mortality. Although diagnosis and treatment of CVD have achieved huge breakthrough in recent years, it still needs additional enhancements, which result in the demand for new techniques. Artificial intelligence (AI) is an emerging science field that has been widely used to guide diseases diagnosis, evaluation and treatment. AI techniques are promising in CVD to explore novel pathogenic genes phenotype, guide optimal individualized therapeutic strategy, improve the management and quality of discharged patients, predict disease prognosis, and as adjuvant therapy tool. Thus, we summarize the latest application of AI techniques in clinical diagnosis, evaluation and treatment of CVD, aiming to provide novel beneficial evidence of AI and promote its application in CVD.

Chen Zhu, Xiao Changhu, Qiu Haihua, Tan Xinxing, Jin Lian, He Yi, Guo Yuan, He Nongyue

2020-Jul-01

Oncology

Second-order grey-scale texture analysis of pleural ultrasound images to differentiate acute respiratory distress syndrome and cardiogenic pulmonary edema.

In Journal of clinical monitoring and computing

Discriminating acute respiratory distress syndrome (ARDS) from acute cardiogenic pulmonary edema (CPE) may be challenging in critically ill patients. Aim of this study was to investigate if gray-level co-occurrence matrix (GLCM) analysis of lung ultrasound (LUS) images can differentiate ARDS from CPE. The study population consisted of critically ill patients admitted to intensive care unit (ICU) with acute respiratory failure and submitted to LUS and extravascular lung water monitoring, and of a healthy control group (HCG). A digital analysis of pleural line and subpleural space, based on the GLCM with second order statistical texture analysis, was tested. We prospectively evaluated 47 subjects: 16 with a clinical diagnosis of CPE, 8 of ARDS, and 23 healthy subjects. By comparing ARDS and CPE patients' subgroups with HCG, the one-way ANOVA models found a statistical significance in 9 out of 11 GLCM textural features. Post-hoc pairwise comparisons found statistical significance within each matrix feature for ARDS vs. CPE and CPE vs. HCG (P ≤ 0.001 for all). For ARDS vs. HCG a statistical significance occurred only in two matrix features (correlation: P = 0.005; homogeneity: P = 0.048). The quantitative method proposed has shown high diagnostic accuracy in differentiating normal lung from ARDS or CPE, and good diagnostic accuracy in differentiating CPE and ARDS. Gray-level co-occurrence matrix analysis of LUS images has the potential to aid pulmonary edemas differential diagnosis.

Brusasco Claudia, Santori Gregorio, Tavazzi Guido, Via Gabriele, Robba Chiara, Gargani Luna, Mojoli Francesco, Mongodi Silvia, Bruzzo Elisa, Trò Rosella, Boccacci Patrizia, Isirdi Alessandro, Forfori Francesco, Corradi Francesco

2020-Dec-12

Acute respiratory failure, Artificial intelligence, Computer aided diagnosis, Heart failure, Lung ultrasonography, Quantitative lung ultrasonography

General

High-throughput phenotyping with temporal sequences.

In Journal of the American Medical Informatics Association : JAMIA

OBJECTIVE : High-throughput electronic phenotyping algorithms can accelerate translational research using data from electronic health record (EHR) systems. The temporal information buried in EHRs is often underutilized in developing computational phenotypic definitions. This study aims to develop a high-throughput phenotyping method, leveraging temporal sequential patterns from EHRs.

MATERIALS AND METHODS : We develop a representation mining algorithm to extract 5 classes of representations from EHR diagnosis and medication records: the aggregated vector of the records (aggregated vector representation), the standard sequential patterns (sequential pattern mining), the transitive sequential patterns (transitive sequential pattern mining), and 2 hybrid classes. Using EHR data on 10 phenotypes from the Mass General Brigham Biobank, we train and validate phenotyping algorithms.

RESULTS : Phenotyping with temporal sequences resulted in a superior classification performance across all 10 phenotypes compared with the standard representations in electronic phenotyping. The high-throughput algorithm's classification performance was superior or similar to the performance of previously published electronic phenotyping algorithms. We characterize and evaluate the top transitive sequences of diagnosis records paired with the records of risk factors, symptoms, complications, medications, or vaccinations.

DISCUSSION : The proposed high-throughput phenotyping approach enables seamless discovery of sequential record combinations that may be difficult to assume from raw EHR data. Transitive sequences offer more accurate characterization of the phenotype, compared with its individual components, and reflect the actual lived experiences of the patients with that particular disease.

CONCLUSION : Sequential data representations provide a precise mechanism for incorporating raw EHR records into downstream machine learning. Our approach starts with user interpretability and works backward to the technology.

Estiri Hossein, Strasser Zachary H, Murphy Shawn N

2020-Dec-14

electronic health records, phenotyping, sequential pattern mining, temporal data representation

Oncology

Automated model versus treating physician for predicting survival time of patients with metastatic cancer.

In Journal of the American Medical Informatics Association : JAMIA

OBJECTIVE : Being able to predict a patient's life expectancy can help doctors and patients prioritize treatments and supportive care. For predicting life expectancy, physicians have been shown to outperform traditional models that use only a few predictor variables. It is possible that a machine learning model that uses many predictor variables and diverse data sources from the electronic medical record can improve on physicians' performance. For patients with metastatic cancer, we compared accuracy of life expectancy predictions by the treating physician, a machine learning model, and a traditional model.

MATERIALS AND METHODS : A machine learning model was trained using 14 600 metastatic cancer patients' data to predict each patient's distribution of survival time. Data sources included note text, laboratory values, and vital signs. From 2015-2016, 899 patients receiving radiotherapy for metastatic cancer were enrolled in a study in which their radiation oncologist estimated life expectancy. Survival predictions were also made by the machine learning model and a traditional model using only performance status. Performance was assessed with area under the curve for 1-year survival and calibration plots.

RESULTS : The radiotherapy study included 1190 treatment courses in 899 patients. A total of 879 treatment courses in 685 patients were included in this analysis. Median overall survival was 11.7 months. Physicians, machine learning model, and traditional model had area under the curve for 1-year survival of 0.72 (95% CI 0.63-0.81), 0.77 (0.73-0.81), and 0.68 (0.65-0.71), respectively.

CONCLUSIONS : The machine learning model's predictions were more accurate than those of the treating physician or a traditional model.

Gensheimer Michael F, Aggarwal Sonya, Benson Kathryn R K, Carter Justin N, Henry A Solomon, Wood Douglas J, Soltys Scott G, Hancock Steven, Pollom Erqi, Shah Nigam H, Chang Daniel T

2020-Dec-14

machine learning, natural language processing, neoplasms, prognosis, radiotherapy

General

DReSS: a method to quantitatively describe the influence of structural perturbations on state spaces of genetic regulatory networks.

In Briefings in bioinformatics

Structures of genetic regulatory networks are not fixed. These structural perturbations can cause changes to the reachability of systems' state spaces. As system structures are related to genotypes and state spaces are related to phenotypes, it is important to study the relationship between structures and state spaces. However, there is still no method can quantitively describe the reachability differences of two state spaces caused by structural perturbations. Therefore, Difference in Reachability between State Spaces (DReSS) is proposed. DReSS index family can quantitively describe differences of reachability, attractor sets between two state spaces and can help find the key structure in a system, which may influence system's state space significantly. First, basic properties of DReSS including non-negativity, symmetry and subadditivity are proved. Then, typical examples are shown to explain the meaning of DReSS and the differences between DReSS and traditional graph distance. Finally, differences of DReSS distribution between real biological regulatory networks and random networks are compared. Results show most structural perturbations in biological networks tend to affect reachability inside and between attractor basins rather than to affect attractor set itself when compared with random networks, which illustrates that most genotype differences tend to influence the proportion of different phenotypes and only a few ones can create new phenotypes. DReSS can provide researchers with a new insight to study the relation between genotypes and phenotypes.

Yin Ziqiao, Guo Binghui, Ma Shuangge, Sun Yifan, Mi Zhilong, Zheng Zhiming

2020-Dec-14

dynamical analysis, genetic regulatory network, state space, structural perturbation

Public Health

EWASex: an efficient R-package to predict sex in epigenome-wide association studies.

In Bioinformatics (Oxford, England)

SUMMARY : Epigenome-Wide Association Study (EWAS) has become a powerful approach to identify epigenetic variations associated with diseases or health traits. Sex is an important variable to include in EWAS to ensure unbiased data processing and statistical analysis. We introduce the R-package EWASex, which allows for fast and highly accurate sex-estimation using DNA methylation data on a small set of CpG sites located on the X-chromosome under stable X-chromosome inactivation in females.

RESULTS : We demonstrate that EWASex outperforms the current state of the art tools by using different EWAS datasets. With EWASex, we offer an efficient way to predict and to verify sex that can be easily implemented in any EWAS using blood samples or even other tissue types. It comes with pre-trained weights to work without prior sex labels and without requiring access to RAW data, which is a necessity for all currently available methods.

AVAILABILITY AND IMPLEMENTATION : The EWASex R-package along with tutorials, documentation and source code are available at https://github.com/Silver-Hawk/EWASex.

SUPPLEMENTARY INFORMATION : Supplementary data are available at Bioinformatics online.

Lund Jesper Beltoft, Li Weilong, Mohammadnejad Afsaneh, Li Shuxia, Baumbach Jan, Tan Qihua

2020-Dec-11

General

Do we need different machine learning algorithms for QSAR modeling? A comprehensive assessment of 16 machine learning algorithms on 14 QSAR data sets.

In Briefings in bioinformatics

Although a wide variety of machine learning (ML) algorithms have been utilized to learn quantitative structure-activity relationships (QSARs), there is no agreed single best algorithm for QSAR learning. Therefore, a comprehensive understanding of the performance characteristics of popular ML algorithms used in QSAR learning is highly desirable. In this study, five linear algorithms [linear function Gaussian process regression (linear-GPR), linear function support vector machine (linear-SVM), partial least squares regression (PLSR), multiple linear regression (MLR) and principal component regression (PCR)], three analogizers [radial basis function support vector machine (rbf-SVM), K-nearest neighbor (KNN) and radial basis function Gaussian process regression (rbf-GPR)], six symbolists [extreme gradient boosting (XGBoost), Cubist, random forest (RF), multiple adaptive regression splines (MARS), gradient boosting machine (GBM), and classification and regression tree (CART)] and two connectionists [principal component analysis artificial neural network (pca-ANN) and deep neural network (DNN)] were employed to learn the regression-based QSAR models for 14 public data sets comprising nine physicochemical properties and five toxicity endpoints. The results show that rbf-SVM, rbf-GPR, XGBoost and DNN generally illustrate better performances than the other algorithms. The overall performances of different algorithms can be ranked from the best to the worst as follows: rbf-SVM > XGBoost > rbf-GPR > Cubist > GBM > DNN > RF > pca-ANN > MARS > linear-GPR ≈ KNN > linear-SVM ≈ PLSR > CART ≈ PCR ≈ MLR. In terms of prediction accuracy and computational efficiency, SVM and XGBoost are recommended to the regression learning for small data sets, and XGBoost is an excellent choice for large data sets. We then investigated the performances of the ensemble models by integrating the predictions of multiple ML algorithms. The results illustrate that the ensembles of two or three algorithms in different categories can indeed improve the predictions of the best individual ML algorithms.

Wu Zhenxing, Zhu Minfeng, Kang Yu, Leung Elaine Lai-Han, Lei Tailong, Shen Chao, Jiang Dejun, Wang Zhe, Cao Dongsheng, Hou Tingjun

2020-Dec-14

QSAR, XGBoost, ensemble learning, machine learning, support vector machine

General

ITP-Pred: an interpretable method for predicting, therapeutic peptides with fused features low-dimension representation.

In Briefings in bioinformatics

The peptide therapeutics market is providing new opportunities for the biotechnology and pharmaceutical industries. Therefore, identifying therapeutic peptides and exploring their properties are important. Although several studies have proposed different machine learning methods to predict peptides as being therapeutic peptides, most do not explain the decision factors of model in detail. In this work, an Interpretable Therapeutic Peptide Prediction (ITP-Pred) model based on efficient feature fusion was developed. First, we proposed three kinds of feature descriptors based on sequence and physicochemical property encoded, namely amino acid composition (AAC), group AAC and coding autocorrelation, and concatenated them to obtain the feature representation of therapeutic peptide. Then, we input it into the CNN-Bi-directional Long Short-Term Memory (BiLSTM) model to automatically learn recognition of therapeutic peptides. The cross-validation and independent verification experiments results indicated that ITP-Pred has a higher prediction performance on the benchmark dataset than other comparison methods. Finally, we analyzed the output of the model from two aspects: sequence order and physical and chemical properties, mining important features as guidance for the design of better models that can complement existing methods.

Cai Lijun, Wang Li, Fu Xiangzheng, Xia Chenxing, Zeng Xiangxiang, Zou Quan

2020-Dec-14

CNN-BiLSTM, feature fusion, interpretability analysis, therapeutic peptides prediction

Public Health

The Value of Artificial Intelligence in Laboratory Medicine.

In American journal of clinical pathology ; h5-index 39.0

OBJECTIVES : As laboratory medicine continues to undergo digitalization and automation, clinical laboratorians will likely be confronted with the challenges associated with artificial intelligence (AI). Understanding what AI is good for, how to evaluate it, what are its limitations, and how it can be implemented are not well understood. With a survey, we aimed to evaluate the thoughts of stakeholders in laboratory medicine on the value of AI in the diagnostics space and identify anticipated challenges and solutions to introducing AI.

METHODS : We conducted a web-based survey on the use of AI with participants from Roche's Strategic Advisory Network that included key stakeholders in laboratory medicine.

RESULTS : In total, 128 of 302 stakeholders responded to the survey. Most of the participants were medical practitioners (26%) or laboratory managers (22%). AI is currently used in the organizations of 15.6%, while 66.4% felt they might use it in the future. Most had an unsure attitude on what they would need to adopt AI in the diagnostics space. High investment costs, lack of proven clinical benefits, number of decision makers, and privacy concerns were identified as barriers to adoption. Education in the value of AI, streamlined implementation and integration into existing workflows, and research to prove clinical utility were identified as solutions needed to mainstream AI in laboratory medicine.

CONCLUSIONS : This survey demonstrates that specific knowledge of AI in the medical community is poor and that AI education is much needed. One strategy could be to implement new AI tools alongside existing tools.

Paranjape Ketan, Schinkel Michiel, Hammer Richard D, Schouten Bo, Nannan Panday R S, Elbers Paul W G, Kramer Mark H H, Nanayakkara Prabath

2020-Dec-14

Artificial intelligence, Diagnostics, Laboratory medicine, Medical education

General

Prediction of bio-sequence modifications and the associations with diseases.

In Briefings in functional genomics

Modifications of protein, RNA and DNA play an important role in many biological processes and are related to some diseases. Therefore, accurate identification and comprehensive understanding of protein, RNA and DNA modification sites can promote research on disease treatment and prevention. With the development of sequencing technology, the number of known sequences has continued to increase. In the past decade, many computational tools that can be used to predict protein, RNA and DNA modification sites have been developed. In this review, we comprehensively summarized the modification site predictors for three different biological sequences and the association with diseases. The relevant web server is accessible at http://lab.malab.cn/∼acy/PTM_data/ some sample data on protein, RNA and DNA modification can be downloaded from that website.

Ao Chunyan, Yu Liang, Zou Quan

2020-Dec-14

bio-sequence modifications, diseases, machine learning, prediction tool

General

Automated Measurements of Key Morphological Features of Human Embryos for IVF.

In Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention

A major challenge in clinical In-Vitro Fertilization (IVF) is selecting the highest quality embryo to transfer to the patient in the hopes of achieving a pregnancy. Time-lapse microscopy provides clinicians with a wealth of information for selecting embryos. However, the resulting movies of embryos are currently analyzed manually, which is time consuming and subjective. Here, we automate feature extraction of time-lapse microscopy of human embryos with a machine-learning pipeline of five convolutional neural networks (CNNs). Our pipeline consists of (1) semantic segmentation of the regions of the embryo, (2) regression predictions of fragment severity, (3) classification of the developmental stage, and object instance segmentation of (4) cells and (5) pronuclei. Our approach greatly speeds up the measurement of quantitative, biologically relevant features that may aid in embryo selection.

Leahy B D, Jang W-D, Yang H Y, Struyven R, Wei D, Sun Z, Lee K R, Royston C, Cam L, Kalma Y, Azem F, Ben-Yosef D, Pfister H, Needleman D

2020-Oct

Deep Learning, Human Embryos, In-Vitro Fertilization

Public Health

Aligning text mining and machine learning algorithms with best practices for study selection in systematic literature reviews.

In Systematic reviews

BACKGROUND : Despite existing research on text mining and machine learning for title and abstract screening, the role of machine learning within systematic literature reviews (SLRs) for health technology assessment (HTA) remains unclear given lack of extensive testing and of guidance from HTA agencies. We sought to address two knowledge gaps: to extend ML algorithms to provide a reason for exclusion-to align with current practices-and to determine optimal parameter settings for feature-set generation and ML algorithms.

METHODS : We used abstract and full-text selection data from five large SLRs (n = 3089 to 12,769 abstracts) across a variety of disease areas. Each SLR was split into training and test sets. We developed a multi-step algorithm to categorize each citation into the following categories: included; excluded for each PICOS criterion; or unclassified. We used a bag-of-words approach for feature-set generation and compared machine learning algorithms using support vector machines (SVMs), naïve Bayes (NB), and bagged classification and regression trees (CART) for classification. We also compared alternative training set strategies: using full data versus downsampling (i.e., reducing excludes to balance includes/excludes because machine learning algorithms perform better with balanced data), and using inclusion/exclusion decisions from abstract versus full-text screening. Performance comparisons were in terms of specificity, sensitivity, accuracy, and matching the reason for exclusion.

RESULTS : The best-fitting model (optimized sensitivity and specificity) was based on the SVM algorithm using training data based on full-text decisions, downsampling, and excluding words occurring fewer than five times. The sensitivity and specificity of this model ranged from 94 to 100%, and 54 to 89%, respectively, across the five SLRs. On average, 75% of excluded citations were excluded with a reason and 83% of these citations matched the reviewers' original reason for exclusion. Sensitivity significantly improved when both downsampling and abstract decisions were used.

CONCLUSIONS : ML algorithms can improve the efficiency of the SLR process and the proposed algorithms could reduce the workload of a second reviewer by identifying exclusions with a relevant PICOS reason, thus aligning with HTA guidance. Downsampling can be used to improve study selection, and improvements using full-text exclusions have implications for a learn-as-you-go approach.

Popoff E, Besada M, Jansen J P, Cope S, Kanters S

2020-Dec-13

Classification, Downsampling, Machine learning, Methods, Reasons for exclusion, Study selection, Systematic literature reviews, Text mining, Updates

Pathology

Machine Learning-Based Presymptomatic Detection of Rice Sheath Blight Using Spectral Profiles.

In Plant phenomics (Washington, D.C.)

Early detection of plant diseases, prior to symptom development, can allow for targeted and more proactive disease management. The objective of this study was to evaluate the use of near-infrared (NIR) spectroscopy combined with machine learning for early detection of rice sheath blight (ShB), caused by the fungus Rhizoctonia solani. We collected NIR spectra from leaves of ShB-susceptible rice (Oryza sativa L.) cultivar, Lemont, growing in a growth chamber one day following inoculation with R. solani, and prior to the development of any disease symptoms. Support vector machine (SVM) and random forest, two machine learning algorithms, were used to build and evaluate the accuracy of supervised classification-based disease predictive models. Sparse partial least squares discriminant analysis was used to confirm the results. The most accurate model comparing mock-inoculated and inoculated plants was SVM-based and had an overall testing accuracy of 86.1% (N = 72), while when control, mock-inoculated, and inoculated plants were compared the most accurate SVM model had an overall testing accuracy of 73.3% (N = 105). These results suggest that machine learning models could be developed into tools to diagnose infected but asymptomatic plants based on spectral profiles at the early stages of disease development. While testing and validation in field trials are still needed, this technique holds promise for application in the field for disease diagnosis and management.

Conrad Anna O, Li Wei, Lee Da-Young, Wang Guo-Liang, Rodriguez-Saona Luis, Bonello Pierluigi

2020

General

TasselGAN: An Application of the Generative Adversarial Model for Creating Field-Based Maize Tassel Data.

In Plant phenomics (Washington, D.C.)

Machine learning-based plant phenotyping systems have enabled high-throughput, non-destructive measurements of plant traits. Tasks such as object detection, segmentation, and localization of plant traits in images taken in field conditions need the machine learning models to be developed on training datasets that contain plant traits amidst varying backgrounds and environmental conditions. However, the datasets available for phenotyping are typically limited in variety and mostly consist of lab-based images in controlled conditions. Here, we present a new method called TasselGAN, using a variant of a deep convolutional generative adversarial network, to synthetically generate images of maize tassels against sky backgrounds. Both foreground tassel images and background sky images are generated separately and merged together to form artificial field-based maize tassel data to aid the training of machine learning models, where there is a paucity of field-based data. The effectiveness of the proposed method is demonstrated using quantitative and perceptual qualitative experiments.

Shete Snehal, Srinivasan Srikant, Gonsalves Timothy A

2020

General

Global Wheat Head Detection (GWHD) Dataset: A Large and Diverse Dataset of High-Resolution RGB-Labelled Images to Develop and Benchmark Wheat Head Detection Methods.

In Plant phenomics (Washington, D.C.)

The detection of wheat heads in plant images is an important task for estimating pertinent wheat traits including head population density and head characteristics such as health, size, maturity stage, and the presence of awns. Several studies have developed methods for wheat head detection from high-resolution RGB imagery based on machine learning algorithms. However, these methods have generally been calibrated and validated on limited datasets. High variability in observational conditions, genotypic differences, development stages, and head orientation makes wheat head detection a challenge for computer vision. Further, possible blurring due to motion or wind and overlap between heads for dense populations make this task even more complex. Through a joint international collaborative effort, we have built a large, diverse, and well-labelled dataset of wheat images, called the Global Wheat Head Detection (GWHD) dataset. It contains 4700 high-resolution RGB images and 190000 labelled wheat heads collected from several countries around the world at different growth stages with a wide range of genotypes. Guidelines for image acquisition, associating minimum metadata to respect FAIR principles, and consistent head labelling methods are proposed when developing new head detection datasets. The GWHD dataset is publicly available at http://www.global-wheat.com/and aimed at developing and benchmarking methods for wheat head detection.

David Etienne, Madec Simon, Sadeghi-Tehran Pouria, Aasen Helge, Zheng Bangyou, Liu Shouyang, Kirchgessner Norbert, Ishikawa Goro, Nagasawa Koichi, Badhon Minhajul A, Pozniak Curtis, de Solan Benoit, Hund Andreas, Chapman Scott C, Baret Frédéric, Stavness Ian, Guo Wei

2020

General

A Deep Learning-Based Phenotypic Analysis of Rice Root Distribution from Field Images.

In Plant phenomics (Washington, D.C.)

Root distribution in the soil determines plants' nutrient and water uptake capacity. Therefore, root distribution is one of the most important factors in crop production. The trench profile method is used to observe the root distribution underground by making a rectangular hole close to the crop, providing informative images of the root distribution compared to other root phenotyping methods. However, much effort is required to segment the root area for quantification. In this study, we present a promising approach employing a convolutional neural network for root segmentation in trench profile images. We defined two parameters, Depth50 and Width50, representing the vertical and horizontal centroid of root distribution, respectively. Quantified parameters for root distribution in rice (Oryza sativa L.) predicted by the trained model were highly correlated with parameters calculated by manual tracing. These results indicated that this approach is useful for rapid quantification of the root distribution from the trench profile images. Using the trained model, we quantified the root distribution parameters among 60 rice accessions, revealing the phenotypic diversity of root distributions. We conclude that employing the trench profile method and a convolutional neural network is reliable for root phenotyping and it will furthermore facilitate the study of crop roots in the field.

Teramoto S, Uga Y

2020

General

Computing on Phenotypic Descriptions for Candidate Gene Discovery and Crop Improvement.

In Plant phenomics (Washington, D.C.)

Many newly observed phenotypes are first described, then experimentally manipulated. These language-based descriptions appear in both the literature and in community datastores. To standardize phenotypic descriptions and enable simple data aggregation and analysis, controlled vocabularies and specific data architectures have been developed. Such simplified descriptions have several advantages over natural language: they can be rigorously defined for a particular context or problem, they can be assigned and interpreted programmatically, and they can be organized in a way that allows for semantic reasoning (inference of implicit facts). Because researchers generally report phenotypes in the literature using natural language, curators have been translating phenotypic descriptions into controlled vocabularies for decades to make the information computable. Unfortunately, this methodology is highly dependent on human curation, which does not scale to the scope of all publications available across all of plant biology. Simultaneously, researchers in other domains have been working to enable computation on natural language. This has resulted in new, automated methods for computing on language that are now available, with early analyses showing great promise. Natural language processing (NLP) coupled with machine learning (ML) allows for the use of unstructured language for direct analysis of phenotypic descriptions. Indeed, we have found that these automated methods can be used to create data structures that perform as well or better than those generated by human curators on tasks such as predicting gene function and biochemical pathway membership. Here, we describe current and ongoing efforts to provide tools for the plant phenomics community to explore novel predictions that can be generated using these techniques. We also describe how these methods could be used along with mobile speech-to-text tools to collect and analyze in-field spoken phenotypic descriptions for association genetics and breeding applications.

Braun Ian R, Yanarella Colleen F, Lawrence-Dill Carolyn J

2020

General

Soybean Root System Architecture Trait Study through Genotypic, Phenotypic, and Shape-Based Clusters.

In Plant phenomics (Washington, D.C.)

We report a root system architecture (RSA) traits examination of a larger scale soybean accession set to study trait genetic diversity. Suffering from the limitation of scale, scope, and susceptibility to measurement variation, RSA traits are tedious to phenotype. Combining 35,448 SNPs with an imaging phenotyping platform, 292 accessions (replications = 14) were studied for RSA traits to decipher the genetic diversity. Based on literature search for root shape and morphology parameters, we used an ideotype-based approach to develop informative root (iRoot) categories using root traits. The RSA traits displayed genetic variability for root shape, length, number, mass, and angle. Soybean accessions clustered into eight genotype- and phenotype-based clusters and displayed similarity. Genotype-based clusters correlated with geographical origins. SNP profiles indicated that much of US origin genotypes lack genetic diversity for RSA traits, while diverse accession could infuse useful genetic variation for these traits. Shape-based clusters were created by integrating convolution neural net and Fourier transformation methods, enabling trait cataloging for breeding and research applications. The combination of genetic and phenotypic analyses in conjunction with machine learning and mathematical models provides opportunities for targeted root trait breeding efforts to maximize the beneficial genetic diversity for future genetic gains.

Falk Kevin G, Jubery Talukder Zaki, O’Rourke Jamie A, Singh Arti, Sarkar Soumik, Ganapathysubramanian Baskar, Singh Asheesh K

2020

General

High-Throughput Rice Density Estimation from Transplantation to Tillering Stages Using Deep Networks.

In Plant phenomics (Washington, D.C.)

Rice density is closely related to yield estimation, growth diagnosis, cultivated area statistics, and management and damage evaluation. Currently, rice density estimation heavily relies on manual sampling and counting, which is inefficient and inaccurate. With the prevalence of digital imagery, computer vision (CV) technology emerges as a promising alternative to automate this task. However, challenges of an in-field environment, such as illumination, scale, and appearance variations, render gaps for deploying CV methods. To fill these gaps towards accurate rice density estimation, we propose a deep learning-based approach called the Scale-Fusion Counting Classification Network (SFC²Net) that integrates several state-of-the-art computer vision ideas. In particular, SFC²Net addresses appearance and illumination changes by employing a multicolumn pretrained network and multilayer feature fusion to enhance feature representation. To ameliorate sample imbalance engendered by scale, SFC²Net follows a recent blockwise classification idea. We validate SFC²Net on a new rice plant counting (RPC) dataset collected from two field sites in China from 2010 to 2013. Experimental results show that SFC²Net achieves highly accurate counting performance on the RPC dataset with a mean absolute error (MAE) of 25.51, a root mean square error (MSE) of 38.06, a relative MAE of 3.82%, and a R² of 0.98, which exhibits a relative improvement of 48.2% w.r.t. MAE over the conventional counting approach CSRNet. Further, SFC²Net provides high-throughput processing capability, with 16.7 frames per second on 1024 × 1024 images. Our results suggest that manual rice counting can be safely replaced by SFC²Net at early growth stages. Code and models are available online at https://git.io/sfc2net.

Liu Liang, Lu Hao, Li Yanan, Cao Zhiguo

2020

General

A hydrogel-based in vitro assay for the fast prediction of antibiotic accumulation in Gram-negative bacteria.

In Materials today. Bio

The pipeline of antibiotics has been for decades on an alarmingly low level. Considering the steadily emerging antibiotic resistance, novel tools are needed for early and easy identification of effective anti-infective compounds. In Gram-negative bacteria, the uptake of anti-infectives is especially limited. We here present a surprisingly simple in vitro model of the Gram-negative bacterial envelope, based on 20% (w/v) potato starch gel, printed on polycarbonate 96-well filter membranes. Rapid permeability measurements across this polysaccharide hydrogel allowed to correctly predict either high or low accumulation for all 16 tested anti-infectives in living Escherichia coli. Freeze-fracture TEM supports that the macromolecular network structure of the starch hydrogel may represent a useful surrogate of the Gram-negative bacterial envelope. A random forest analysis of in vitro data revealed molecular mass, minimum projection area, and rigidity as the most critical physicochemical parameters for hydrogel permeability, in agreement with reported structural features needed for uptake into Gram-negative bacteria. Correlating our dataset of 27 antibiotics from different structural classes to reported MIC values of nine clinically relevant pathogens allowed to distinguish active from nonactive compounds based on their low in vitro permeability specifically for Gram-negatives. The model may help to identify poorly permeable antimicrobial candidates before testing them on living bacteria.

Richter Robert, Kamal Mohamed A M, García-Rivera Mariel A, Kaspar Jerome, Junk Maximilian, Elgaher Walid A M, Srikakulam Sanjay Kumar, Gress Alexander, Beckmann Anja, Grißmer Alexander, Meier Carola, Vielhaber Michael, Kalinina Olga, Hirsch Anna K H, Hartmann Rolf W, Brönstrup Mark, Schneider-Daum Nicole, Lehr Claus-Michael

2020-Sep

3D-printing, Antibiotic screening, Machine learning, Starch hydrogel, Structure–permeability relationships

Radiology

Advances in imaging for lung emphysema.

In Annals of translational medicine

Lung emphysema represents a major public health burden and still accounts for five percent of all deaths worldwide. Hence, it is essential to further understand this disease in order to develop effective diagnostic and therapeutic strategies. Lung emphysema is an irreversible enlargement of the airways distal to the terminal bronchi (i.e., the alveoli) due to the destruction of the alveolar walls. The two most important causes of emphysema are (I) smoking and (II) α1-antitrypsin-deficiency. In the former lung emphysema is predominant in the upper lung parts, the latter is characterized by a predominance in the basal areas of the lungs. Since quantification and evaluation of the distribution of lung emphysema is crucial in treatment planning, imaging plays a central role. Imaging modalities in lung emphysema are manifold: computed tomography (CT) imaging is nowadays the gold standard. However, emerging imaging techniques like dynamic or functional magnetic resonance imaging (MRI), scintigraphy and lately also the implementation of radiomics and artificial intelligence are more and more diffused in the evaluation, diagnosis and quantification of lung emphysema. The aim of this review is to shortly present the different subtypes of lung emphysema, to give an overview on prediction and risk assessment in emphysematous disease and to discuss not only the traditional, but also the new imaging techniques for diagnosis, quantification and evaluation of lung emphysema.

Martini Katharina, Frauenfelder Thomas

2020-Nov

Emphysema, chronic obstructive pulmonary disease (COPD), imaging

Radiology

Chest CT imaging features and severity scores as biomarkers for prognostic prediction in patients with COVID-19.

In Annals of translational medicine

Background : Coronavirus disease 2019 (COVID-19) has become a pandemic. Few studies have explored the role of chest computed tomography (CT) features and severity scores for prognostic prediction. In this study, we aimed to investigate the role of chest CT severity score and imaging features in the prediction of the prognosis of COVID-19 patients.

Methods : A total of 134 patients (62 recovered and 72 deceased patients) with confirmed COVID-19 were enrolled. The clinical, laboratory, and chest CT (316 scans) data were retrospectively reviewed. Demographics, symptoms, comorbidities, and temporal changes of laboratory results, CT features, and severity scores were compared between recovered and deceased groups using the Mann-Whitney U test and logistic regression to identify the risk factors for poor prognosis.

Results : Median age was 48 and 58 years for recovered and deceased patients, respectively. More patients had at least one comorbidity in the deceased group than the recovered group (60% vs. 29%). Leukocytes, neutrophil, high-sensitivity C-reactive protein (hsCRP), prothrombin, D-dimer, serum ferritin, interleukin (IL)-2, and IL-6 were significantly elevated in the deceased group than the recovered group at different stages. The total CT score at the peak stage was significantly greater in the deceased group than the recovered group (20 vs. 11 points). The optimal cutoff value of the total CT scores was 16.5 points, achieving 69.4% sensitivity and 82.2% specificity for the prognostic prediction. The crazy-paving pattern and consolidation were more common in the deceased patients than those in the recovered patients. Linear opacities significantly increased with the disease course in the recovered patients. Sex, age, neutrophil, IL-2, IL-6, and total CT scores were independent risk factors for the prognosis with odds ratios of 3.8 to 8.7.

Conclusions : Sex (male), older age (>60 years), elevated neutrophil, IL-2, IL-6 level, and total CT scores (≥16) were independent risk factors for poor prognosis in patients with COVID-19. Temporal changes of chest CT features and severity scores could be valuable for early identification of severe cases and eventually reducing the mortality rate of COVID-19.

Zhou Shuchang, Chen Chengyang, Hu Yiqi, Lv Wenzhi, Ai Tao, Xia Liming

2020-Nov

Coronavirus disease, computed tomography, prognosis, risk factor, severity score

Radiology

Reference ranges ("normal values") for cardiovascular magnetic resonance (CMR) in adults and children: 2020 update.

In Journal of cardiovascular magnetic resonance : official journal of the Society for Cardiovascular Magnetic Resonance

Cardiovascular magnetic resonance (CMR) enables assessment and quantification of morphological and functional parameters of the heart, including chamber size and function, diameters of the aorta and pulmonary arteries, flow and myocardial relaxation times. Knowledge of reference ranges ("normal values") for quantitative CMR is crucial to interpretation of results and to distinguish normal from disease. Compared to the previous version of this review published in 2015, we present updated and expanded reference values for morphological and functional CMR parameters of the cardiovascular system based on the peer-reviewed literature and current CMR techniques. Further, databases and references for deep learning methods are included.

Kawel-Boehm Nadine, Hetzel Scott J, Ambale-Venkatesh Bharath, Captur Gabriella, Francois Christopher J, Jerosch-Herold Michael, Salerno Michael, Teague Shawn D, Valsangiacomo-Buechel Emanuela, van der Geest Rob J, Bluemke David A

2020-Dec-14

Cardiac magnetic resonance, Normal values, Reference range

Public Health

Evaluating the impact of mobility on COVID-19 pandemic with machine learning hybrid predictions.

In The Science of the total environment

COVID-19 pandemic had expanded to the US since early 2020 and has caused nationwide economic loss and public health crisis. Until now, although the US has the most confirmed cases in the world and are still experiencing an increasing pandemic, several states insisted to re-open business activities and colleges while announced strict control measures. To provide a quantitative reference for official strategies, predicting the near future trend based on finer spatial resolution data and presumed scenarios are urgently needed. In this study, the first attempted COVID-19 case predicting model based on county-level demographic, environmental, and mobility data was constructed with multiple machine learning techniques and a hybrid framework. Different scenarios were also applied to selected metropolitan counties including New York City, Cook County in Illinois, Los Angeles County in California, and Miami-Dade County in Florida to assess the impact from lockdown, Phase I, and Phase III re-opening. Our results showed that, for selected counties, the mobility decreased substantially after the lockdown but kept increasing with an apparent weekly pattern, and the weekly pattern of mobility and infections implied high infections during the weekend. Meanwhile, our model was successfully built up, and the scenario assessment results indicated that, compared with Phase I re-opening, a 1-week and a 2-week lockdown could reduce 4%-29% and 15%-55% infections, respectively, in the future week, while 2-week Phase III re-opening could increase 16%-80% infections. We concluded that the mandatory orders in metropolitan counties such lockdown should last longer than one week, the effect could be observed. The impact of lockdown or re-opening was also county-dependent and varied with the local pandemic. In future works, we expect to involve a longer period of data, consider more county-dependent factors, and employ more sophisticated techniques to decrease the modeling uncertainty and apply it to counties nationally and other countries.

Kuo Cheng-Pin, Fu Joshua S

2020-Nov-28

County-level, Forecasting, Lockdown, Pandemic, Re-opening

Surgery

Screening and identification of a CD44v6 specific peptide using improved phage display for gastric cancer targeting.

In Annals of translational medicine

Background : Peptide probes can be applied for biomarker targeting to improve the diagnostic accuracy. Cluster of differentiation 44 (CD44) is up-regulated in gastric cancer (GC). Among all the variants of CD44, CD44v6 is reported the most promising biomarker for GC. The purpose of this study was generating and identification a peptide ligand specific to CD44v6.

Methods : A 12-mer phage peptide library was screened on CD44v overexpressed HEK-293 cells with an improved subtractive method. Five candidate sequences emerged. Candidate phages were selected using enzyme-linked immunosorbent assay and competitive inhibition assays. Then the sequence (designated ELT) was chosen for further study. Its binding affinity and specificity were verified on recombinant protein, GC cells, GC tissues and xenograft models based on BALB/c-nu/nu mice using dissociation constant calculation, immunofluorescence, immunohistochemistry and in vivo imaging separately.

Results : The dissociation constant of ELT with recombinant protein was 611.2 nM. ELT stained CD44v overexpressed HEK-293 but not the cell expressing wild-type CD44s. On GC cell lines, ELT co-stained with anti-CD44v6 antibody. ELT binding on tumor tissues significantly increased compared with that of paracancer tissues, also showed a linear positive correlation with CD44v6 expression. ELT specifically accumulated in tumor and eliminated in short time in vivo.

Conclusions : ELT can target GC in vitro and in vivo via CD44v6, indicating its potential to serve as a probe for GC targeting diagnosis and therapy.

Zhang Dan, Huang Jin, Li Weiming, Zhang Zhiyong, Zhu Meng, Feng Yun, Zhao Yan, Li Yarui, Lu Shaoying, He Shuixiang

2020-Nov

CD44v6, Peptide probe, gastric cancer (GC), in vivo imaging, phage display

General

Development of a scoring tool for predicting prolonged length of hospital stay in peritoneal dialysis patients through data mining.

In Annals of translational medicine

Background : The hospital admission rate is high in patients treated with peritoneal dialysis (PD), and the length of stay (LOS) in the hospital is a key indicator of medical resource allocation. This study aimed to develop a scoring tool for predicting prolonged LOS (pLOS) in PD patients by combining machine learning and traditional logistic regression (LR).

Methods : This study was based on patient data collected using the Hospital Quality Monitoring System (HQMS) in China. Three machine learning methods, classification and regression tree (CART), random forest (RF), and gradient boosting decision tree (GBDT), were used to develop models to predict pLOS, which is longer than the average LOS, in PD patients. The model with the best prediction performance was used to identify predictive factors contributing to the outcome. A multivariate LR model based on the identified predictors was then built to derive the score assigned to each predictor. Finally, a scoring tool was developed, and it was tested by stratifying PD patients into different pLOS risk groups.

Results : A total of 22,859 PD patients were included in our study, with 25.2% having pLOS. Among the three machine learning models, the RF model achieved the best prediction performance and thus was used to identify the 10 most predictive variables for building the scoring system. The multivariate LR model based on the identified predictors showed good discrimination power with an AUROC of 0.721 in the test dataset, and its coefficients were used as a basis for scoring tool development. On the basis of the developed scoring tool, PD patients were divided into three groups: low risk (≤5), median risk [5-10], and high risk (>10). The observed pLOS proportions in the low-risk, median-risk, and high-risk groups in the test dataset were 11.4%, 29.5%, and 54.7%, respectively.

Conclusions : This study developed a scoring tool to predict pLOS in PD patients. The scoring tool can effectively discriminate patients with different pLOS risks and be easily implemented in clinical practice. The pLOS scoring tool has a great potential to help physicians allocate medical resources optimally and achieve improved clinical outcomes.

Wu Jingyi, Kong Guilan, Lin Yu, Chu Hong, Yang Chao, Shi Ying, Wang Haibo, Zhang Luxia

2020-Nov

Length of stay (LOS), logistic regression (LR), machine learning, peritoneal dialysis (PD), scoring methods

General

The role of artificial intelligence in identifying asthma in pediatric inpatient setting.

In Annals of translational medicine

Background : The incidence of asthma in Chinese children has rapidly increased as a result of inadequate management. This is mainly due to the failure of many primary-level pediatricians to distinguish asthma from common respiratory diseases, such as bronchitis and pneumonia. Such misdiagnoses often lead to the abuse of antibiotics and systemic glucocorticoids. Additionally, if asthma is not diagnosed early, chronic airway inflammation results in lesions that not only hamper children's athletic abilities, but serve as the primary cause for adult chronic airway diseases, such as chronic obstructive pulmonary disease (COPD).

Methods : A number of machine learning-based models including CatBoost, Logistic Regression, Naïve Bayes, and Support Vector Machines (SVM) have been developed to identify asthma via utilizing retrospective electronic medical records (EMRs) of patients. These models were evaluated independently using EMRs from both the Pulmonology Department and other departments of the Children's Hospital, Zhejiang University School of Medicine, China.

Results : Two independent test sets were applied for performance evaluation. TestSet-1 consisted of 325 positive asthma cases and 428 negative cases from the Pulmonology Department. TestSet-2 was composed of 2,123 cases from non-pulmonology departments, and included 337 positive and 1,786 negative cases. Experimental results showed that the CatBoost model outperformed other models on both test sets with an accuracy of 84.7% and an area under the curve (AUC) of 90.9% on TestSet-1, and an accuracy of 96.7% and an AUC of 98.1% on TestSet-2.

Conclusions : The artificial intelligence (AI) model could rapidly and accurately identify asthma in general medical wards of children, and may aid primary pediatricians in the correct diagnoses of asthma. It possesses great clinical value and practical significance in improving the control rate of asthma in children, optimizing medical resources, and limiting the abuse of antibiotics and systemic glucocorticoids.

Yu Gang, Li Zheming, Li Shuxian, Liu Jingling, Sun Moyuan, Liu Xiaoqing, Sun Fenglei, Zheng Jie, Li Yiming, Yu Yizhou, Shu Qiang, Wang Yingshuo

2020-Nov

Pediatric, artificial intelligence (AI), asthma, diagnostic assistant, machine learning

Pathology

Pathological Cluster Identification by Unsupervised Analysis in 3,822 UK Biobank Cardiac MRIs.

In Frontiers in cardiovascular medicine

We perform unsupervised analysis of image-derived shape and motion features extracted from 3,822 cardiac Magnetic resonance imaging (MRIs) of the UK Biobank. First, with a feature extraction method previously published based on deep learning models, we extract from each case 9 feature values characterizing both the cardiac shape and motion. Second, a feature selection is performed to remove highly correlated feature pairs. Third, clustering is carried out using a Gaussian mixture model on the selected features. After analysis, we identify 2 small clusters that probably correspond to 2 pathological categories. Further confirmation using a trained classification model and dimensionality reduction tools is carried out to support this finding. Moreover, we examine the differences between the other large clusters and compare our measures with the ground truth.

Zheng Qiao, Delingette Hervé, Fung Kenneth, Petersen Steffen E, Ayache Nicholas

2020

UK Biobank, cardiac pathology, cine MRI, cluster analysis, feature extraction

General

Leveraging Data Science for a Personalized Haemodialysis.

In Kidney diseases (Basel, Switzerland)

Background : The 2019 Science for Dialysis Meeting at Bellvitge University Hospital was devoted to the challenges and opportunities posed by the use of data science to facilitate precision and personalized medicine in nephrology, and to describe new approaches and technologies. The meeting included separate sections for issues in data collection and data analysis. As part of data collection, we presented the institutional ARGOS e-health project, which provides a common model for the standardization of clinical practice. We also pay specific attention to the way in which randomized controlled trials offer data that may be critical to decision-making in the real world. The opportunities of open source software (OSS) for data science in clinical practice were also discussed.

Summary : Precision medicine aims to provide the right treatment for the right patients at the right time and is deeply connected to data science. Dialysis patients are highly dependent on technology to live, and their treatment generates a huge volume of data that has to be analysed. Data science has emerged as a tool to provide an integrated approach to data collection, storage, cleaning, processing, analysis, and interpretation from potentially large volumes of information. This is meant to be a perspective article about data science based on the experience of the experts invited to the Science for Dialysis Meeting and provides an up-to-date perspective of the potential of data science in kidney disease and dialysis.

Key messages : Healthcare is quickly becoming data-dependent, and data science is a discipline that holds the promise of contributing to the development of personalized medicine, although nephrology still lags behind in this process. The key idea is to ensure that data will guide medical decisions based on individual patient characteristics rather than on averages over a whole population usually based on randomized controlled trials that excluded kidney disease patients. Furthermore, there is increasing interest in obtaining data about the effectiveness of available treatments in current patient care based on pragmatic clinical trials. The use of data science in this context is becoming increasingly feasible in part thanks to the swift developments in OSS.

Hueso Miguel, de Haro Lluís, Calabia Jordi, Dal-Ré Rafael, Tebé Cristian, Gibert Karina, Cruzado Josep M, Vellido Alfredo

2020-Nov

Artificial intelligence, Data science, Haemodialysis, Machine learning, Personalized medicine, Pragmatic clinical trials

Radiology

The Modified Heidelberg and the AI Appendicitis Score Are Superior to Current Scores in Predicting Appendicitis in Children: A Two-Center Cohort Study.

In Frontiers in pediatrics

Background: Acute appendicitis represents the most frequent reason for abdominal surgery in children. Since diagnosis can be challenging various scoring systems have been published. The aim of this study was to evaluate and validate (and improve) different appendicitis scores in a very large cohort of children with abdominal pain. Methods: Retrospective analysis of all children that have been hospitalized due to suspected appendicitis at the Pediatric Surgery Department of the Altonaer Children's Hospital and University Medical Center Hamburg-Eppendorf from 01/2018 until 11/2019. Four different appendicitis scores (Heidelberg Appendicitis Score, Alvarado Score, Pediatric Appendicitis Score and Tzanakis Score) were applied to all data sets. Furthermore, the best score was improved and artificial intelligence (AI) was applied and compare the current scores. Results: In 23 months, 463 patients were included in the study. Of those 348 (75.2%) were operated for suspected appendicitis and in 336 (96.6%) patients the diagnosis was confirmed histopathologically. The best predictors of appendicitis (simple and perforated) were rebound tenderness, cough/hopping tenderness, ultrasound, and laboratory results. After modifying the HAS, it provided excellent results for simple (PPV 95.0%, NPV 70.0%) and very good for perforated appendicitis (PPV 34.4%, NPV 93.8%), outperforming all other appendicitis score. Discussion: The modified HAS and the AI score show excellent predictive capabilities and may be used to identify most cases of appendicitis and more important to rule out perforated appendicitis. The new scores outperform all other scores and are simple to apply. The modified HAS comprises five features that can all be assessed in the emergency department as opposed to current scores that are relatively complex to utilize in a clinical setting as they include of up to eight features with various weighting factors. In conclusion, the modified HAS and the AI score may be used to identify children with appendicitis, yet prospective studies to validate our findings in a large mutli-center cohorts are needed.

Stiel Carolin, Elrod Julia, Klinke Michaela, Herrmann Jochen, Junge Carl-Martin, Ghadban Tarik, Reinshagen Konrad, Boettcher Michael

2020

appendicitis, children, diagnosis, predicition, scores

Public Health

Artificial intelligence in dentistry: Harnessing big data to predict oral cancer survival.

In World journal of clinical oncology

BACKGROUND : Oral cancer is the sixth most prevalent cancer worldwide. Public knowledge in oral cancer risk factors and survival is limited.

AIM : To come up with machine learning (ML) algorithms to predict the length of survival for individuals diagnosed with oral cancer, and to explore the most important factors that were responsible for shortening or lengthening oral cancer survival.

METHODS : We used the Surveillance, Epidemiology, and End Results database from the years 1975 to 2016 that consisted of a total of 257880 cases and 94 variables. Four ML techniques in the area of artificial intelligence were applied for model training and validation. Model accuracy was evaluated using mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE), R² and adjusted R².

RESULTS : The most important factors predictive of oral cancer survival time were age at diagnosis, primary cancer site, tumor size and year of diagnosis. Year of diagnosis referred to the year when the tumor was first diagnosed, implying that individuals with tumors that were diagnosed in the modern era tend to have longer survival than those diagnosed in the past. The extreme gradient boosting ML algorithms showed the best performance, with the MAE equaled to 13.55, MSE 486.55 and RMSE 22.06.

CONCLUSION : Using artificial intelligence, we developed a tool that can be used for oral cancer survival prediction and for medical-decision making. The finding relating to the year of diagnosis represented an important new discovery in the literature. The results of this study have implications for cancer prevention and education for the public.

Hung Man, Park Jungweon, Hon Eric S, Bounsanga Jerry, Moazzami Sara, Ruiz-Negrón Bianca, Wang Dawei

2020-Nov-24

Artificial intelligence, Dental medicine, Machine learning, Oral cancer survival, Public health, Quality of life, Surveillance, Epidemiology, and End Results

General

STAN: SMALL TUMOR-AWARE NETWORK FOR BREAST ULTRASOUND IMAGE SEGMENTATION.

In Proceedings. IEEE International Symposium on Biomedical Imaging

Breast tumor segmentation provides accurate tumor boundary, and serves as a key step toward further cancer quantification. Although deep learning-based approaches have been proposed and achieved promising results, existing approaches have difficulty in detecting small breast tumors. The capacity to detecting small tumors is particularly important in finding early stage cancers using computer-aided diagnosis (CAD) systems. In this paper, we propose a novel deep learning architecture called Small Tumor-Aware Network (STAN), to improve the performance of segmenting tumors with different size. The new architecture integrates both rich context information and high-resolution image features. We validate the proposed approach using seven quantitative metrics on two public breast ultrasound datasets. The proposed approach outperformed the state-of-the-art approaches in segmenting small breast tumors.

Shareef Bryar, Xian Min, Vakanski Aleksandar

2020-Apr

STAN, breast ultrasound, deep learning, multi-scale features, small tumor segmentation

General

BENDING LOSS REGULARIZED NETWORK FOR NUCLEI SEGMENTATION IN HISTOPATHOLOGY IMAGES.

In Proceedings. IEEE International Symposium on Biomedical Imaging

Separating overlapped nuclei is a major challenge in histopathology image analysis. Recently published approaches have achieved promising overall performance on public datasets; however, their performance in segmenting overlapped nuclei are limited. To address the issue, we propose the bending loss regularized network for nuclei segmentation. The proposed bending loss defines high penalties to contour points with large curvatures, and applies small penalties to contour points with small curvature. Minimizing the bending loss can avoid generating contours that encompass multiple nuclei. The proposed approach is validated on the MoNuSeg dataset using five quantitative metrics. It outperforms six state-of-the-art approaches on the following metrics: Aggregate Jaccard Index, Dice, Recognition Quality, and Panoptic Quality.

Wang Haotian, Xian Min, Vakanski Aleksandar

2020-Apr

Nuclei segmentation, bending loss, histopathology images, multitask deep learning

Oncology

Why we do what we do. A brief analysis of cancer therapies.

In EXCLI journal

The goal of all medical activity is to preserve health in fit people, and to restore the sick into a state of complete physical, mental and social wellbeing. In an effort to determine whether we are achieving this last goal in oncology, herein we review the biological and clinical framework that has led to the foundations of the current anticancer treatment paradigm. Currently, cancer therapy is still based on the ancient axiom that states that the complete eradication of the tumor burden is the only way to achieve a cure. This strategy has led to a substantial improvement in survival rates as cancer mortality rates have dropped in an unprecedented way. Despite this progress, more than 9 million people still die from cancer every year, indicating that the current treatment strategy is not leading to a cancer cure, but to a cancer remission, that is "the temporary absence of manifestations of a particular disease"; after months or years of remission, in most patients, cancer will inevitably recur. Our critical analysis indicates that it is time to discuss about the new key challenges and future directions in clinical oncology. We need to generate novel treatment strategies more suited to the current clinical reality.

Galmarini Carlos M

2020

antiangiogenic agents, artificial intelligence, immunotherapy, molecular targeted therapies, neoplasm

Oncology

Deep learning for sensitive detection of Helicobacter Pylori in gastric biopsies.

In BMC gastroenterology

BACKGROUND : Helicobacter pylori, a 2 × 1 μm spiral-shaped bacterium, is the most common risk factor for gastric cancer worldwide. Clinically, patients presenting with symptoms of gastritis, routinely undergo gastric biopsies. The following histo-morphological evaluation dictates therapeutic decisions, where antibiotics are used for H. pylori eradication. There is a strong rational to accelerate the detection process of H. pylori on histological specimens, using novel technologies, such as deep learning.

METHODS : We designed a deep-learning-based decision support algorithm that can be applied on regular whole slide images of gastric biopsies. In detail, we can detect H. pylori both on Giemsa- and regular H&E stained whole slide images.

RESULTS : With the help of our decision support algorithm, we show an increased sensitivity in a subset of 87 cases that underwent additional PCR- and immunohistochemical testing to define a sensitive ground truth of HP presence. For Giemsa stained sections, the decision support algorithm achieved a sensitivity of 100% compared to 68.4% (microscopic diagnosis), with a tolerable specificity of 66.2% for the decision support algorithm compared to 92.6 (microscopic diagnosis).

CONCLUSION : Together, we provide the first evidence of a decision support algorithm proving as a sensitive screening option for H. pylori that can potentially aid pathologists to accurately diagnose H. pylori presence on gastric biopsies.

Klein Sebastian, Gildenblat Jacob, Ihle Michaele Angelika, Merkelbach-Bruse Sabine, Noh Ka-Won, Peifer Martin, Quaas Alexander, Büttner Reinhard

2020-Dec-11

Artificial intelligence, Convolutional neural networks, Deep learning, Gastric cancer prevention, Helicobacter pylori, Screening

General

Corticosteroid therapy is associated with improved outcome in critically ill COVID-19 patients with hyperinflammatory phenotype.

In Chest ; h5-index 81.0

BACKGROUND : Corticosteroid therapy is commonly used in patients with coronavirus disease 2019 (COVID-19), while its impact on outcomes and which patients could benefit from corticosteroid therapy are uncertain.

RESEARCH QUESTION : Whether clinical phenotypes of COVID-19 were associated with differential response to corticosteroid therapy.

STUDY DESIGN AND METHODS : Critically ill patients with COVID-19 from Tongji hospital between Jan 2020 and Feb 2020 were included, and the main exposure of interest was the administration of intravenous corticosteroids. The primary outcome was 28-day mortality. Marginal structural modeling was used to account for baseline and time-dependent confounders. An unsupervised machine learning approach was carried out to identify phenotypes of COVID-19.

RESULTS : A total of 428 patients were included, and 280/428 (65.4%) patients received corticosteroid therapy. The 28-day mortality was significantly higher in patients who received corticosteroid therapy than in those who did not (53.9% vs. 19.6%; p<0.0001). After marginal structural modeling, corticosteroid therapy was not significantly associated with 28-day mortality (HR 0.80, 95% CI 0.54-1.18; p=0.26). Our analysis identified two phenotypes of COVID-19, and compared to the hypoinflammatory phenotype, the hyperinflammatory phenotype was characterized by elevated levels of proinflammatory cytokines, higher SOFA scores and higher rates of complications. Corticosteroid therapy was associated with a reduced 28-day mortality (HR 0.45; 95% CI 0.25-0.80; p=0.0062) in patients with hyperinflammatory phenotype.

INTERPRETATION : For critically ill patients with COVID-19, corticosteroid therapy was not associated with 28-day mortality, but the use of corticosteroids showed significant survival benefits in patients with the hyperinflammatory phenotype.

Chen Hui, Xie Jianfeng, Su Nan, Wang Jun, Sun Qin, Li Shusheng, Jin Jun, Zhou Jing, Mo Min, Wei Yao, Chao Yali, Hu Weiwei, Du Bin, Qiu Haibo

2020-Dec-11

COVID-19, Corticosteroid, Phenotype

Surgery

Artificial intelligence based real-time microcirculation analysis system for laparoscopic colorectal surgery.

In World journal of gastroenterology ; h5-index 103.0

BACKGROUND : Colonic perfusion status can be assessed easily by indocyanine green (ICG) angiography to predict ischemia related anastomotic complications during laparoscopic colorectal surgery. Recently, various parameter-based perfusion analysis have been studied for quantitative evaluation, but the analysis results differ depending on the use of quantitative parameters due to differences in vascular anatomical structure. Therefore, it can help improve the accuracy and consistency by artificial intelligence (AI) based real-time analysis microperfusion (AIRAM).

AIM : To evaluate the feasibility of AIRAM to predict the risk of anastomotic complication in the patient with laparoscopic colorectal cancer surgery.

METHODS : The ICG curve was extracted from the region of interest (ROI) set in the ICG fluorescence video of the laparoscopic colorectal surgery. Pre-processing was performed to reduce AI performance degradation caused by external environment such as background, light source reflection, and camera shaking using MATLAB 2019 on an I7-8700k Intel central processing unit (CPU) PC. AI learning and evaluation were performed by dividing into a training patient group (n = 50) and a test patient group (n = 15). Training ICG curve data sets were classified and machine learned into 25 ICG curve patterns using a self-organizing map (SOM) network. The predictive reliability of anastomotic complications in a trained SOM network is verified using test set.

RESULTS : AI-based risk and the conventional quantitative parameters including T_1/2max , time ratio (TR), and rising slope (RS) were consistent when colonic perfusion was favorable as steep increasing ICG curve pattern. When the ICG graph pattern showed stepped rise, the accuracy of conventional quantitative parameters decreased, but the AI-based classification maintained accuracy consistently. The receiver operating characteristic curves for conventional parameters and AI-based classification were comparable for predicting the anastomotic complication risks. Statistical performance verifications were improved in the AI-based analysis. AI analysis was evaluated as the most accurate parameter to predict the risk of anastomotic complications. The F1 score of the AI-based method increased by 31% for T_1/2max , 8% for TR, and 8% for RS. The processing time of AIRAM was measured as 48.03 s, which was suitable for real-time processing.

CONCLUSION : In conclusion, AI-based real-time microcirculation analysis had more accurate and consistent performance than the conventional parameter-based method.

Park Sang-Ho, Park Hee-Min, Baek Kwang-Ryul, Ahn Hong-Min, Lee In Young, Son Gyung Mo

2020-Nov-28

Anastomotic complications, Artificial intelligent, Colorectal surgery, Indocyanine green, Laparoscopic, Microcirculation analysis

Internal Medicine

Emerging use of artificial intelligence in inflammatory bowel disease.

In World journal of gastroenterology ; h5-index 103.0

Inflammatory bowel disease (IBD) is a complex, immune-mediated gastrointestinal disorder with ill-defined etiology, multifaceted diagnostic criteria, and unpredictable treatment response. Innovations in IBD diagnostics, including developments in genomic sequencing and molecular analytics, have generated tremendous interest in leveraging these large data platforms into clinically meaningful tools. Artificial intelligence, through machine learning facilitates the interpretation of large arrays of data, and may provide insight to improving IBD outcomes. While potential applications of machine learning models are vast, further research is needed to generate standardized models that can be adapted to target IBD populations.

Kohli Arushi, Holzwanger Erik A, Levy Alexander N

2020-Nov-28

Artificial intelligence, Automated diagnostics, Colorectal neoplasia screening, Machine learning, Multiomic data, Predictive models

General

Model-size reduction for reservoir computing by concatenating internal states through time.

In Scientific reports ; h5-index 158.0

Reservoir computing (RC) is a machine learning algorithm that can learn complex time series from data very rapidly based on the use of high-dimensional dynamical systems, such as random networks of neurons, called "reservoirs." To implement RC in edge computing, it is highly important to reduce the amount of computational resources that RC requires. In this study, we propose methods that reduce the size of the reservoir by inputting the past or drifting states of the reservoir to the output layer at the current time step. To elucidate the mechanism of model-size reduction, the proposed methods are analyzed based on information processing capacity proposed by Dambre et al. (Sci Rep 2:514, 2012). In addition, we evaluate the effectiveness of the proposed methods on time-series prediction tasks: the generalized Hénon-map and NARMA. On these tasks, we found that the proposed methods were able to reduce the size of the reservoir up to one tenth without a substantial increase in regression error.

Sakemi Yusuke, Morino Kai, Leleu Timothée, Aihara Kazuyuki

2020-Dec-11

General

Symmetry prediction and knowledge discovery from X-ray diffraction patterns using an interpretable machine learning approach.

In Scientific reports ; h5-index 158.0

Determination of crystal system and space group in the initial stages of crystal structure analysis forms a bottleneck in material science workflow that often requires manual tuning. Herein we propose a machine-learning (ML)-based approach for crystal system and space group classification based on powder X-ray diffraction (XRD) patterns as a proof of concept using simulated patterns. Our tree-ensemble-based ML model works with nearly or over 90% accuracy for crystal system classification, except for triclinic cases, and with 88% accuracy for space group classification with five candidates. We also succeeded in quantifying empirical knowledge vaguely shared among experts, showing the possibility for data-driven discovery of unrecognised characteristics embedded in experimental data by using an interpretable ML approach.

Suzuki Yuta, Hino Hideitsu, Hawai Takafumi, Saito Kotaro, Kotsugi Masato, Ono Kanta

2020-Dec-11

Radiology

Fully automated detection and segmentation of intracranial aneurysms in subarachnoid hemorrhage on CTA using deep learning.

In Scientific reports ; h5-index 158.0

In aneurysmal subarachnoid hemorrhage (aSAH), accurate diagnosis of aneurysm is essential for subsequent treatment to prevent rebleeding. However, aneurysm detection proves to be challenging and time-consuming. The purpose of this study was to develop and evaluate a deep learning model (DLM) to automatically detect and segment aneurysms in patients with aSAH on computed tomography angiography. In this retrospective single-center study, three different DLMs were trained on 68 patients with 79 aneurysms treated for aSAH (2016-2017) using five-fold-cross-validation. Their outputs were combined to a single DLM via ensemble-learning. The DLM was evaluated on an independent test set consisting of 185 patients with 215 aneurysms (2010-2015). Independent manual segmentations of aneurysms in a 3D voxel-wise manner by two readers (neurosurgeon, radiologist) provided the reference standard. For aneurysms > 30 mm³ (mean diameter of ~ 4 mm) on the test set, the DLM provided a detection sensitivity of 87% with false positives (FPs)/scan of 0.42. Automatic segmentations achieved a median dice similarity coefficient (DSC) of 0.80 compared to the reference standard. Aneurysm location (anterior vs. posterior circulation; P = .07) and bleeding severity (Fisher grade ≤ 3 vs. 4; P = .33) did not impede detection sensitivity or segmentation performance. For aneurysms > 100 mm³ (mean diameter of ~ 6 mm), a sensitivity of 96% with DSC of 0.87 and FPs/scan of 0.14 were obtained. In the present study, we demonstrate that the proposed DLM detects and segments aneurysms > 30 mm³ in patients with aSAH with high sensitivity independent of cerebral circulation and bleeding severity while producing FP findings of less than one per scan. Hence, the DLM can potentially assist treating physicians in aSAH by providing automated detection and segmentations of aneurysms.

Shahzad Rahil, Pennig Lenhard, Goertz Lukas, Thiele Frank, Kabbasch Christoph, Schlamann Marc, Krischek Boris, Maintz David, Perkuhn Michael, Borggrefe Jan

2020-Dec-11

General

Automatic seizure detection based on imaged-EEG signals through fully convolutional networks.

In Scientific reports ; h5-index 158.0

Seizure detection is a routine process in epilepsy units requiring manual intervention of well-trained specialists. This process could be extensive, inefficient and time-consuming, especially for long term recordings. We proposed an automatic method to detect epileptic seizures using an imaged-EEG representation of brain signals. To accomplish this, we analyzed EEG signals from two different datasets: the CHB-MIT Scalp EEG database and the EPILEPSIAE project that includes scalp and intracranial recordings. We used fully convolutional neural networks to automatically detect seizures. For our best model, we reached average accuracy and specificity values of 99.3% and 99.6%, respectively, for the CHB-MIT dataset, and corresponding values of 98.0% and 98.3% for the EPILEPSIAE patients. For these patients, the inclusion of intracranial electrodes together with scalp ones increased the average accuracy and specificity values to 99.6% and 58.3%, respectively. Regarding the other metrics, our best model reached average precision of 62.7%, recall of 58.3%, F-measure of 59.0% and AP of 54.5% on the CHB-MIT recordings, and comparatively lowers performances for the EPILEPSIAE dataset. For both databases, the number of false alarms per hour reached values less than 0.5/h for 92% of the CHB-MIT patients and less than 1.0/h for 80% of the EPILEPSIAE patients. Compared to recent studies, our lightweight approach does not need any estimation of pre-selected features and demonstrates high performances with promising possibilities for the introduction of such automatic methods in the clinical practice.

Gómez Catalina, Arbeláez Pablo, Navarrete Miguel, Alvarado-Rojas Catalina, Le Van Quyen Michel, Valderrama Mario

2020-Dec-11

General

Learning grain boundary segregation energy spectra in polycrystals.

In Nature communications ; h5-index 260.0

The segregation of solute atoms at grain boundaries (GBs) can profoundly impact the structural properties of metallic alloys, and induce effects that range from strengthening to embrittlement. And, though known to be anisotropic, there is a limited understanding of the variation of solute segregation tendencies across the full, multidimensional GB space, which is critically important in polycrystals where much of that space is represented. Here we develop a machine learning framework that can accurately predict the segregation tendency-quantified by the segregation enthalpy spectrum-of solute atoms at GB sites in polycrystals, based solely on the undecorated (pre-segregation) local atomic environment of such sites. We proceed to use the learning framework to scan across the alloy space, and build an extensive database of segregation energy spectra for more than 250 metal-based binary alloys. The resulting machine learning models and segregation database are key to unlocking the full potential of GB segregation as an alloy design tool, and enable the design of microstructures that maximize the useful impacts of segregation.

Wagih Malik, Larsen Peter M, Schuh Christopher A

2020-Dec-11

General

Machine learning uncovers independently regulated modules in the Bacillus subtilis transcriptome.

In Nature communications ; h5-index 260.0

The transcriptional regulatory network (TRN) of Bacillus subtilis coordinates cellular functions of fundamental interest, including metabolism, biofilm formation, and sporulation. Here, we use unsupervised machine learning to modularize the transcriptome and quantitatively describe regulatory activity under diverse conditions, creating an unbiased summary of gene expression. We obtain 83 independently modulated gene sets that explain most of the variance in expression and demonstrate that 76% of them represent the effects of known regulators. The TRN structure and its condition-dependent activity uncover putative or recently discovered roles for at least five regulons, such as a relationship between histidine utilization and quorum sensing. The TRN also facilitates quantification of population-level sporulation states. As this TRN covers the majority of the transcriptome and concisely characterizes the global expression state, it could inform research on nearly every aspect of transcriptional regulation in B. subtilis.

Rychel Kevin, Sastry Anand V, Palsson Bernhard O

2020-12-11

General

Integrative analysis of DNA methylation and gene expression identified cervical cancer-specific diagnostic biomarkers.

In Signal transduction and targeted therapy

Cervical cancer is the leading cause of death among women with cancer worldwide. Here, we performed an integrative analysis of Illumina HumanMethylation450K and RNA-seq data from TCGA to identify cervical cancer-specific DNA methylation markers. We first identified differentially methylated and expressed genes and examined the correlation between DNA methylation and gene expression. The DNA methylation profiles of 12 types of cancers, including cervical cancer, were used to generate a candidate set, and machine-learning techniques were adopted to define the final cervical cancer-specific markers in the candidate set. Then, we assessed the protein levels of marker genes by immunohistochemistry by using tissue arrays containing 93 human cervical squamous cell carcinoma samples and cancer-adjacent normal tissues. Promoter methylation was negatively correlated with the local regulation of gene expression. In the distant regulation of gene expression, the methylation of hypermethylated genes was more likely to be negatively correlated with gene expression, while the methylation of hypomethylated genes was more likely to be positively correlated with gene expression. Moreover, we identified four cervical cancer-specific methylation markers, cg07211381 (RAB3C), cg12205729 (GABRA2), cg20708961 (ZNF257), and cg26490054 (SLC5A8), with 96.2% sensitivity and 95.2% specificity by using the tenfold cross-validation of TCGA data. The four markers could distinguish tumors from normal tissues with a 94.2, 100, 100, and 100% AUC in four independent validation sets from the GEO database. Overall, our study demonstrates the potential use of methylation markers in cervical cancer diagnosis and may boost the development of new epigenetic therapies.

Xu Wanxue, Xu Mengyao, Wang Longlong, Zhou Wei, Xiang Rong, Shi Yi, Zhang Yunshan, Piao Yongjun

2019-Dec-13

General

A social engineering model for poverty alleviation.

In Nature communications ; h5-index 260.0

Poverty, the quintessential denominator of a developing nation, has been traditionally defined against an arbitrary poverty line; individuals (or countries) below this line are deemed poor and those above it, not so! This has two pitfalls. First, absolute reliance on a single poverty line, based on basic food consumption, and not on total consumption distribution, is only a partial poverty index at best. Second, a single expense descriptor is an exogenous quantity that does not evolve from income-expenditure statistics. Using extensive income-expenditure statistics from India, here we show how a self-consistent endogenous poverty line can be derived from an agent-based stochastic model of market exchange, combining all expenditure modes (basic food, other food and non-food), whose parameters are probabilistically estimated using advanced Machine Learning tools. Our mathematical study establishes a consumption based poverty measure that combines labor, commodity, and asset market outcomes, delivering an excellent tool for economic policy formulation.

Chattopadhyay Amit K, Kumar T Krishna, Rice Iain

2020-12-11

General

Potential predictors of type-2 diabetes risk: machine learning, synthetic data and wearable health devices.

In BMC bioinformatics

BACKGROUND : The aim of a recent research project was the investigation of the mechanisms involved in the onset of type 2 diabetes in the absence of familiarity. This has led to the development of a computational model that recapitulates the aetiology of the disease and simulates the immunological and metabolic alterations linked to type-2 diabetes subjected to clinical, physiological, and behavioural features of prototypical human individuals.

RESULTS : We analysed the time course of 46,170 virtual subjects, experiencing different lifestyle conditions. We then set up a statistical model able to recapitulate the simulated outcomes.

CONCLUSIONS : The resulting machine learning model adequately predicts the synthetic dataset and can, therefore, be used as a computationally-cheaper version of the detailed mathematical model, ready to be implemented on mobile devices to allow self-assessment by informed and aware individuals. The computational model used to generate the dataset of this work is available as a web-service at the following address: http://kraken.iac.rm.cnr.it/T2DM .

Stolfi Paola, Valentini Ilaria, Palumbo Maria Concetta, Tieri Paolo, Grignolio Andrea, Castiglione Filippo

2020-Dec-14

Computational modeling, Emulator, Machine learning, Random forest, Synthetic data, T2D

General

Correction: Identifying and Ranking Common COVID-19 Symptoms From Tweets in Arabic: Content Analysis.

In Journal of medical Internet research ; h5-index 88.0

[This corrects the article DOI: 10.2196/21329.].

Alanazi Eisa, Alashaikh Abdulaziz, Alqurashi Sarah, Alanazi Aued

2020-Dec-14

General

Indecision Modeling

ArXiv Preprint

AI systems are often used to make or contribute to important decisions in a growing range of applications, including criminal justice, hiring, and medicine. Since these decisions impact human lives, it is important that the AI systems act in ways which align with human values. Techniques for preference modeling and social choice help researchers learn and aggregate peoples' preferences, which are used to guide AI behavior; thus, it is imperative that these learned preferences are accurate. These techniques often assume that people are willing to express strict preferences over alternatives; which is not true in practice. People are often indecisive, and especially so when their decision has moral implications. The philosophy and psychology literature shows that indecision is a measurable and nuanced behavior -- and that there are several different reasons people are indecisive. This complicates the task of both learning and aggregating preferences, since most of the relevant literature makes restrictive assumptions on the meaning of indecision. We begin to close this gap by formalizing several mathematical \emph{indecision} models based on theories from philosophy, psychology, and economics; these models can be used to describe (indecisive) agent decisions, both when they are allowed to express indecision and when they are not. We test these models using data collected from an online survey where participants choose how to (hypothetically) allocate organs to patients waiting for a transplant.

Duncan C McElfresh, Lok Chan, Kenzie Doyle, Walter Sinnott-Armstrong, Vincent Conitzer, Jana Schaich Borg, John P Dickerson

2020-12-15

Pathology

Deep learning-based cross-classifications reveal conserved spatial behaviors within tumor histological images.

In Nature communications ; h5-index 260.0

Histopathological images are a rich but incompletely explored data type for studying cancer. Manual inspection is time consuming, making it challenging to use for image data mining. Here we show that convolutional neural networks (CNNs) can be systematically applied across cancer types, enabling comparisons to reveal shared spatial behaviors. We develop CNN architectures to analyze 27,815 hematoxylin and eosin scanned images from The Cancer Genome Atlas for tumor/normal, cancer subtype, and mutation classification. Our CNNs are able to classify TCGA pathologist-annotated tumor/normal status of whole slide images (WSIs) in 19 cancer types with consistently high AUCs (0.995 ± 0.008), as well as subtypes with lower but significant accuracy (AUC 0.87 ± 0.1). Remarkably, tumor/normal CNNs trained on one tissue are effective in others (AUC 0.88 ± 0.11), with classifier relationships also recapitulating known adenocarcinoma, carcinoma, and developmental biology. Moreover, classifier comparisons reveal intra-slide spatial similarities, with an average tile-level correlation of 0.45 ± 0.16 between classifier pairs. Breast cancers, bladder cancers, and uterine cancers have spatial patterns that are particularly easy to detect, suggesting these cancers can be canonical types for image analysis. Patterns for TP53 mutations can also be detected, with WSI self- and cross-tissue AUCs ranging from 0.65-0.80. Finally, we comparatively evaluate CNNs on 170 breast and colon cancer images with pathologist-annotated nuclei, finding that both cellular and intercellular regions contribute to CNN accuracy. These results demonstrate the power of CNNs not only for histopathological classification, but also for cross-comparisons to reveal conserved spatial behaviors across tumors.

Noorbakhsh Javad, Farahmand Saman, Foroughi Pour Ali, Namburi Sandeep, Caruana Dennis, Rimm David, Soltanieh-Ha Mohammad, Zarringhalam Kourosh, Chuang Jeffrey H

2020-12-11

General

3D PBV-Net: An automated prostate MRI data segmentation method.

In Computers in biology and medicine

Prostate cancer is one of the most common deadly diseases in men worldwide, which is seriously affecting people's life and health. Reliable and automated segmentation of the prostate gland in MRI data is exceptionally critical for diagnosis and treatment planning of prostate cancer. Although many automated segmentation methods have emerged, including deep learning based approaches, segmentation performance is still poor due to the large variability of image appearance, anisotropic spatial resolution, and imaging interference. This study proposes an automated prostate MRI data segmentation approach using bicubic interpolation with improved 3D V-Net (dubbed 3D PBV-Net). Considering the low-frequency components in the prostate gland, the bicubic interpolation is applied to preprocess the MRI data. On this basis, a 3D PBV-Net is developed to perform prostate MRI data segmentation. To illustrate the effectiveness of our approach, we evaluate the proposed 3D PBV-Net on two clinical prostate MRI data datasets, i.e., PROMISE 12 and TPHOH, with the manual delineations available as the ground truth. Our approach generates promising segmentation results, which have achieved 97.65% and 98.29% of average accuracy, 0.9613 and 0.9765 of Dice metric, 3.120 mm and 0.9382 mm of Hausdorff distance, and average boundary distance of 1.708, 0.7950 on PROMISE 12 and TPHOH datasets, respectively. Our method has effectively improved the accuracy of automated segmentation of the prostate MRI data and is promising to meet the accuracy requirements for telehealth applications.

Jin Yao, Yang Guang, Fang Ying, Li Ruipeng, Xu Xiaomei, Liu Yongkai, Lai Xiaobo

2020-Dec-07

Automated segmentation, Enabling technology, MRI, Prostate cancer, Telehealth care

General

Enhancing phage therapy through synthetic biology and genome engineering.

In Current opinion in biotechnology

The antimicrobial and therapeutic efficacy of bacteriophages is currently limited, mostly due to rapid emergence of phage-resistance and the inability of most phage isolates to bind and infect a broad range of clinical strains. Here, we discuss how phage therapy can be improved through recent advances in genetic engineering. First, we outline how receptor-binding proteins and their relevant structural domains are engineered to redirect phage specificity and to avoid resistance. Next, we summarize how phages are reprogrammed as prokaryotic gene therapy vectors that deliver antimicrobial 'payload' proteins, such as sequence-specific nucleases, to target defined cells within complex microbiomes. Finally, we delineate big data- and novel artificial intelligence-driven approaches that may guide the design of improved synthetic phage in the future.

Lenneman Bryan R, Fernbach Jonas, Loessner Martin J, Lu Timothy K, Kilcher Samuel

2020-Dec-10

Radiology

Classification of evoked responses to inverted faces reveals both spatial and temporal cortical response abnormalities in Autism spectrum disorder.

In NeuroImage. Clinical

The neurophysiology of face processing has been studied extensively in the context of social impairments associated with autism spectrum disorder (ASD), but the existing studies have concentrated mainly on univariate analyses of responses to upright faces, and, less frequently, inverted faces. The small number of existing studies on neurophysiological responses to inverted face in ASD have used univariate approaches, with divergent results. Here, we used a data-driven, classification-based, multivariate machine learning decoding approach to investigate the temporal and spatial properties of the neurophysiological evoked response for upright and inverted faces, relative to the neurophysiological evoked response for houses, a neutral stimulus. 21 (2 females) ASD and 29 (4 females) TD participants ages 7 to 19 took part in this study. Group level classification accuracies were obtained for each condition, using first the temporal domain of the evoked responses, and then the spatial distribution of the evoked responses on the cortical surface, each separately. We found that classification of responses to inverted neutral faces vs. houses was less accurate in ASD compared to TD, in both the temporal and spatial domains. In contrast, there were no group differences in the classification of evoked responses to upright neutral faces relative to houses. Using the classification in the temporal domain, lower decoding accuracies in ASD were found around 120 ms and 170 ms, corresponding the known components of the evoked responses to faces. Using the classification in the spatial domain, lower decoding accuracies in ASD were found in the right superior marginal gyrus (SMG), intra-parietal sulcus (IPS) and posterior superior temporal sulcus (pSTS), but not in core face processing areas. Importantly, individual classification accuracies from both the temporal and spatial classifiers correlated with ASD severity, confirming the relevance of the results to the ASD phenotype.

Nunes Adonay S, Mamashli Fahimeh, Kozhemiako Nataliia, Khan Sheraz, McGuiggan Nicole M, Losh Ainsley, Joseph Robert M, Ahveninen Jyrki, Doesburg Sam M, Hämäläinen Matti S, Kenet Tal

2020-Nov-30

Autism, Faces, Inverted faces, Machine learning, Magenetoencephalography

General

Monitoring urban black-odorous water by using hyperspectral data and machine learning.

In Environmental pollution (Barking, Essex : 1987)

Economic development, population growth, industrialization, and urbanization dramatically increase urban water quality deterioration, and thereby endanger human life and health. However, there are not many efficient methods and techniques to monitor urban black and odorous water (BOW) pollution. Our research aims at identifying primary indicators of urban BOW through their spectral characteristics and differentiation. This research combined ground in-situ water quality data with ground hyperspectral data collected from main urban BOWs in Guangzhou, China, and integrated factorial data mining and machine learning techniques to investigate how to monitor urban BOW. Eight key water quality parameters at 52 sample sites were used to retrieve three latent dimensions of urban BOW quality by factorial data mining. The synchronically measured hyperspectral bands along with the band combinations were examined by the machine learning technique, Lasso regression, to identify the most correlated bands and band combinations, over which three multiple regression models were fitted against three latent water quality indicators to determine which spectral bands were highly sensitive to three dimensions of urban BOW pollution. The findings revealed that the many sensitive bands were concentrated in higher hyperspectral band ranges, which supported the unique contribution of hyperspectral data for monitoring water quality. In addition, this integrated data mining and machine learning approach overcame the limitations of conventional band selection, which focus on a limited number of band ratios, band differences, and reflectance bands in the lower range of infrared region. The outcome also indicated that the integration of dimensionality reduction with feature selection shows good potential for monitoring urban BOW. This new analysis framework can be used in urban BOW monitoring and provides scientific data for policymakers to monitor it.

Sarigai Yang, Ji Zhou, Alicia Han, Liusheng Li, Yong Xie

2020-Nov-27

BOW, Factorial data mining, Hyperspectral data, Lasso-based machine learning, Water quality parameters

General

Injury severity analysis of pedestrian and bicyclist trespassing crashes at non-crossings: A hybrid predictive text analytics and heterogeneity-based statistical modeling approach.

In Accident; analysis and prevention

Non-motorists involved in rail-trespassing crashes are usually more vulnerable to receiving major or fatal injuries. Previous research has used traditional quantitative crash data for understanding factors contributing to injury outcomes of non-motorists in train involved collisions. However, usually overlooked crash narratives can provide useful and unique contextual crash-specific information regarding factors associated with injury outcomes. The main objective of this study is to harness the rapid advancements in more sophisticated qualitative analysis procedures for identifying thematic concepts in unstructured crash narrative data. A two-staged hybrid approach is proposed where text mining is applied first to extract valuable information from crash narratives followed by inclusion of the new variables derived from text mining in formulation of advanced statistical models for injury outcomes. By using ten-year (2006-2015) non-motorist non-crossing trespassing injury data obtained from the Federal Railroad Administration, statistical procedures and advanced machine learning text analytics are applied to extract unique information on contributory factors of trespassers' injury outcomes. The key concepts are systematically categorized into trespasser, injury, train, medical, and location related factors. A total of 13 unique variables are extracted from the thematic concepts that are not present in traditional tabular crash data. The analysis reveals a positive statistically significant association between presence of crash narrative and trespasser's injury outcome (coded as minor, major, and fatal injury). Compared to crashes with minor injuries, crashes involving major and fatal injuries are more likely to be reported with crash narratives. A crosstabulation of new variables derived from text mining with injury outcomes revealed that trespassers with confirmed suicide attempts, trespassers wearing headphones, or talking on cell phones are more likely to receive fatal injuries. Among other factors identified, trespassers under alcohol influence, trespasser hit by commuter train, and advance warnings by engineer are associated with more severe (major and fatal) trespasser injury outcomes. Accounting for unobserved heterogeneity and controlling for other factors, fixed and random parameter discrete outcome models are developed to understand the heterogeneous correlations between trespasser injury outcomes and the new crash specific explanatory variables derived from text mining - providing deeper insights. Practical implications and future research directions are discussed.

Wali Behram, Khattak Asad J, Ahmad Numan

2020-Dec-09

Concept/Entity extraction, Dynamic factor analysis, Heterogeneity-based discrete outcome modeling, Injury severity, Machine learning, Non-crossings, Non-motorist trespassing crashes, Text analysis

General

Automated detection of pneumoconiosis with multilevel deep features learned from chest X-Ray radiographs.

In Computers in biology and medicine

Early detection of pneumoconiosis in X-Rays has been a challenging task that leads to high inter- and intra-reader variability. Motivated by the success of deep learning in general and medical image classification, this paper proposes an approach to automatically detect pneumoconiosis using a deep feature based binary classifier. The features are extracted from X-rays using deep transfer learning, comprising both low and high-level feature sets. For this, a CNN model pre-trained with a transfer learning from a CheXNet model was initially used to extract deep features from the X-Ray images, then the deep features were mapped to higher-dimensional feature spaces for classification using Support Vector Machine (SVM) and CNN based feature aggregation methods. In order to cross validate the proposed method, the training and testing images were randomly split into three folds before each experiment. Nine evaluation metrics were employed to compare the performance of the proposed method and state-of-the-art methods from the literature that used the same datasets. The experimental results show that the proposed framework outperformed others, achieving an accuracy of 92.68% in the automated detection of pneumoconiosis.

Devnath Liton, Luo Suhuai, Summons Peter, Wang Dadong

2020-Nov-21

Black lung, “Coal workers pneumoconiosis (CWP)”, Computer-aided diagnosis, Deep transfer learning, Support vector machine, X-rays

Public Health

A practical framework for predicting residential indoor PM2.5 concentration using land-use regression and machine learning methods.

In Chemosphere

People typically spend most of their time indoors. It is of importance to establish prediction models to estimate PM_2.5 concentration in indoor environments (e.g., residential households) to allow accurate assessments of exposure in epidemiological studies. This study aimed to develop models to predict PM_2.5 concentration in residential households. PM_2.5 concentration and related parameters (e.g., basic information about the households and ventilation settings) were collected in 116 households during the winter and summer seasons in Hong Kong. Outdoor PM_2.5 concentration at households was estimated using a land-use regression model. The random forest machine learning algorithm was then applied to develop indoor PM_2.5 prediction models. The results show that the random forest model achieved a promising predictive accuracy, with R² and cross-validation R² values of 0.93 and 0.65, respectively. Outdoor PM_2.5 concentration was the most important predictor variable, followed in descending order by the household marked number, outdoor temperature, outdoor relative humidity, average household area and air conditioning. The external validation result using an independent dataset confirmed the potential application of the random forest model, with an R² value of 0.47. Overall, this study shows the value of a combined land-use regression and machine learning approach in establishing indoor PM_2.5 prediction models that provide a relatively accurate assessment of exposure for use in epidemiological studies.

Li Zhiyuan, Tong Xinning, Ho Jason Man Wai, Kwok Timothy C Y, Dong Guanghui, Ho Kin-Fai, Yim Steve Hung Lam

2020-Dec-02

Households, Indoor air, PM(2.5), Prediction model, Random forest

General

T lymphocytes from malignant hyperthermia-susceptible mice display aberrations in intracellular calcium signaling and mitochondrial function.

In Cell calcium

Gain-of-function RyR1-p.R163C mutation in ryanodine receptors type 1 (RyR1) deregulates Ca²⁺ signaling and mitochondrial function in skeletal muscle and causes malignant hyperthermia in humans and mice under triggering conditions. We investigated whether T lymphocytes from heterozygous RyR1-p.R163C knock-in mutant mice (HET T cells) display measurable aberrations in resting cytosolic Ca²⁺ concentration ([Ca²⁺]_i), Ca²⁺ release from the store, store-operated Ca²⁺ entry (SOCE), and mitochondrial inner membrane potential (ΔΨ_m) compared with T lymphocytes from wild-type mice (WT T cells). We explored whether these variables can be used to distinguish between T cells with normal and altered RyR1 genotype. HET and WT T cells were isolated from spleen and lymph nodes and activated in vitro using phytohemagglutinin P. [Ca²⁺]_i and ΔΨ_m dynamics were examined using Fura 2 and tetramethylrhodamine methyl ester fluorescent dyes, respectively. Activated HET T cells displayed elevated resting [Ca²⁺]_i, diminished responses to Ca²⁺ mobilization with thapsigargin, and decreased rate of [Ca²⁺]_i elevation in response to SOCE compared with WT T cells. Pretreatment of HET T cells with ryanodine or dantrolene sodium reduced disparities in the resting [Ca²⁺]_i and ability of thapsigargin to mobilize Ca²⁺ between HET and WT T cells. While SOCE elicited dissipation of the ΔΨ_m in WT T cells, it produced ΔΨ_m hyperpolarization in HET T cells. When used as the classification variable, the amplitude of thapsigargin-induced Ca²⁺ transient showed the best promise in predicting the presence of RyR1-p.R163C mutation. Other significant variables identified by machine learning analysis were the ratio of resting cytosolic Ca²⁺ level to the amplitude of thapsigargin-induced Ca²⁺ transient and an integral of changes in ΔΨ_m in response to SOCE. Our study demonstrated that gain-of-function mutation in RyR1 significantly affects Ca²⁺ signaling and mitochondrial fiction in T lymphocytes, which suggests that this mutation may cause altered immune responses in its carrier. Our data link the RyR1-p.R163C mutation, which causes inherited skeletal muscle diseases, to deregulation of Ca²⁺ signaling and mitochondrial function in immune T cells and establish proof-of-principle for in vitro T cell-based diagnostic assay for hereditary RyR1 hyperfunction.

Yang Lukun, Dedkova Elena N, Allen Paul D, Jafri M Saleet, Fomina Alla F

2020-Dec-01

Dantrolene sodium, Intracellular Ca(2+), Mitochondrial potential, RYR1-p.R163C knock-in mice, Ryanodine receptor, T lymphocytes

Surgery

"Robotic surgery: the impact of simulation and other innovative platforms on performance and training".

In Journal of minimally invasive gynecology ; h5-index 40.0

OBJECTIVE : To review the current status of robotic training and the impact of various training platforms on the performance of robotic surgical trainees.

DATA SOURCES : Literature review of Google Scholar and PubMed. Search terms included a combination of the following: "robotic training", "simulation", "robotic curriculum", "obgyn residency robotic training", "virtual reality robotic training", "DaVinci training", "surgical simulation", "gyn surgical training". Sources considered for inclusion included peer reviewed articles, literature reviews, textbook chapters, and statements from various institutions involved in resident training.

METHODS OF STUDY SELECTION : A literature search of Google Scholar and PubMed using terms related to robotic surgery and robotics training, as mentioned above.

RESULTS : Multiple novel platforms that utilize machine learning and real time video feedback to teach and evaluate robotic surgical skills have been developed over recent years. Various training curricula, VR simulators, and other robotic training tools have shown to enhance robotic surgical education and improve surgical skills. Integration of didactic learning, simulation, and intraoperative teaching into more comprehensive training curricula shows positive effects on robotic skills proficiency. Few robotic surgery training curricula have been validated through peer reviewed study, and there is more work to be completed in this area. In addition, there is a lack of information about how skills obtained through robotics curricula and simulation translates into operating room performance and patient outcomes.

CONCLUSION : Data collected to date shows promising advances in training of robotic surgeons. A diverse array of curricula for training robotic surgeons continues to emerge, and existing teaching modalities are evolving to keep up with the rapid demand for proficient robotic surgeons. Futures areas of growth include establishing competency benchmarks for existing training tools, validating existing curricula, and determining how to translate acquired skills in simulation to performance in the operating room and patient outcomes. Many surgical training platforms are beginning to expand beyond discreet robotic skills training to procedure-specific and team training. There is still a wealth of research to be done to understand how to create an effective training experience for gyn surgical trainees and robotics teams.

Azadi Shirin, Green Isabel, Arnold Anne, Truong Mireille, Potts Jacqueline, Martino Martin A

2020-Dec-09

General

Toward computational modelling on immune system function.

In BMC bioinformatics

The 3rd edition of the computational methods for the immune system function workshop has been held in San Diego, CA, in conjunction with the IEEE International Conference on Bioinformatics and Biomedicine (BIBM 2019) from November 18 to 21, 2019. The workshop has continued its growing tendency, with a total of 18 accepted papers that have been presented in a full day workshop. Among these, the best 10 papers have been selected and extended for presentation in this special issue. The covered topics range from computer-aided identification of T cell epitopes to the prediction of heart rate variability to prevent brain injuries, from In Silico modeling of Tuberculosis and generation of digital patients to machine learning applied to predict type-2 diabetes risk.

Pappalardo Francesco, Russo Giulia, Reche Pedro A

2020-Dec-14

Radiology

Chest CT imaging features and severity scores as biomarkers for prognostic prediction in patients with COVID-19.

In Annals of translational medicine

Background : Coronavirus disease 2019 (COVID-19) has become a pandemic. Few studies have explored the role of chest computed tomography (CT) features and severity scores for prognostic prediction. In this study, we aimed to investigate the role of chest CT severity score and imaging features in the prediction of the prognosis of COVID-19 patients.

Methods : A total of 134 patients (62 recovered and 72 deceased patients) with confirmed COVID-19 were enrolled. The clinical, laboratory, and chest CT (316 scans) data were retrospectively reviewed. Demographics, symptoms, comorbidities, and temporal changes of laboratory results, CT features, and severity scores were compared between recovered and deceased groups using the Mann-Whitney U test and logistic regression to identify the risk factors for poor prognosis.

Results : Median age was 48 and 58 years for recovered and deceased patients, respectively. More patients had at least one comorbidity in the deceased group than the recovered group (60% vs. 29%). Leukocytes, neutrophil, high-sensitivity C-reactive protein (hsCRP), prothrombin, D-dimer, serum ferritin, interleukin (IL)-2, and IL-6 were significantly elevated in the deceased group than the recovered group at different stages. The total CT score at the peak stage was significantly greater in the deceased group than the recovered group (20 vs. 11 points). The optimal cutoff value of the total CT scores was 16.5 points, achieving 69.4% sensitivity and 82.2% specificity for the prognostic prediction. The crazy-paving pattern and consolidation were more common in the deceased patients than those in the recovered patients. Linear opacities significantly increased with the disease course in the recovered patients. Sex, age, neutrophil, IL-2, IL-6, and total CT scores were independent risk factors for the prognosis with odds ratios of 3.8 to 8.7.

Conclusions : Sex (male), older age (>60 years), elevated neutrophil, IL-2, IL-6 level, and total CT scores (≥16) were independent risk factors for poor prognosis in patients with COVID-19. Temporal changes of chest CT features and severity scores could be valuable for early identification of severe cases and eventually reducing the mortality rate of COVID-19.

Zhou Shuchang, Chen Chengyang, Hu Yiqi, Lv Wenzhi, Ai Tao, Xia Liming

2020-Nov

Coronavirus disease, computed tomography, prognosis, risk factor, severity score

Pathology

Enriched Annotations for Tumor Attribute Classification from Pathology Reports with Limited Labeled Data

ArXiv Preprint

Precision medicine has the potential to revolutionize healthcare, but much of the data for patients is locked away in unstructured free-text, limiting research and delivery of effective personalized treatments. Generating large annotated datasets for information extraction from clinical notes is often challenging and expensive due to the high level of expertise needed for high quality annotations. To enable natural language processing for small dataset sizes, we develop a novel enriched hierarchical annotation scheme and algorithm, Supervised Line Attention (SLA), and apply this algorithm to predicting categorical tumor attributes from kidney and colon cancer pathology reports from the University of California San Francisco (UCSF). Whereas previous work only annotated document level labels, we in addition ask the annotators to enrich the traditional label by asking them to also highlight the relevant line or potentially lines for the final label, which leads to a 20% increase of annotation time required per document. With the enriched annotations, we develop a simple and interpretable machine learning algorithm that first predicts the relevant lines in the document and then predicts the tumor attribute. Our results show across the small dataset sizes of 32, 64, 128, and 186 labeled documents per cancer, SLA only requires half the number of labeled documents as state-of-the-art methods to achieve similar or better micro-f1 and macro-f1 scores for the vast majority of comparisons that we made. Accounting for the increased annotation time, this leads to a 40% reduction in total annotation time over the state of the art.

Nick Altieri, Briton Park, Mara Olson, John DeNero, Anobel Odisho, Bin Yu

2020-12-15

General

Recent advances in constraint and machine learning-based metabolic modeling by leveraging stoichiometric balances, thermodynamic feasibility and kinetic law formalisms.

In Metabolic engineering

Understanding the governing principles behind organisms' metabolism and growth underpins their effective deployment as bioproduction chassis. A central objective of metabolic modeling is predicting how metabolism and growth are affected by both external environmental factors and internal genotypic perturbations. The fundamental concepts of reaction stoichiometry, thermodynamics, and mass action kinetics have emerged as the foundational principles of many modeling frameworks designed to describe how and why organisms allocate resources towards both growth and bioproduction. This review focuses on the latest algorithmic advancements that have integrated these foundational principles into increasingly sophisticated quantitative frameworks.

Suthers Patrick F, Foster Charles J, Sarkar Debolina, Wang Lin, Maranas Costas D

2020-Dec-10

General

Distribution of agitation and related symptoms among hospitalized patients using a scalable natural language processing method.

In General hospital psychiatry

BACKGROUND : Agitation is a common feature of many neuropsychiatric disorders.

OBJECTIVE : Understanding the prevalence, implications, and characteristics of agitation among hospitalized populations can facilitate more precise recognition of disability arising from neuropsychiatric diseases.

METHODS : We developed two agitation phenotypes using an expansion of expert curated term lists. These phenotypes were used to characterize five years of psychiatric admissions. The relationship of agitation symptoms and length of stay was examined.

RESULTS : Among 4548 psychiatric admissions, 1134 (24.9%) included documentation of agitation based on the primary agitation phenotype. These symptoms were greater among individuals with public insurance, and those with mania and psychosis compared to major depressive disorder. Greater symptoms were associated with longer hospital stay, with ~0.9 day increase in stay for every 10% increase in agitation phenotype.

CONCLUSION : Agitation was common at hospital admission and associated with diagnosis and longer length of stay. Characterizing agitation-related symptoms through natural language processing may provide new tools for understanding agitated behaviors and their relationship to delirium.

Hart Kamber L, Pellegrini Amelia M, Forester Brent P, Berretta Sabina, Murphy Shawn N, Perlis Roy H, McCoy Thomas H

2020-Nov-10

Data mining, Electronic health records, Machine learning, Natural language processing, Psychomotor agitation

Radiology

Comparison of Segmentation-Free and Segmentation-Dependent Computer-Aided Diagnosis of Breast Masses on a Public Mammography Dataset.

In Journal of biomedical informatics ; h5-index 55.0

PURPOSE : To compare machine learning methods for classifying mass lesions on mammography images that use predefined image features computed over lesion segmentations to those that leverage segmentation-free representation learning on a standard, public evaluation dataset.

METHODS : We apply several classification algorithms to the public Curated Breast Imaging Subset of the Digital Database for Screening Mammography (CBIS-DDSM), in which each image contains a mass lesion. Segmentation-free representation learning techniques for classifying lesions as benign or malignant include both a Bag-of-Visual-Words (BoVW) method and a Convolutional Neural Network (CNN). We compare classification performance of these techniques to that obtained using two different segmentation-dependent approaches from the literature that rely on specific combinations of end classifiers (e.g. linear discriminant analysis, neural networks) and predefined features computed over the lesion segmentation (e.g. spiculation measure, morphological characteristics, intensity metrics).

RESULTS : We report area under the receiver operating characteristic curve (A_Z) values for malignancy classification on CBIS-DDSM for each technique. We find average A_Z values of 0.73 for a segmentation-free BoVW method, 0.86 for a segmentation-free CNN method, 0.75 for a segmentation-dependent linear discriminant analysis of Rubber-Band Straightening Transform features, and 0.58 for a hybrid rule-based neural network classification using a small number of hand-designed features.

CONCLUSIONS : We find that malignancy classification performance on the CBIS-DDSM dataset using segmentation-free BoVW features is comparable to that of the best segmentation-dependent methods we study, but also observe that a common segmentation-free CNN model substantially and significantly outperforms each of these (p<0.05). These results reinforce recent findings suggesting that representation learning techniques such as BoVW and CNNs are advantageous for mammogram analysis because they do not require lesion segmentation, the quality and specific characteristics of which can vary substantially across datasets. We further observe that segmentation-dependent methods achieve performance levels on CBIS-DDSM inferior to those achieved on the original evaluation datasets reported in the literature. Each of these findings reinforces the need for standardization of datasets, segmentation techniques, and model implementations in performance assessments of automated classifiers for medical imaging.

Sawyer Lee Rebecca, Dunnmon Jared A, He Ann, Tang Siyi, Ré Christopher, Rubin Daniel L

2020-Dec-10

Computer Assisted Diagnosis, Deep Learning, Mammography, Segmentation

General

Population-based Screening for Hereditary Colorectal Cancer Variants in Japan.

In Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association

BACKGROUND & AIMS : Colorectal cancer (CRC) is one of the most common cancers in the world. A small proportion of CRCs can be attributed to recognizable hereditary germline variants of known CRC susceptibility genes. To better understand cancer risk, it is necessary to explore the prevalence of hereditary CRC and pathogenic variants of multiple cancer-predisposing genes in non-European populations.

METHODS : We analyzed the coding regions of 27 cancer-predisposing genes in 12,503 unselected Japanese CRC patients and 23,705 controls by target sequencing and genome-wide SNP chip. Their clinical significance was assessed using ClinVar and the guidelines by ACMG/AMP.

RESULTS : We identified 4,804 variants in the 27 genes and annotated them as pathogenic in 397, and benign variants in 941, of which 43.6% were novel. In total, 3.3% of the unselected CRC patients and 1.5% of the controls had a pathogenic variant. The pathogenic variants of MSH2 (odds ratio (OR) =18.1), MLH1 (OR=8.6), MSH6 (OR=4.9), APC (OR=49.4), BRIP1 (OR=3.6), BRCA1 (OR=2.6), BRCA2 (OR=1.9), and TP53 (OR=1.7) were significantly associated with CRC development in the Japanese population (P-values<0.01, FDR<0.05). These pathogenic variants were significantly associated with diagnosis age and personal/family history of cancer. In total, at least 3.5% of the Japanese CRC population had a pathogenic variant or CNV of the 27 cancer-predisposing genes, indicating hereditary cancers.

CONCLUSIONS : This largest study of CRC heredity in Asia can contribute to the development of guidelines for genetic testing and variant interpretation for heritable CRCs.

Fujita Masashi, Liu Xiaoxi, Iwasaki Yusuke, Terao Chikashi, Mizukami Keijiro, Kawakami Eiryo, Takata Sadaaki, Inai Chihiro, Aoi Tomomi, Mizukoshi Misaki, Maejima Kazuhiro, Hirata Makoto, Murakami Yoshinori, Kamatani Yoichiro, Kubo Michiaki, Akagi Kiwamu, Matsuda Koichi, Nakagawa Hidewaki, Momozawa Yukihide

2020-Dec-10

BRCA1/2, BRIP1, CNV, hereditary colorectal cancer, pathogenic variant

General

The role of explainability in creating trustworthy artificial intelligence for health care: a comprehensive survey of the terminology, design choices, and evaluation strategies.

In Journal of biomedical informatics ; h5-index 55.0

Artificial intelligence (AI) has huge potential to improve the health and well-being of people, but adoption in clinical practice is still limited. Lack of transparency is identified as one of the main barriers to implementation, as clinicians should be confident the AI system can be trusted. Explainable AI has the potential to overcome this issue and can be a step towards trustworthy AI. In this paper we review the recent literature to provide guidance to researchers and practitioners on the design of explainable AI systems for the health-care domain and contribute to formalization of the field of explainable AI. We argue the reason to demand explainability determines what should be explained as this determines the relative importance of the properties of explainability (i.e. interpretability and fidelity). Based on this, we propose a framework to guide the choice between classes of explainable AI methods (explainable modelling versus post-hoc explanation; model-based, attribution-based, or example-based explanations; global and local explanations). Furthermore, we find that quantitative evaluation metrics, which are important for objective standardized evaluation, are still lacking for some properties (e.g. clarity) and types of explanations (e.g. example-based methods). We conclude that explainable modelling can contribute to trustworthy AI, but the benefits of explainability still need to be proven in practice and complementary measures might be needed to create trustworthy AI in health care (e.g. reporting data quality, performing extensive (external) validation, and regulation).

Markus Aniek F, Kors Jan A, Rijnbeek Peter R

2020-Dec-09

Explainable Artificial Intelligence, Explainable Modelling, Interpretability, Post-hoc Explanation, Trustworthy Artificial Intelligence

Surgery

One of the first validations of an artificial intelligence algorithm for clinical use: The impact on intraoperative hypotension prediction and clinical decision-making.

In Surgery ; h5-index 54.0

This review describes the steps and conclusions from the development and validation of an artificial intelligence algorithm (the Hypotension Prediction Index), one of the first machine learning predictive algorithms used in the operating room environment. The algorithm has been demonstrated to reduce intraoperative hypotension in two randomized controlled trials via real-time prediction of upcoming hypotensive events prompting anesthesiologists to act earlier, more often, and differently in managing impending hypotension. However, the algorithm entails no dynamic learning process that evolves from use in clinical patient care, meaning the algorithm is fixed, and furthermore provides no insight in the decisional process that leads to an early warning for intraoperative hypotension, which makes the algorithm a "black box." Many other artificial intelligence machine learning algorithms have these same disadvantages. Clinical validation of such algorithms is relatively new and requires more standardization, as guidelines are lacking or only now start to be drafted. Before adaptation in clinical practice, impact of artificial intelligence algorithms on clinical behavior, outcomes and economic advantages should be studied too.

van der Ven Ward H, Veelo Denise P, Wijnberge Marije, van der Ster Björn J P, Vlaar Alexander P J, Geerts Bart F

2020-Dec-11

General

Population attitudes toward contraceptive methods over time on a social media platform.

In American journal of obstetrics and gynecology

BACKGROUND : Contraceptive method choice is often strongly influenced by the experiences and opinions of one's social network. Though social media, including Twitter, increasingly influences reproductive-age individuals, discussion of contraception in this setting has yet to be characterized. Natural language processing, a type of machine learning in which computers analyze natural language data, enables this analysis.

OBJECTIVES : To illuminate temporal trends in attitudes toward long- and short-acting reversible contraceptive methods in tweets between 2006 and 2019 and establish social media platforms as alternate data sources for large-scale sentiment analysis about contraception.

STUDY DESIGN : We studied English language tweets mentioning reversible prescription contraceptive methods between March 2006 (founding of Twitter) and December 2019. Tweets mentioning contraception were extracted using search terms including generic or brand names, colloquial names, and abbreviations. We characterized and performed sentiment analysis on tweets. We used Mann-Kendall nonparametric tests to assess temporal trends in the overall number and the number of positive, negative, and neutral tweets referring to each method. The code to reproduce this analysis is available at https://github.com/hms-dbmi/contraceptionOnTwitter.

RESULTS : We extracted 838,739 tweets mentioning at least one contraceptive method. The annual number of contraception-related tweets increased significantly over the study period. The intrauterine device was the most commonly-referenced method (45.9%). Long-acting methods were mentioned more often than short-acting ones (58% vs. 42%), and the annual proportion of long-acting reversible contraception-related tweets increased over time. In sentiment analysis of tweets mentioning a single contraceptive method (n=665,064), the greatest proportion of all tweets was negative (65,339 of 160,713 tweets with at least 95% confident sentiment, or 40.66%). Tweets mentioning long-acting methods were nearly twice as likely to be positive compared to tweets mentioning short-acting methods (19.65% vs. 10.21%, p<0.002).

CONCLUSION : Recognizing the influence of social networks on contraceptive decision-making, social media platforms may be useful in the collection and dissemination of information about contraception.

Merz Allison A, Gutiérrez-Sacristán Alba, Bartz Deborah, Williams Natalie E, Ojo Ayotomiwa, Schaefer Kimberly M, Huang Melody, Li Chloe Y, Sofia Sandoval Raquel, Ye Sonya, Cathcart Ann M, Starosta Anabel, Avillach Paul

2020-Dec-09

Twitter, birth control, contraception, long-acting reversible contraception, natural language processing, sentiment analysis, social media

General

Fully automatic volume segmentation of infra-renal abdominal aortic aneurysm CT images with deep learning approaches versus physician controlled manual segmentation.

In Journal of vascular surgery ; h5-index 87.0

OBJECTIVE : Imaging softwares have become critical tools in the diagnosis and decision making for the treatment of abdominal aortic aneurysms (AAA). However, the inter-observer reproducibility of maximum cross-section diameter is poor. This study aimed to present and assess the quality of a new fully automated software (PRAEVAorta) that enables fast and robust detection of the aortic lumen and the infra-renal AAA characteristics including the presence of thrombus.

METHODS : To evaluate the segmentation obtained with this new software, we performed a quantitative comparison with the results obtained from a semi-automatic segmentation manually corrected by a senior and a junior surgeon on a dataset of 100 pre-operative CTAs from patients with infrarenal AAAs (i.e. 13465 slices). The Dice Similarity Coefficient (DSC), Jaccard index (JAC), Sensitivity, Specificity, volumetric similarity (VS), Hausdorff distance (HD), maximum aortic transverse diameter, and the duration of segmentation were calculated between the two methods and, for the semi-automatic software, also between the two observers.

RESULTS : The analyses demonstrated an excellent correlation of the volumes, surfaces, and diameters measured with the fully automatic and manually corrected segmentation methods, with a Pearson's coefficient correlation >0.90, P<0.0001. Overall, comparison between the fully automatic and manually corrected segmentation method by the senior surgeon revealed a mean DSC of 0.95±0.01, JAC of 0.91±0.02, sensitivity of 0.94±0.02, specificity of 0.97±0.01, VS of 0.98±0.01, and mean HD/slice of 4.61±7.26mm. The mean VS reached 0.95±0.04 for the lumen and 0.91±0.07 for the thrombus. For the fully automatic method, the segmentation time varied from 27 seconds to 4 minutes per patient vs 5 minutes to 80 minutes for the manually corrected methods (P<0.0001).

CONCLUSION : By enabling a fast and fully automated detailed analysis of the anatomic characteristics of infra-renal AAAs, this software could have strong applications in daily clinical practice and clinical research.

Caradu Caroline, Spampinato Benedetta, Vrancianu Ana Maria, Bérard Xavier, Ducasse Eric

2020-Dec-09

abdominal aortic aneurysm, artificial intelligence, automatic segmentation, deep learning, endovascular aortic repair, thrombus, volume

General

Asynchrony Between Individual and Government Actions Accounts for Disproportionate Impact of COVID-19 on Vulnerable Communities.

In American journal of preventive medicine ; h5-index 75.0

INTRODUCTION : Previously estimated effects of social distancing do not account for changes in individual behavior before the implementation of stay-at-home policies or model this behavior in relation to the burden of disease. This study aims to assess the asynchrony between individual behavior and government stay-at-home orders, quantify the true impact of social distancing using mobility data, and explore the sociodemographic variables linked to variation in social distancing practices.

METHODS : This study was a retrospective investigation that leveraged mobility data to quantify the time to behavioral change in relation to the initial presence of COVID-19 and the implementation of government stay-at-home orders. The impact of social distancing that accounts for both individual behavior and testing data was calculated using generalized mixed models. The role of sociodemographics in accounting for variation in social distancing behavior was modeled using a 10-fold cross-validated elastic net (linear machine learning model). Analysis was conducted in April‒July 2020.

RESULTS : Across all the 1,124 counties included in this analysis, individuals began to socially distance at a median of 5 days (IQR=3-8) after 10 cumulative cases of COVID-19 were confirmed in their state, with state governments taking a median of 15 days (IQR=12-19) to enact stay-at-home orders. Overall, people began social distancing at a median of 12 days (IQR=8-17) before their state enacted stay-at-home orders. Of the 16 studies included in the review, 13 exclusively used government dates as a proxy for social distancing behavior, and none accounted for both testing and mobility. Using government stay-at-home dates as a proxy for social distancing (10.2% decrease in the number of daily cases) accounted for only 55% of the true impact of the intervention when compared with estimates using mobility (18.6% reduction). Using 10-fold cross-validation, 23 of 43 sociodemographic variables were significantly and independently predictive of variation in individual social distancing, with delays corresponding to an increase in a county's proportion of people without a high school diploma and proportion of racial and ethnic minorities.

CONCLUSIONS : This retrospective analysis of mobility patterns found that social distancing behavior occurred well before the onset of government stay-at-home dates. This asynchrony leads to the underestimation of the impact of social distancing. Sociodemographic characteristics associated with delays in social distancing can help explain the disproportionate case burden and mortality among vulnerable communities.

Abdalla Moustafa, Abar Arjan, Beiter Evan R, Saad Mohamed

2020-Nov-13

Public Health

Accurate real-time monitoring of high particulate matter concentration based on holographic speckles and deep learning.

In Journal of hazardous materials

Accurate real-time monitoring of particulate matter (PM) has emerged as a global issue due to the hazardous effects of PM on public health and industry. However, conventional PM monitoring techniques are usually cumbersome and require expensive equipments. In this study, Holo-SpeckleNet is proposed as a fast and accurate PM concentration measurement technique with high throughput using a deep learning based holographic speckle pattern analysis. Speckle pattern datasets of PMs for a wide range of PM concentrations were acquired by using a digital in-line holography microscopy system. Deep autoencoder and regression algorithms were trained with the captured speckle pattern datasets to directly measure PM concentration from speckle pattern images without any air intake device and time-consuming post image processing. The proposed technique was applied to predict various PM concentrations using the test datasets, optimize hyperparameters, and compare its performance with a convolutional neural network (CNN) algorithm. As a result, high PM concentration values can be measured over air quality index of 150, above which human exposure is unhealthy. In addition, the proposed technique exhibits higher measurement accuracy and less overfitting than the CNN with a relative error of 7.46 ± 3.92%. It can be applied to design a compact air quality monitoring device for highly accurate and real-time measurement of PM concentrations under hazardous environment, such as factories or construction sites.

Kim Jihwan, Go Taesik, Lee Sang Joon

2020-Nov-19

Deep learning, Digital holographic microscopy, Particulate matter, Speckle pattern

General

Toward Healthcare Diagnoses by Machine-Learning-Enabled Volatile Organic Compound Identification.

In ACS nano ; h5-index 203.0

As a natural monitor of health conditions for human beings, volatile organic compounds (VOCs) act as significant biomarkers for healthcare monitoring and early stage diagnosis of diseases. Most existing VOC sensors use semiconductors, optics, and electrochemistry, which are only capable of measuring the total concentration of VOCs with slow response, resulting in the lack of selectivity and low efficiency for VOC detection. Infrared (IR) spectroscopy technology provides an effective solution to detect chemical structures of VOC molecules by absorption fingerprints induced by the signature vibration of chemical stretches. However, traditional IR spectroscopy for VOC detection is limited by the weak light-matter interaction, resulting in large optical paths. Leveraging the ultrahigh electric field induced by plasma, the vibration of the molecules is enhanced to improve the light-matter interaction. Herein, we report a plasma-enhanced IR absorption spectroscopy with advantages of fast response, accurate quantization, and good selectivity. An order of ∼kV voltage was achieved from the multiswitched manipulation of the triboelectric nanogenerator by repeated sliding. The VOC species and their concentrations were well-quantified from the wavelength and intensity of spectra signals with the enhancement from plasma. Furthermore, machine learning has visualized the relationship of different VOCs in the mixture, which demonstrated the feasibility of the VOC identification to mimic patients.

Zhu Jianxiong, Ren Zhihao, Lee Chengkuo

2020-Dec-14

healthcare diagnosis, machine learning, mid-infrared spectroscopy, triboelectric nanogenerator, volatile organic compound

General

Asynchrony Between Individual and Government Actions Accounts for Disproportionate Impact of COVID-19 on Vulnerable Communities.

In American journal of preventive medicine ; h5-index 75.0

INTRODUCTION : Previously estimated effects of social distancing do not account for changes in individual behavior before the implementation of stay-at-home policies or model this behavior in relation to the burden of disease. This study aims to assess the asynchrony between individual behavior and government stay-at-home orders, quantify the true impact of social distancing using mobility data, and explore the sociodemographic variables linked to variation in social distancing practices.

METHODS : This study was a retrospective investigation that leveraged mobility data to quantify the time to behavioral change in relation to the initial presence of COVID-19 and the implementation of government stay-at-home orders. The impact of social distancing that accounts for both individual behavior and testing data was calculated using generalized mixed models. The role of sociodemographics in accounting for variation in social distancing behavior was modeled using a 10-fold cross-validated elastic net (linear machine learning model). Analysis was conducted in April‒July 2020.

RESULTS : Across all the 1,124 counties included in this analysis, individuals began to socially distance at a median of 5 days (IQR=3-8) after 10 cumulative cases of COVID-19 were confirmed in their state, with state governments taking a median of 15 days (IQR=12-19) to enact stay-at-home orders. Overall, people began social distancing at a median of 12 days (IQR=8-17) before their state enacted stay-at-home orders. Of the 16 studies included in the review, 13 exclusively used government dates as a proxy for social distancing behavior, and none accounted for both testing and mobility. Using government stay-at-home dates as a proxy for social distancing (10.2% decrease in the number of daily cases) accounted for only 55% of the true impact of the intervention when compared with estimates using mobility (18.6% reduction). Using 10-fold cross-validation, 23 of 43 sociodemographic variables were significantly and independently predictive of variation in individual social distancing, with delays corresponding to an increase in a county's proportion of people without a high school diploma and proportion of racial and ethnic minorities.

CONCLUSIONS : This retrospective analysis of mobility patterns found that social distancing behavior occurred well before the onset of government stay-at-home dates. This asynchrony leads to the underestimation of the impact of social distancing. Sociodemographic characteristics associated with delays in social distancing can help explain the disproportionate case burden and mortality among vulnerable communities.

Abdalla Moustafa, Abar Arjan, Beiter Evan R, Saad Mohamed

2020-Nov-13

General

Designing a Mobile Social and Vocational Reintegration Assistant for Burn-out Outpatient Treatment

ArXiv Preprint

Using Social Agents as health-care assistants or trainers is one focus area of IVA research. While their use as physical health-care agents is well established, their employment in the field of psychotherapeutic care comes with daunting challenges. This paper presents our mobile Social Agent EmmA in the role of a vocational reintegration assistant for burn-out outpatient treatment. We follow a typical participatory design approach including experts and patients in order to address requirements from both sides. Since the success of such treatments is related to a patients emotion regulation capabilities, we employ a real-time social signal interpretation together with a computational simulation of emotion regulation that influences the agent's social behavior as well as the situational selection of verbal treatment strategies. Overall, our interdisciplinary approach enables a novel integrative concept for Social Agents as assistants for burn-out patients.

Patrick Gebhard, Tanja Schneeberger, Michael Dietz, Elisabeth André, Nida ul Habib Bajwa

2020-12-15

Ophthalmology

Current Application of Digital Diagnosing Systems for Retinopathy of Prematurity.

In Computer methods and programs in biomedicine

BACKGROUND AND OBJECTIVE : Retinopathy of prematurity (ROP), a proliferative vascular eye disease, is one of the leading causes of blindness in childhood and prevails in premature infants with low-birth-weight. The recent progress in digital image analysis offers novel strategies for ROP diagnosis. This paper provides a comprehensive review on the development of digital diagnosing systems for ROP to software researchers. It may also be adopted as a guide to ophthalmologists for selecting the most suitable diagnostic software in the clinical setting, particularly for the remote ophthalmic support.

METHODS : We review the latest literatures concerning the application of digital diagnosing systems for ROP. The diagnosing systems are analyzed and categorized. Articles published between 1998 and 2020 were screened with the two searching engines Pubmed and Google Scholar.

RESULTS : Telemedicine is a method of remote image interpretation that can provide medical service to remote regions, and yet requires training to local operators. On the basis of image collection in telemedicine, computer-based image analytical systems for ROP were later developed. So far, the aforementioned systems have been mainly developed by virtue of classic machine learning, deep learning (DL) and multiple machine learning. During the past two decades, various computer-aided systems for ROP based on classic machine learning (e.g. RISA, ROPtool, CAIER) became available and have achieved satisfactory performance. Further, automated systems for ROP diagnosis based on DL are developed for clinical applications and exhibit high accuracy. Moreover, multiple instance learning is another method to establish an automated system for ROP detection besides DL, which, however, warrants further investigation in future.

CONCLUSION : At present, the incorporation of computer-based image analysis with telemedicine potentially enables the detection, supervision and in-time treatment of ROP for the preterm babies.

Bao Yuekun, Ming Wai-Kit, Mou Zhi-Wei, Kong Qi-Hang, Li Ang, Yuan Ti-Fei, Mi Xue-Song

2020-Nov-23

computer-based image analysis, deep learning, machine learning, multiple instance learning, retinopathy of prematurity

Oncology

Lung Nodule Detection based on Faster R-CNN Framework.

In Computer methods and programs in biomedicine

BACKGROUND : Lung cancer is a worldwide high-risk disease, and lung nodules are the main manifestation of early lung cancer. Automatic detection of lung nodules reduces the workload of radiologists, the rate of misdiagnosis and missed diagnosis. For this purpose, we propose a Faster R-CNN algorithm for the detection of these lung nodules.

METHOD : Faster R-CNN algorithm can detect lung nodules, and the training set is used to prove the feasibility of this technique. In theory, parameter optimization can improve network structure, as well as detection accuracy.

RESULT : Through experiments, the best parameters are that the basic learning rate is 0.001, step size is 70,000, attenuation coefficient is 0.1, the value of Dropout is 0.5, and the value of Batch Size is 64. Compared with other networks for detecting lung nodules, the optimized and improved algorithm proposed in this paper generally improves detection accuracy by more than 20% when compared with the other traditional algorithms.

CONCLUSION : Our experimental results have proved that the method of detecting lung nodules based on Faster R-CNN algorithm has good accuracy and therefore, presents potential clinical value in lung disease diagnosis. This method can further assist radiologists, and also for researchers in the design and development of the detection system for lung nodules.

Su Ying, Li Dan, Chen Xiaodong

2020-Nov-22

CT images, Computer-aided diagnosis, Deep learning, Fast R-CNN, Lung nodules

General

A machine learning method for acute hypotensive episodes prediction using only non-invasive parameters.

In Computer methods and programs in biomedicine

BACKGROUND AND OBJECTIVES : Accurate prediction of acute hypotensive episodes (AHE) is fundamentally important for timely and appropriate clinical decision-making, as it can provide medical professionals with sufficient time to accurately select more efficient therapeutic interventions for each specific condition. However, existing methods are invasive, easily affected by artifacts and can be difficult to acquire in a pre-hospital setting.

METHODS : In this study, 1055 patients' records were extracted from the Multiparameter Intelligent Monitoring in Intensive Care II database (MIMIC II), comprising of 388 AHE records and 667 non-AHE records. Six commonly used machine learning algorithms were selected and used to develop an AHE prediction model based on features extracted from seven types of non-invasive physiological parameters.

RESULTS : The optimal observation window and prediction gap were selected as 300 minutes and 60 minutes, respectively. For GBDT, XGB and AdaBoost, the optimal feature subsets contained only 39% of the overall features. An ensemble prediction model was developed using the voting method to achieve a more robust performance with an accuracy (ACC) of 0.822 and area under the receiver operating characteristic curve (AUC) of 0.878.

CONCLUSION : A novel machine learning method that uses only noninvasive physiological parameters offers a promising solution for easy and prompt AHE prediction in widespread scenario applications, including pre-hospital and in-hospital care.

Zhang Guang, Yuan Jing, Yu Ming, Wu Taihu, Luo Xi, Chen Feng

2020-Nov-23

Acute hypotensive episodes (AHE), Data mining, Feature extraction methods, Machine learning algorithms, Non-invasive physiological parameters (NIPPs), Observation window, Prediction, Prediction gap

General

Convolutional neural network for automatically segmenting magnetic resonance images of the shoulder joint.

In Computer methods and programs in biomedicine

BACKGROUND : Magnetic resonance imaging (MRI) has been known to replace computed tomography (CT) for bone and skeletal joint examination. The accurate automatic segmentation of bone structure in shoulder MRI is important for the measurement and diagnosis of bone injury and disease. Existing bone segmentation algorithms cannot achieve automatic segmentation without any prior knowledge, and their versatility and accuracy are relatively low. Therefore, an automatic segmentation combining pulse coupled neural network (PCNN) and full convolutional neural networks (FCN) is proposed.

METHODOLOGY : By constructing the block-based AlexNet segmentation model and U-Net-based bone segmentation module, we implemented the humeral segmentation model, articular bone segmentation model, humeral head and articular bone segmentation model synthesis model. We use this four kinds of segmentation models to obtain candidate bone regions, and accurately detect the positions of humerus and articular bone by voting. Finally, we perform an AlexNet segmentation model in the detected bone area in one step to segment accuracy at the pixel level.

RESULTS : The experimental data came from 8 groups of patients in Shengjing Hospital affiliated to China Medical University. The scanning volume of each group is approximately 100 images. Five fold cross-validations and for training were recorded, and five sets of data were carefully separated. After using our technique in the three groups of patients tested, the positive predictive value of dice coefficient (PPV) and the average accuracy of sensitivity were very significant, which reached 0.96±0.02, 0.97±0.02 and 0.94±0.03, respectively.

CONCLUSION : The method used in the experiment in this paper is based on a small amount of patient sample data. The deep learning required for the experiment needs to be performed through 2D medical images. The shoulder segmentation data obtained in this way can be very accurate.

Wang Guangbin, Han Yaxin

2020-Nov-23

Convolutional neural network, Deep learning, Magnetic resonance image, Medical image examination, Orthopedic diagnosis

General

Pay-for-performance reduces bypassing of health facilities: Evidence from Tanzania.

In Social science & medicine (1982)

Many patients and expectant mothers in low-income countries bypass local health facilities in search of better-quality services. This study examines the impact of a payment-for-performance (P4P) scheme on bypassing practices among expectant women in Tanzania. We expect the P4P intervention to reduce incidences of bypassing by improving the quality of services in local health facilities, thereby reducing the incentive to migrate. We used a difference-in-difference regression model to assess the impact of P4P on bypassing after one year and after three years. In addition, we implemented a machine learning approach to identify factors that predict bypassing. Overall, 38% of women bypassed their local health service provider to deliver in another facility. Our analysis shows that the P4P scheme significantly reduced bypassing. On average, P4P reduced bypassing in the study area by 17% (8 percentage points) over three years. We also identified two main predictors of bypassing - facility type and the distance to the closest hospital. Women are more likely to bypass if their local facility is a dispensary instead of a hospital or a health center. Women are less likely to bypass if they live close to a hospital.

Bezu Sosina, Binyaruka Peter, Mæstad Ottar, Somville Vincent

2020-Nov-25

Africa, Health financing, Health governance, Health service use, Maternal care, Pay for performance, Tanzania, bypassing

General

Computational Analysis of Multidimensional Behavioral Alterations After Chronic Social Defeat Stress.

In Biological psychiatry ; h5-index 105.0

BACKGROUND : The study of depression in humans depends on animal models that attempt to mimic specific features of the human syndrome. Most studies focus on one or a few behavioral domains, with time and practical considerations prohibiting a comprehensive evaluation. Although machine learning has enabled unbiased analysis of behavior in animals, this has not yet been applied to animal models of psychiatric disease.

METHODS : We performed chronic social defeat stress (CSDS) in mice and evaluated behavior with PsychoGenics' SmartCube, a high-throughput unbiased automated phenotyping platform that collects >2000 behavioral features based on machine learning. We evaluated group differences at several times post-CSDS and after administration of the antidepressant medication imipramine.

RESULTS : SmartCube analysis after CSDS successfully separated control and defeated-susceptible mice, and defeated-resilient mice more resembled control mice. We observed a potentiation of CSDS effects over time. Treatment of susceptible mice with imipramine induced a 40.2% recovery of the defeated-susceptible phenotype as assessed by SmartCube.

CONCLUSIONS : High-throughput analysis can simultaneously evaluate multiple behavioral alterations in an animal model for the study of depression, which provides a more unbiased and holistic approach to evaluating group differences after CSDS and perhaps can be applied to other mouse models of psychiatric disease.

Lorsch Zachary S, Ambesi-Impiombato Alberto, Zenowich Rebecca, Morganstern Irene, Leahy Emer, Bansal Mukesh, Nestler Eric J, Hanania Taleen

2020-Oct-24

Antidepressants, Behavior, Bioinformatics, Chronic social defeat stress, Depression, Translational models

Oncology

Status and perspectives of biomarker validation for diagnosis, stratification, and treatment.

In Public health

OBJECTIVES : The aim of this study was to discuss the status of and perspective for biomarker validation in view of the challenges imposed on national healthcare systems due to an increasing number of citizens with chronic diseases and new expensive drugs with effects that are sometimes poorly documented. The demand for a paradigm shift toward stratification of patients or even 'personalized medicine' (PM) is rising, and the implementation of such novel strategies has the potential to increase patient outcomes and cost efficiency of treatments. The implementation of PM depends on relevant and reliable biomarkers correlated to disease states, prognosis, or effect of treatment. Beyond biomarkers of disease, personalized prevention strategies (such as individualized nutrition guidance) are likely to depend on novel biomarkers.

STUDY DESIGN : We discuss the current status of the use of biomarkers and the need for standardization and integration of biomarkers based on multi-omics approaches.

METHODS : We present representative cases from laboratory medicine, oncology, and nutrition, where present and emerging biomarkers have or may present opportunities for PM or prevention.

RESULTS : Biomarkers vary greatly in complexity, from single genomic mutations to metagenomic analyses of the composition of the gut microbiota and comprehensive analyses of metabolites, metabolomics. Using biomarkers for decision-making has previously often relied on measurements of single biomolecules. The current development now moves toward the use of multiple biomarkers requiring the use of machine learning or artificial intelligence. Still, the usefulness of biomarkers is often challenged by suboptimal validation, and the discovery of new biomarkers moves much faster than standardization efforts. To reap the potential benefits of personalization of treatment and prevention, healthcare systems and regulatory authorities need to focus on validation and standardization of biomarkers.

CONCLUSION : There is a great public health need for better understanding of the usefulness, but also limitations, of biomarkers among policy makers, clinicians, and scientists, and efforts securing effective validation are key to the future use of novel sets of complex biomarkers.

Skov J, Kristiansen K, Jespersen J, Olesen P

2020-Dec-09

Biomarker, Genomics, Metabolomics, Metagenomics, Multi-omics, Nutrigenomics, Personalized medicine, Prevention

Public Health

Racialized algorithms for kidney function: Erasing social experience.

In Social science & medicine (1982)

The rise of evidence-based medicine, medical informatics, and genomics --- together with growing enthusiasm for machine learning and other types of algorithms to standardize medical decision-making --- has lent increasing credibility to biomedical knowledge as a guide to the practice of medicine. At the same time, concern over the lack of attention to the underlying assumptions and unintended health consequences of such practices, particularly the widespread use of race-based algorithms, from the simple to the complex, has caught the attention of both physicians and social scientists. Epistemological debates over the meaning of "the social" and "the scientific" are consequential in discussions of race and racism in medicine. In this paper, we examine the socio-scientific processes by which one algorithm that "corrects" for kidney function in African Americans became central to knowledge production about chronic kidney disease (CKD). Correction factors are now used extensively and routinely in clinical laboratories and medical practices throughout the US. Drawing on close textual analysis of the biomedical literature, we use the theoretical frameworks of science and technology studies to critically analyze the initial development of the race-based algorithm, its uptake, and its normalization. We argue that race correction of kidney function is a racialized biomedical practice that contributes to the consolidation of a long-established hierarchy of difference in medicine. Consequentially, correcting for race in the assessment of kidney function masks the complexity of the lived experience of societal neglect that damages health.

Braun Lundy, Wentz Anna, Baker Reuben, Richardson Ellen, Tsai Jennifer

2020-Nov-23

Algorithms, Chronic kidney disease, Estimated glomerular filtration rate, Racialization

Radiology

Medical image segmentation and reconstruction of prostate tumor based on 3D AlexNet.

In Computer methods and programs in biomedicine

BACKGROUND : Prostate cancer is a disease with a high incidence of tumors in men. Due to the long incubation time and insidious condition, early diagnosis is difficult; especially imaging diagnosis is more difficult. In actual clinical practice, the method of manual segmentation by medical experts is mainly used, which is time-consuming and labor-intensive and relies heavily on the experience and ability of medical experts. The rapid, accurate and repeatable segmentation of the prostate area is still a challenging problem. It is important to explore the automated segmentation of prostate images based on the 3D AlexNet network.

METHOD : Taking the medical image of prostate cancer as the entry point, the three-dimensional data is introduced into the deep learning convolutional neural network. This paper proposes a 3D AlexNet method for the automatic segmentation of prostate cancer magnetic resonance images, and the general network ResNet 50, Inception -V4 compares network performance.

RESULTS : Based on the training samples of magnetic resonance images of 500 prostate cancer patients, a set of 3D AlexNet with simple structure and excellent performance was established through adaptive improvement on the basis of classic AlexNet. The accuracy rate was as high as 0.921, the specificity was 0.896, and the sensitivity It is 0.902 and the area under the receiver operating characteristic curve (AUC) is 0.964. The Mean Absolute Distance (MAD) between the segmentation result and the medical expert's gold standard is 0.356 mm, and the Hausdorff distance (HD) is 1.024 mm, the Dice similarity coefficient is 0.9768.

CONCLUSION : The improved 3D AlexNet can automatically complete the structured segmentation of prostate magnetic resonance images. Compared with traditional segmentation methods and depth segmentation methods, the performance of the 3D AlexNet network is superior in terms of training time and parameter amount, or network performance evaluation. Compared with the algorithm, it proves the effectiveness of this method.

Chen Jun, Wan Zhechao, Zhang Jiacheng, Li Wenhua, Chen Yanbing, Li Yuebing, Duan Yue

2020-Nov-27

3D AlexNet, Convolutional Neural Network, Prostate Cancer, Three-dimensional reconstruction

Public Health

Combining Data, Machine Learning, and Visual Analytics to Improve Detection of Disease Re-emergence: The Re-emerging Disease Alert Tool.

In JMIR public health and surveillance

BACKGROUND : Currently, the identification of infectious disease re-emergence is performed without describing specific quantitative criteria that can be used to identify re-emergence events consistently. This practice may lead to ineffective mitigation. In addition, identification of factors contributing to local disease re-emergence and assessment of global disease re-emergence require access to data about disease incidence and a large number of factors at the local level for the entire world. This paper presents Re-emerging Disease Alert (RED Alert), a web-based tool designed to help public health officials detect and understand infectious disease re-emergence.

OBJECTIVE : Our objective is to bring together variety of disease-related data and analytics needed to help public health analysts answer following three primary questions for detecting and understanding disease re-emergence: Is there a potential disease re-emergence at the local (country) level? What are the potential contributing factors for this re-emergence? Is there a potential for global re-emergence?

METHODS : We collected and cleaned disease-related data (e.g., case counts, vaccination rates, and indicators related to disease transmission) from several data sources including the WHO, PAHO, World Bank, and Gideon. We combined these data with machine learning and visual analytics into a tool called RED Alert to detect re-emergence for following four diseases: measles, cholera, dengue, and yellow fever. We evaluated the performance of the machine learning models for re-emergence detection and reviewed the output of the tool through a number of case studies.

RESULTS : Our supervised learning models were able to identify 82-90% of the local re-emergence events, although with 18-31% (except 46% for dengue) false positives. This is consistent with our goal of identifying all possible re-emergences while allowing some false positives. The review of the web-based tool through case studies showed that local re-emergence detection was possible, the tool provided actionable information about potential factors contributing to the local disease re-emergence, and trends in global disease re-emergence.

CONCLUSIONS : To the best of our knowledge, this is the first tool that focuses specifically on disease re-emergence and addresses the important challenges mentioned above.

CLINICALTRIAL :

Parikh Nidhi Kiranbhai, Daughton Ashlynn Rae, Rosenberger William Earl, Aberle Derek Jacob, Chitanvis Maneesha Elizabeth, Altherr Forest Michael, Velappan Nileena, Fairchild Geoffrey, Deshpande Alina

2020-Dec-14

General

State bounding for fuzzy memristive neural networks with bounded input disturbances.

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

This paper investigates the state bounding problem of fuzzy memristive neural networks (FMNNs) with bounded input disturbances. By using the characters of Metzler, Hurwitz and nonnegative matrices, this paper obtains the exact delay-independent and delay-dependent boundary ranges of the solution, which have less conservatism than the results in existing literatures. The validity of the results is verified by two numerical examples.

Gao Yu, Zhu Song, Yang Chunyu, Wen Shiping

2020-Dec-03

Bounded disturbances, Fuzzy systems, Memristor, Neural networks, State bounding

Public Health

Evaluating the impact of mobility on COVID-19 pandemic with machine learning hybrid predictions.

In The Science of the total environment

COVID-19 pandemic had expanded to the US since early 2020 and has caused nationwide economic loss and public health crisis. Until now, although the US has the most confirmed cases in the world and are still experiencing an increasing pandemic, several states insisted to re-open business activities and colleges while announced strict control measures. To provide a quantitative reference for official strategies, predicting the near future trend based on finer spatial resolution data and presumed scenarios are urgently needed. In this study, the first attempted COVID-19 case predicting model based on county-level demographic, environmental, and mobility data was constructed with multiple machine learning techniques and a hybrid framework. Different scenarios were also applied to selected metropolitan counties including New York City, Cook County in Illinois, Los Angeles County in California, and Miami-Dade County in Florida to assess the impact from lockdown, Phase I, and Phase III re-opening. Our results showed that, for selected counties, the mobility decreased substantially after the lockdown but kept increasing with an apparent weekly pattern, and the weekly pattern of mobility and infections implied high infections during the weekend. Meanwhile, our model was successfully built up, and the scenario assessment results indicated that, compared with Phase I re-opening, a 1-week and a 2-week lockdown could reduce 4%-29% and 15%-55% infections, respectively, in the future week, while 2-week Phase III re-opening could increase 16%-80% infections. We concluded that the mandatory orders in metropolitan counties such lockdown should last longer than one week, the effect could be observed. The impact of lockdown or re-opening was also county-dependent and varied with the local pandemic. In future works, we expect to involve a longer period of data, consider more county-dependent factors, and employ more sophisticated techniques to decrease the modeling uncertainty and apply it to counties nationally and other countries.

Kuo Cheng-Pin, Fu Joshua S

2020-Nov-28

County-level, Forecasting, Lockdown, Pandemic, Re-opening

General

Development of an absolute assignment predictor for triple-negative breast cancer subtyping using machine learning approaches.

In Computers in biology and medicine

Triple-negative breast cancer (TNBC) heterogeneity represents one of the main obstacles to precision medicine for this disease. Recent concordant transcriptomics studies have shown that TNBC could be divided into at least three subtypes with potential therapeutic implications. Although a few studies have been conducted to predict TNBC subtype using transcriptomics data, the subtyping was partially sensitive and limited by batch effect and dependence on a given dataset, which may penalize the switch to routine diagnostic testing. Therefore, we sought to build an absolute predictor (i.e., intra-patient diagnosis) based on machine learning algorithms with a limited number of probes. To that end, we started by introducing probe binary comparison for each patient (indicators). We based the predictive analysis on this transformed data. Probe selection was first involved combining both filter and wrapper methods for variable selection using cross-validation. We tested three prediction models (random forest, gradient boosting [GB], and extreme gradient boosting) using this optimal subset of indicators as inputs. Nested cross-validation consistently allowed us to choose the best model. The results showed that the fifty selected indicators highlighted the biological characteristics associated with each TNBC subtype. The GB based on this subset of indicators performs better than other models.

Ben Azzouz Fadoua, Michel Bertrand, Lasla Hamza, Gouraud Wilfried, François Anne-Flore, Girka Fabien, Lecointre Théo, Guérin-Charbonnel Catherine, Juin Philippe P, Campone Mario, Jézéquel Pascal

2020-Dec-09

Machine learning, Prediction models, TNBC subtype, Transcriptomics data, Variable selection

General

Enabling Personalized Decision Support with Patient-Generated Data and Attributable Components.

In Journal of biomedical informatics ; h5-index 55.0

Decision-making related to health is complex. Machine learning (ML) and patient generated data can identify patterns and insights at the individual level, where human cognition falls short, but not all ML-generated information is of equal utility for making health-related decisions. We develop and apply attributable components analysis (ACA), a method inspired by optimal transport theory, to type 2 diabetes self-monitoring data to identify patterns of association between nutrition and blood glucose control. In comparison with linear regression, we found that ACA offers a number of characteristics that make it promising for use in decision support applications. For example, ACA was able to identify non-linear relationships, was more robust to outliers, and offered broader and more expressive uncertainty estimates. In addition, our results highlight a tradeoff between model accuracy and interpretability, and we discuss implications for ML-driven decision support systems.

Mitchell Elliot G, Tabak Esteban G, Levine Matthew E, Mamykina Lena, Albers David J

2020-Dec-11

General

Challenges and Solutions to Employing Natural Language Processing and Machine Learning to Measure Patients' Health Literacy and Physician Writing Complexity: The ECLIPPSE Study.

In Journal of biomedical informatics ; h5-index 55.0

OBJECTIVE : In the National Library of Medicine funded ECLIPPSE Project (Employing Computational Linguistics to Improve Patient-Provider Secure Emails exchange), we attempted to create novel, valid, and scalable measures of both patients' health literacy (HL) and physicians' linguistic complexity by employing natural language processing (NLP) techniques and machine learning (ML). We applied these techniques to >400,000 patients' and physicians' secure messages (SMs) exchanged via an electronic patient portal, developing and validating an automated patient literacy profile (LP) and physician complexity profile (CP). Herein, we describe the challenges faced and the solutions implemented during this innovative endeavor.

MATERIALS AND METHODS : To describe challenges and solutions, we used two data sources: study documents and interviews with study investigators. Over the five years of the project, the team tracked their research process using a combination of Google Docs tools and an online team organization, tracking, and management tool (Asana). In year 5, the team convened a number of times to discuss, categorize, and code primary challenges and solutions.

RESULTS : We identified 23 challenges and associated approaches that emerged from three overarching process domains: (1) Data Mining related to the SM corpus; (2) Analyses using NLP indices on the SM corpus; and (3) Interdisciplinary Collaboration. With respect to Data Mining, problems included cleaning SMs to enable analyses, removing hidden caregiver proxies (e.g., other family members) and Spanish language SMs, and culling SMs to ensure that only patients' primary care physicians were included. With respect to Analyses, critical decisions needed to be made as to which computational linguistic indices and ML approaches should be selected; how to enable the NLP-based linguistic indices tools to run smoothly and to extract meaningful data from a large corpus of medical text; and how to best assess content and predictive validities of both the LP and the CP. With respect to the Interdisciplinary Collaboration, because the research required engagement between clinicians, health services researchers, biomedical informaticians, linguists, and cognitive scientists, continual effort was needed to identify and reconcile differences in scientific terminologies and resolve confusion; arrive at common understanding of tasks that needed to be completed and priorities therein; reach compromises regarding what represents "meaningful findings" in health services vs. cognitive science research; and address constraints regarding potential transportability of the final LP and CP to different health care settings.

DISCUSSION : Our study represents a process evaluation of an innovative research initiative to harness "big linguistic data" to estimate patient HL and physician linguistic complexity. Any of the challenges we identified, if left unaddressed, would have either rendered impossible the effort to generate LPs and CPs, or invalidated analytic results related to the LPs and CPs. Investigators undertaking similar research in HL or using computational linguistic methods to assess patient-clinician exchange will face similar challenges and may find our solutions helpful when designing and executing their health communications research.

Brown William, Balyan Renu, Karter Andrew J, Crossley Scott, Semere Wagahta, Duran Nicholas D, Lyles Courtney, Liu Jennifer, Moffet Howard H, Daniels Ryane, McNamara Danielle S, Schillinger Dean

2020-Dec-11

diabetes health care quality, digital health and health services research, electronic health records, health literacy, machine learning, natural language processing

Surgery

Neural endophenotypes and predictors of laryngeal dystonia penetrance and manifestation.

In Neurobiology of disease

Focal dystonias are the most common forms of isolated dystonia; however, the etiopathophysiological signatures of disorder penetrance and clinical manifestation remain unclear. Using an imaging genetics approach, we investigated functional and structural representations of neural endophenotypes underlying the penetrance and manifestation of laryngeal dystonia in families, including 21 probands and 21 unaffected relatives, compared to 32 unrelated healthy controls. We further used a supervised machine-learning algorithm to predict the risk for dystonia development in susceptible individuals based on neural features of identified endophenotypes. We found that abnormalities in the prefrontal-parietal cortex, thalamus, and caudate nucleus were commonly shared between patients and their unaffected relatives, representing an intermediate endophenotype of laryngeal dystonia. Machine-learning classification of 95.2% of unaffected relatives together with patients rather than healthy controls substantiated these neural alterations as the endophenotypic marker of dystonia penetrance, independent of its symptomatology. Additional abnormalities in premotor-parietal-temporal cortical regions, caudate nucleus, and cerebellum were present only in patients but not their unaffected relatives, likely representing a secondary endophenotype of dystonia manifestation. Based on alterations in the parietal cortex and caudate nucleus, the machine-learning categorization of 28.6% of unaffected relative as patients indicated their increased lifetime risk for developing clinical manifestation of dystonia. The identified endophenotypic neural markers may be implemented for screening of at-risk individuals for dystonia development, selection of families for genetic studies of novel variants based on their risk for disease penetrance, or stratification of patients who would respond differently to a particular treatment in clinical trials.

Khosravani Sanaz, Chen Gang, Ozelius Laurie J, Simonyan Kristina

2020-Dec-11

Brain imaging, Dystonia, Endophenotypes

Radiology

Consistency of Independent Component Analysis for FMRI.

In Journal of neuroscience methods

BACKGROUND : Independent component analysis (ICA) has been widely used for blind source separation in the field of medical imaging. However, despite of previous substantial efforts, the stability of ICA components remains a critical issue which has not been adequately addressed, despite numerous previous efforts. Most critical is the inconsistency of some of the extracted components when ICA is run with different model orders (MOs).

NEW METHOD : In this study, a novel method of determining the consistency of component analysis (CoCA) is proposed to evaluate the consistency of extracted components with different model orders. In the method, "consistent components" (CCs) are defined as those which can be extracted repeatably over a range of model orders.

RESULT : The efficacy of the method was evaluated with simulation data and fMRI datasets. With our method, the simulation result showed a clear difference of consistency between ground truths and noise.

COMPARISON WITH EXISTING METHODS : The information criteria were implemented to provide suggestions for the optimal model order, where some of the ICs were revealed inconsistent in our proposed method.

CONCLUSIONS : This method provided an objective protocol for choosing CCs of an ICA decomposition of a data matrix, independent of model order. This is especially useful with high model orders, where noise or other disturbances could possibly lead to an instability of the components.

Zhao Wei, Li Huanjie, Hu Guoqiang, Hao Yuxing, Zhang Qing, Wu Jianlin, Frederick Blaise B, Cong Fengyu

2020-Dec-11

ICA, consistency, fMRI, model order

General

Quantitative Analysis of Brain Herniation from Non-Contrast CT Images using Deep Learning.

In Journal of neuroscience methods

BACKGROUND : Brain herniation is one of the fatal outcomes of increased intracranial pressure (ICP). It is caused due to the presence of hematoma or tumor mass in the brain. Ideal midline (iML) divides the healthy brain into two (right and left) nearly equal hemispheres. In the presence of hematoma, the midline tends to shift from its original position to the contralateral side of the mass and thus develops a deformed midline (dML).

NEW METHOD : In this study, a convolutional neural network (CNN) was used to predict the deformed left and right hemispheres. The proposed algorithm was validated with non-contrast computed tomography (NCCT) of (n = 45) subjects with two types of brain hemorrhages - epidural hemorrhage (EDH): (n = 5) and intra-parenchymal hemorrhage (IPH): (n = 40)).

RESULTS : The method demonstrated excellent potential in automatically predicting MLS with the average errors of 1.29 mm by location, 66.4 mm² by 2D area, and 253.73 mm³ by 3D volume. Estimated MLS could be well correlated with other clinical markers including hematoma volume - R² = 0.86 (EDH); 0.48 (IPH) and a Radiologist-defined severity score (RSS) - R² = 0.62 (EDH); 0.57 (IPH). RSS was found to be even better correlated (R² = 0.98 (EDH); 0.70 (IPH)), hence better predictable by a joint correlation between hematoma volume, midline pixel- or voxel-shift, and minimum distance of (ideal or deformed) midline from the hematoma (boundary or centroid).

CONCLUSION : All these predictors were computed automatically, which highlighted the excellent clinical potential of the proposed automated method in midline shift (MLS) estimation and severity prediction in hematoma decision support systems.

Nag Manas Kumar, Gupta Akshat, Hariharasudhan A S, Sadhu Anup Kumar, Das Abir, Ghosh Nirmalya

2020-Dec-11

NCCT, convolutional neural network, deformed midline, midline shift

General

Corticosteroid therapy is associated with improved outcome in critically ill COVID-19 patients with hyperinflammatory phenotype.

In Chest ; h5-index 81.0

BACKGROUND : Corticosteroid therapy is commonly used in patients with coronavirus disease 2019 (COVID-19), while its impact on outcomes and which patients could benefit from corticosteroid therapy are uncertain.

RESEARCH QUESTION : Whether clinical phenotypes of COVID-19 were associated with differential response to corticosteroid therapy.

STUDY DESIGN AND METHODS : Critically ill patients with COVID-19 from Tongji hospital between Jan 2020 and Feb 2020 were included, and the main exposure of interest was the administration of intravenous corticosteroids. The primary outcome was 28-day mortality. Marginal structural modeling was used to account for baseline and time-dependent confounders. An unsupervised machine learning approach was carried out to identify phenotypes of COVID-19.

RESULTS : A total of 428 patients were included, and 280/428 (65.4%) patients received corticosteroid therapy. The 28-day mortality was significantly higher in patients who received corticosteroid therapy than in those who did not (53.9% vs. 19.6%; p<0.0001). After marginal structural modeling, corticosteroid therapy was not significantly associated with 28-day mortality (HR 0.80, 95% CI 0.54-1.18; p=0.26). Our analysis identified two phenotypes of COVID-19, and compared to the hypoinflammatory phenotype, the hyperinflammatory phenotype was characterized by elevated levels of proinflammatory cytokines, higher SOFA scores and higher rates of complications. Corticosteroid therapy was associated with a reduced 28-day mortality (HR 0.45; 95% CI 0.25-0.80; p=0.0062) in patients with hyperinflammatory phenotype.

INTERPRETATION : For critically ill patients with COVID-19, corticosteroid therapy was not associated with 28-day mortality, but the use of corticosteroids showed significant survival benefits in patients with the hyperinflammatory phenotype.

Chen Hui, Xie Jianfeng, Su Nan, Wang Jun, Sun Qin, Li Shusheng, Jin Jun, Zhou Jing, Mo Min, Wei Yao, Chao Yali, Hu Weiwei, Du Bin, Qiu Haibo

2020-Dec-11

COVID-19, Corticosteroid, Phenotype

Public Health

Aging and Engaging: A Pilot Randomized Controlled Trial of an Online Conversational Skills Coach for Older Adults.

In The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry

OBJECTIVE : Communication difficulties negatively impact relationship quality and are associated with social isolation and loneliness in later life. There is a need for accessible communication interventions offered outside specialty mental health settings.

DESIGN : Pilot randomized controlled trial.

SETTING : Assessments in the laboratory and intervention completed in-home.

PARTICIPANTS : Twenty adults age 60 and older from the community and a geriatric psychiatry clinic.

INTERVENTION : A web-based communication coach that provides automated feedback on eye contact, facial expressivity, speaking volume, and negative content (Aging and Engaging Program, AEP), delivered with minimal assistance in the home (eight brief sessions over 4-6 weeks) or control (education and videos on communication).

MEASUREMENTS : System Usability Scale and Social Skills Performance Assessment, an observer-rated assessment of social communication elicited through standardized role-plays.

RESULTS : Ninety percent of participants completed all AEP sessions and the System Usability Scale score of 68 was above the cut-off for acceptable usability. Participants randomized to AEP demonstrated statistically and clinically significant improvement in eye contact and facial expressivity.

CONCLUSION : The AEP is acceptable and feasible for older adults with communication difficulties to complete at home and may improve eye contact and facial expressivity, warranting a larger RCT to confirm efficacy and explore potential applications to other populations, including individuals with autism and social anxiety.

Ali Rafayet, Hoque Ehsan, Duberstein Paul, Schubert Lenhart, Razavi Seyedeh Zahra, Kane Benjamin, Silva Caroline, Daks Jennifer S, Huang Meghan, Van Orden Kim

2020-Nov-22

Social communication, artificial intelligence, intervention, social isolation, technology

General

Machine learning for predictive data analytics in medicine: A review illustrated by cardiovascular and nuclear medicine examples.

In Clinical physiology and functional imaging

The evidence-based medicine allows the physician to evaluate the risk-benefit ratio of a treatment through setting and data. Risk-based choices can be done by the doctor using different information. With the emergence of new technologies, a large amount of data is recorded offering interesting perspectives with machine learning for predictive data analytics. Machine learning is an ensemble of methods that process data to model a learning problem. Supervised machine learning algorithms consist in using annotated data to construct the model. This category allows to solve prediction data analytics problems. In this paper, we detail the use of supervised machine learning algorithms for predictive data analytics problems in medicine. In the medical field, data can be split into two categories: medical images and other data. For brevity, our review deals with any kind of medical data excluding images. In this article, we offer a discussion around four supervised machine learning approaches: information-based, similarity-based, probability-based, and error-based approaches. Each method is illustrated with detailed cardiovascular and nuclear medicine examples. Our review shows that model ensemble (ME) and support vector machine (SVM) methods are the most popular. SVM, ME, and artificial neural networks often lead to better results than those given by other algorithms. In the coming years, more studies, more data, more tools, and more methods will, for sure, be proposed.

Jamin Antoine, Abraham Pierre, Humeau-Heurtier Anne

2020-Dec-14

error-based approach, information-based approach, prediction, probability-based approach, similarity-based approach, supervised algorithm

Public Health

Primary care electronic medical records can be used to predict risk and identify potentially modifiable factors for early and late death in adult onset epilepsy.

In Epilepsia

OBJECTIVE : To use clinically informed machine learning to derive prediction models for early and late premature death in epilepsy.

METHODS : This was a population-based primary care observational cohort study. All patients meeting a case definition for incident epilepsy in the Health Improvement Network database for inclusive years 2000-2012 were included. A modified Delphi process identified 30 potential risk factors. Outcome was early (within 4 years of epilepsy diagnosis) and late (4 years or more from diagnosis) mortality. We used regularized logistic regression, support vector machines, Gaussian naive Bayes, and random forest classifiers to predict outcomes. We assessed model calibration, discrimination, and generalizability using the Brier score, mean area under the receiver operating characteristic curve (AUC) derived from stratified fivefold cross-validation, plotted calibration curves, and extracted measures of association where possible.

RESULTS : We identified 10 499 presumed incident cases from 11 194 182 patients. All models performed comparably well following stratified fivefold cross-validation, with AUCs ranging from 0.73 to 0.81 and from 0.71 to 0.79 for early and late death, respectively. In addition to comorbid disease, social habits (alcoholism odds ratio [OR] for early death = 1.54, 95% confidence interval [CI] = 1.12-2.11 and OR for late death = 2.62, 95% CI = 1.66-4.16) and treatment patterns (OR for early death when no antiseizure medication [ASM] was prescribed at baseline = 1.33, 95% CI = 1.07-1.64 and OR for late death after receipt of enzyme-inducing ASM at baseline = 1.32, 95% CI = 1.04-1.66) were significantly associated with increased risk of premature death. Baseline ASM polytherapy (OR = 0.55, 95% CI = 0.36-0.85) was associated with reduced risk of early death.

SIGNIFICANCE : Clinically informed models using routine electronic medical records can be used to predict early and late mortality in epilepsy, with moderate to high accuracy and evidence of generalizability. Medical, social, and treatment-related risk factors, such as delayed ASM prescription and baseline prescription of enzyme-inducing ASMs, were important predictors.

Hrabok Marianne, Engbers Jordan D T, Wiebe Samuel, Sajobi Tolulope T, Subota Ann, Almohawes Amal, Federico Paolo, Hanson Alexandra, Klein Karl Martin, Peedicail Joseph, Pillay Neelan, Singh Shaily, Josephson Colin B

2020-Dec-14

cohort study, epilepsy, machine learning, modified Delphi, prediction, premature death

General

Feasibility Study on Automatic Interpretation of Radiation Dose Using Deep Learning Technique for Dicentric Chromosome Assay.

In Radiation research ; h5-index 33.0

The interpretation of radiation dose is an important procedure for both radiological operators and persons who are exposed to background or artificial radiations. Dicentric chromosome assay (DCA) is one of the representative methods of dose estimation that discriminates the aberration in chromosomes modified by radiation. Despite the DCA-based automated radiation dose estimation methods proposed in previous studies, there are still limitations to the accuracy of dose estimation. In this study, a DCA-based automated dose estimation system using deep learning methods is proposed. The system is comprised of three stages. In the first stage, a classifier based on a deep learning technique is used for filtering the chromosome images that are not appropriate for use in distinguishing the chromosome; 99% filtering accuracy was achieved with 2,040 test images. In the second stage, the dicentric rate is evaluated by counting and identifying chromosomes based on the Feature Pyramid Network, which is one of the object detection algorithms based on deep learning architecture. The accuracies of the neural networks for counting and identifying chromosomes were estimated at over 97% and 90%, respectively. In the third stage, dose estimation is conducted using the dicentric rate and the dose-response curve. The accuracies of the system were estimated using two independent samples; absorbed doses ranging from 1- 4 Gy agreed well within a 99% confidential interval showing highest accuracy compared to those in previous studies. The goal of this study was to provide insights towards achieving complete automation of the radiation dose estimation, especially in the event of a large-scale radiation exposure incident.

Jang Seungsoo, Shin Sung-Gyun, Lee Min-Jae, Han Sangsoo, Choi Chan-Ho, Kim Sungkyum, Cho Woo-Sung, Kim Song-Hyun, Kang Yeong-Rok, Jo Wolsoon, Jeong Sookyung, Oh Sujung

2020-Dec-14

General

Large-scale comparative review and assessment of computational methods for anti-cancer peptide identification.

In Briefings in bioinformatics

Anti-cancer peptides (ACPs) are known as potential therapeutics for cancer. Due to their unique ability to target cancer cells without affecting healthy cells directly, they have been extensively studied. Many peptide-based drugs are currently evaluated in the preclinical and clinical trials. Accurate identification of ACPs has received considerable attention in recent years; as such, a number of machine learning-based methods for in silico identification of ACPs have been developed. These methods promote the research on the mechanism of ACPs therapeutics against cancer to some extent. There is a vast difference in these methods in terms of their training/testing datasets, machine learning algorithms, feature encoding schemes, feature selection methods and evaluation strategies used. Therefore, it is desirable to summarize the advantages and disadvantages of the existing methods, provide useful insights and suggestions for the development and improvement of novel computational tools to characterize and identify ACPs. With this in mind, we firstly comprehensively investigate 16 state-of-the-art predictors for ACPs in terms of their core algorithms, feature encoding schemes, performance evaluation metrics and webserver/software usability. Then, comprehensive performance assessment is conducted to evaluate the robustness and scalability of the existing predictors using a well-prepared benchmark dataset. We provide potential strategies for the model performance improvement. Moreover, we propose a novel ensemble learning framework, termed ACPredStackL, for the accurate identification of ACPs. ACPredStackL is developed based on the stacking ensemble strategy combined with SVM, Naïve Bayesian, lightGBM and KNN. Empirical benchmarking experiments against the state-of-the-art methods demonstrate that ACPredStackL achieves a comparative performance for predicting ACPs. The webserver and source code of ACPredStackL is freely available at http://bigdata.biocie.cn/ACPredStackL/ and https://github.com/liangxiaoq/ACPredStackL, respectively.

Liang Xiao, Li Fuyi, Chen Jinxiang, Li Junlong, Wu Hao, Li Shuqin, Song Jiangning, Liu Quanzhong

2020-Dec-15

anti-cancer peptides, bioinformatics, ensemble learning, performance assessment, prediction, sequence analysis

General

Deep learning for brain disorders: from data processing to disease treatment.

In Briefings in bioinformatics

In order to reach precision medicine and improve patients' quality of life, machine learning is increasingly used in medicine. Brain disorders are often complex and heterogeneous, and several modalities such as demographic, clinical, imaging, genetics and environmental data have been studied to improve their understanding. Deep learning, a subpart of machine learning, provides complex algorithms that can learn from such various data. It has become state of the art in numerous fields, including computer vision and natural language processing, and is also growingly applied in medicine. In this article, we review the use of deep learning for brain disorders. More specifically, we identify the main applications, the concerned disorders and the types of architectures and data used. Finally, we provide guidelines to bridge the gap between research studies and clinical routine.

Burgos Ninon, Bottani Simona, Faouzi Johann, Thibeau-Sutre Elina, Colliot Olivier

2020-Dec-15

deep learning, genomics, medical imaging, neurology

General

Stochastic representation decision theory: How probabilities and values are entangled dual characteristics in cognitive processes.

In PloS one ; h5-index 176.0

Humans are notoriously bad at understanding probabilities, exhibiting a host of biases and distortions that are context dependent. This has serious consequences on how we assess risks and make decisions. Several theories have been developed to replace the normative rational expectation theory at the foundation of economics. These approaches essentially assume that (subjective) probabilities weight multiplicatively the utilities of the alternatives offered to the decision maker, although evidence suggest that probability weights and utilities are often not separable in the mind of the decision maker. In this context, we introduce a simple and efficient framework on how to describe the inherently probabilistic human decision-making process, based on a representation of the deliberation activity leading to a choice through stochastic processes, the simplest of which is a random walk. Our model leads naturally to the hypothesis that probabilities and utilities are entangled dual characteristics of the real human decision making process. It predicts the famous fourfold pattern of risk preferences. Through the analysis of choice probabilities, it is possible to identify two previously postulated features of prospect theory: the inverse S-shaped subjective probability as a function of the objective probability and risk-seeking behavior in the loss domain. It also predicts observed violations of stochastic dominance, while it does not when the dominance is "evident". Extending the model to account for human finite deliberation time and the effect of time pressure on choice, it provides other sound predictions: inverse relation between choice probability and response time, preference reversal with time pressure, and an inverse double-S-shaped probability weighting function. Our theory, which offers many more predictions for future tests, has strong implications for psychology, economics and artificial intelligence.

Ferro Giuseppe M, Sornette Didier

2020

General

A mathematical model of local and global attention in natural scene viewing.

In PLoS computational biology

Understanding the decision process underlying gaze control is an important question in cognitive neuroscience with applications in diverse fields ranging from psychology to computer vision. The decision for choosing an upcoming saccade target can be framed as a selection process between two states: Should the observer further inspect the information near the current gaze position (local attention) or continue with exploration of other patches of the given scene (global attention)? Here we propose and investigate a mathematical model motivated by switching between these two attentional states during scene viewing. The model is derived from a minimal set of assumptions that generates realistic eye movement behavior. We implemented a Bayesian approach for model parameter inference based on the model's likelihood function. In order to simplify the inference, we applied data augmentation methods that allowed the use of conjugate priors and the construction of an efficient Gibbs sampler. This approach turned out to be numerically efficient and permitted fitting interindividual differences in saccade statistics. Thus, the main contribution of our modeling approach is two-fold; first, we propose a new model for saccade generation in scene viewing. Second, we demonstrate the use of novel methods from Bayesian inference in the field of scan path modeling.

Malem-Shinitski Noa, Opper Manfred, Reich Sebastian, Schwetlick Lisa, Seelig Stefan A, Engbert Ralf

2020-Dec-14

General

Mapping molar shapes on signaling pathways.

In PLoS computational biology

A major challenge in evolutionary developmental biology is to understand how genetic mutations underlie phenotypic changes. In principle, selective pressures on the phenotype screen the gene pool of the population. Teeth are an excellent model for understanding evolutionary changes in the genotype-phenotype relationship since they exist throughout vertebrates. Genetically modified mice (mutants) with abnormalities in teeth have been used to explore tooth development. The relationship between signaling pathways and molar shape, however, remains elusive due to the high intrinsic complexity of tooth crowns. This hampers our understanding of the extent to which developmental factors explored in mutants explain developmental and phenotypic variation in natural species that represent the consequence of natural selection. Here we combine a novel morphometric method with two kinds of data mining techniques to extract data sets from the three-dimensional surface models of lower first molars: i) machine learning to maximize classification accuracy of 22 mutants, and ii) phylogenetic signal for 31 Murinae species. Major shape variation among mutants is explained by the number of cusps and cusp distribution on a tooth crown. The distribution of mutant mice in morphospace suggests a nonlinear relationship between the signaling pathways and molar shape variation. Comparative analysis of mutants and wild murines reveals that mutant variation overlaps naturally occurring diversity, including more ancestral and derived morphologies. However, taxa with transverse lophs are not fully covered by mutant variation, suggesting experimentally unexplored developmental factors in the evolutionary radiation of Murines.

Morita Wataru, Morimoto Naoki, Jernvall Jukka

2020-Dec

Surgery

Artificial Neural Networks Predict 30-Day Mortality After Hip Fracture: Insights From Machine Learning.

In The Journal of the American Academy of Orthopaedic Surgeons

OBJECTIVES : Accurately stratifying patients in the preoperative period according to mortality risk informs treatment considerations and guides adjustments to bundled reimbursements. We developed and compared three machine learning models to determine which best predicts 30-day mortality after hip fracture.

METHODS : The 2016 to 2017 National Surgical Quality Improvement Program for hip fracture (AO/OTA 31-A-B-C) procedure-targeted data were analyzed. Three models-artificial neural network, naive Bayes, and logistic regression-were trained and tested using independent variables selected via backward variable selection. The data were split into 80% training and 20% test sets. Predictive accuracy between models was evaluated using area under the curve receiver operating characteristics. Odds ratios were determined using multivariate logistic regression with P < 0.05 for significance.

RESULTS : The study cohort included 19,835 patients (69.3% women). The 30-day mortality rate was 5.3%. In total, 47 independent patient variables were identified to train the testing models. Area under the curve receiver operating characteristics for 30-day mortality was highest for artificial neural network (0.92), followed by the logistic regression (0.87) and naive Bayes models (0.83).

DISCUSSION : Machine learning is an emerging approach to develop accurate risk calculators that account for the weighted interactions between variables. In this study, we developed and tested a neural network model that was highly accurate for predicting 30-day mortality after hip fracture. This was superior to the naive Bayes and logistic regression models. The role of machine learning models to predict orthopaedic outcomes merits further development and prospective validation but shows strong promise for positively impacting patient care.

DeBaun Malcolm R, Chavez Gustavo, Fithian Andrew, Oladeji Kingsley, Van Rysselberghe Noelle, Goodnough L Henry, Bishop Julius A, Gardner Michael J

2020-Dec-10

General

Correction: Identifying and Ranking Common COVID-19 Symptoms From Tweets in Arabic: Content Analysis.

In Journal of medical Internet research ; h5-index 88.0

[This corrects the article DOI: 10.2196/21329.].

Alanazi Eisa, Alashaikh Abdulaziz, Alqurashi Sarah, Alanazi Aued

2020-Dec-14

Pathology

Accurate Cell Segmentation in Digital Pathology Images via Attention Enforced Networks

ArXiv Preprint

Automatic cell segmentation is an essential step in the pipeline of computer-aided diagnosis (CAD), such as the detection and grading of breast cancer. Accurate segmentation of cells can not only assist the pathologists to make a more precise diagnosis, but also save much time and labor. However, this task suffers from stain variation, cell inhomogeneous intensities, background clutters and cells from different tissues. To address these issues, we propose an Attention Enforced Network (AENet), which is built on spatial attention module and channel attention module, to integrate local features with global dependencies and weight effective channels adaptively. Besides, we introduce a feature fusion branch to bridge high-level and low-level features. Finally, the marker controlled watershed algorithm is applied to post-process the predicted segmentation maps for reducing the fragmented regions. In the test stage, we present an individual color normalization method to deal with the stain variation problem. We evaluate this model on the MoNuSeg dataset. The quantitative comparisons against several prior methods demonstrate the superiority of our approach.

Muyi Sun, Zeyi Yao, Guanhong Zhang

2020-12-14

General

FedHome: Cloud-Edge based Personalized Federated Learning for In-Home Health Monitoring

ArXiv Preprint

In-home health monitoring has attracted great attention for the ageing population worldwide. With the abundant user health data accessed by Internet of Things (IoT) devices and recent development in machine learning, smart healthcare has seen many successful stories. However, existing approaches for in-home health monitoring do not pay sufficient attention to user data privacy and thus are far from being ready for large-scale practical deployment. In this paper, we propose FedHome, a novel cloud-edge based federated learning framework for in-home health monitoring, which learns a shared global model in the cloud from multiple homes at the network edges and achieves data privacy protection by keeping user data locally. To cope with the imbalanced and non-IID distribution inherent in user's monitoring data, we design a generative convolutional autoencoder (GCAE), which aims to achieve accurate and personalized health monitoring by refining the model with a generated class-balanced dataset from user's personal data. Besides, GCAE is lightweight to transfer between the cloud and edges, which is useful to reduce the communication cost of federated learning in FedHome. Extensive experiments based on realistic human activity recognition data traces corroborate that FedHome significantly outperforms existing widely-adopted methods.

Qiong Wu, Xu Chen, Zhi Zhou, Junshan Zhang

2020-12-14

General

Towards Accurate Spatiotemporal COVID-19 Risk Scores using High Resolution Real-World Mobility Data

ArXiv Preprint

As countries look towards re-opening of economic activities amidst the ongoing COVID-19 pandemic, ensuring public health has been challenging. While contact tracing only aims to track past activities of infected users, one path to safe reopening is to develop reliable spatiotemporal risk scores to indicate the propensity of the disease. Existing works which aim to develop risk scores either rely on compartmental model-based reproduction numbers (which assume uniform population mixing) or develop coarse-grain spatial scores based on reproduction number (R0) and macro-level density-based mobility statistics. Instead, in this paper, we develop a Hawkes process-based technique to assign relatively fine-grain spatial and temporal risk scores by leveraging high-resolution mobility data based on cell-phone originated location signals. While COVID-19 risk scores also depend on a number of factors specific to an individual, including demography and existing medical conditions, the primary mode of disease transmission is via physical proximity and contact. Therefore, we focus on developing risk scores based on location density and mobility behaviour. We demonstrate the efficacy of the developed risk scores via simulation based on real-world mobility data. Our results show that fine-grain spatiotemporal risk scores based on high-resolution mobility data can provide useful insights and facilitate safe re-opening.

Sirisha Rambhatla, Sepanta Zeighami, Kameron Shahabi, Cyrus Shahabi, Yan Liu

2020-12-14

General

Towards Accurate Spatiotemporal COVID-19 Risk Scores using High Resolution Real-World Mobility Data

ArXiv Preprint

As countries look towards re-opening of economic activities amidst the ongoing COVID-19 pandemic, ensuring public health has been challenging. While contact tracing only aims to track past activities of infected users, one path to safe reopening is to develop reliable spatiotemporal risk scores to indicate the propensity of the disease. Existing works which aim to develop risk scores either rely on compartmental model-based reproduction numbers (which assume uniform population mixing) or develop coarse-grain spatial scores based on reproduction number (R0) and macro-level density-based mobility statistics. Instead, in this paper, we develop a Hawkes process-based technique to assign relatively fine-grain spatial and temporal risk scores by leveraging high-resolution mobility data based on cell-phone originated location signals. While COVID-19 risk scores also depend on a number of factors specific to an individual, including demography and existing medical conditions, the primary mode of disease transmission is via physical proximity and contact. Therefore, we focus on developing risk scores based on location density and mobility behaviour. We demonstrate the efficacy of the developed risk scores via simulation based on real-world mobility data. Our results show that fine-grain spatiotemporal risk scores based on high-resolution mobility data can provide useful insights and facilitate safe re-opening.

Sirisha Rambhatla, Sepanta Zeighami, Kameron Shahabi, Cyrus Shahabi, Yan Liu

2020-12-14

General

Predicting hospitalization following psychiatric crisis care using machine learning.

In BMC medical informatics and decision making ; h5-index 38.0

BACKGROUND : Accurate prediction models for whether patients on the verge of a psychiatric criseis need hospitalization are lacking and machine learning methods may help improve the accuracy of psychiatric hospitalization prediction models. In this paper we evaluate the accuracy of ten machine learning algorithms, including the generalized linear model (GLM/logistic regression) to predict psychiatric hospitalization in the first 12 months after a psychiatric crisis care contact. We also evaluate an ensemble model to optimize the accuracy and we explore individual predictors of hospitalization.

METHODS : Data from 2084 patients included in the longitudinal Amsterdam Study of Acute Psychiatry with at least one reported psychiatric crisis care contact were included. Target variable for the prediction models was whether the patient was hospitalized in the 12 months following inclusion. The predictive power of 39 variables related to patients' socio-demographics, clinical characteristics and previous mental health care contacts was evaluated. The accuracy and area under the receiver operating characteristic curve (AUC) of the machine learning algorithms were compared and we also estimated the relative importance of each predictor variable. The best and least performing algorithms were compared with GLM/logistic regression using net reclassification improvement analysis and the five best performing algorithms were combined in an ensemble model using stacking.

RESULTS : All models performed above chance level. We found Gradient Boosting to be the best performing algorithm (AUC = 0.774) and K-Nearest Neighbors to be the least performing (AUC = 0.702). The performance of GLM/logistic regression (AUC = 0.76) was slightly above average among the tested algorithms. In a Net Reclassification Improvement analysis Gradient Boosting outperformed GLM/logistic regression by 2.9% and K-Nearest Neighbors by 11.3%. GLM/logistic regression outperformed K-Nearest Neighbors by 8.7%. Nine of the top-10 most important predictor variables were related to previous mental health care use.

CONCLUSIONS : Gradient Boosting led to the highest predictive accuracy and AUC while GLM/logistic regression performed average among the tested algorithms. Although statistically significant, the magnitude of the differences between the machine learning algorithms was in most cases modest. The results show that a predictive accuracy similar to the best performing model can be achieved when combining multiple algorithms in an ensemble model.

Blankers Matthijs, van der Post Louk F M, Dekker Jack J M

2020-Dec-10

Acute psychiatry, Machine learning, Prognostic modeling, Psychiatric hospitalization

General

Reporting of screening and diagnostic AI rarely acknowledges ethical, legal, and social implications: a mass media frame analysis.

In BMC medical informatics and decision making ; h5-index 38.0

BACKGROUND : Healthcare is a rapidly expanding area of application for Artificial Intelligence (AI). Although there is considerable excitement about its potential, there are also substantial concerns about the negative impacts of these technologies. Since screening and diagnostic AI tools now have the potential to fundamentally change the healthcare landscape, it is important to understand how these tools are being represented to the public via the media.

METHODS : Using a framing theory approach, we analysed how screening and diagnostic AI was represented in the media and the frequency with which media articles addressed the benefits and the ethical, legal, and social implications (ELSIs) of screening and diagnostic AI.

RESULTS : All the media articles coded (n = 136) fit into at least one of three frames: social progress (n = 131), economic development (n = 59), and alternative perspectives (n = 9). Most of the articles were positively framed, with 135 of the articles discussing benefits of screening and diagnostic AI, and only 9 articles discussing the ethical, legal, and social implications.

CONCLUSIONS : We found that media reporting of screening and diagnostic AI predominantly framed the technology as a source of social progress and economic development. Screening and diagnostic AI may be represented more positively in the mass media than AI in general. This represents an opportunity for health journalists to provide publics with deeper analysis of the ethical, legal, and social implications of screening and diagnostic AI, and to do so now before these technologies become firmly embedded in everyday healthcare delivery.

Frost Emma K, Carter Stacy M

2020-Dec-10

Artificial intelligence, Diagnosis, Ethics, Frame analysis, Media framing, Screening

General

Five novel clinical phenotypes for critically ill patients with mechanical ventilation in intensive care units: a retrospective and multi database study.

In Respiratory research ; h5-index 45.0

BACKGROUND : Although protective mechanical ventilation (MV) has been used in a variety of applications, lung injury may occur in both patients with and without acute respiratory distress syndrome (ARDS). The purpose of this study is to use machine learning to identify clinical phenotypes for critically ill patients with MV in intensive care units (ICUs).

METHODS : A retrospective cohort study was conducted with 5013 patients who had undergone MV and treatment in the Department of Critical Care Medicine, Peking Union Medical College Hospital. Statistical and machine learning methods were used. All the data used in this study, including demographics, vital signs, circulation parameters and mechanical ventilator parameters, etc., were automatically extracted from the electronic health record (EHR) system. An external database, Medical Information Mart for Intensive Care III (MIMIC III), was used for validation.

RESULTS : Phenotypes were derived from a total of 4009 patients who underwent MV using a latent profile analysis of 22 variables. The associations between the phenotypes and disease severity and clinical outcomes were assessed. Another 1004 patients in the database were enrolled for validation. Of the five derived phenotypes, phenotype I was the most common subgroup (n = 2174; 54.2%) and was mostly composed of the postoperative population. Phenotype II (n = 480; 12.0%) led to the most severe conditions. Phenotype III (n = 241; 6.01%) was associated with high positive end-expiratory pressure (PEEP) and low mean airway pressure. Phenotype IV (n = 368; 9.18%) was associated with high driving pressure, and younger patients comprised a large proportion of the phenotype V group (n = 746; 18.6%). In addition, we found that the mortality rate of Phenotype IV was significantly higher than that of the other phenotypes. In this subgroup, the number of patients in the sequential organ failure assessment (SOFA) score segment (9,22] was 198, the number of deaths was 88, and the mortality rate was higher than 44%. However, the cumulative 28-day mortality of Phenotypes IV and II, which were 101 of 368 (27.4%) and 87 of 480 (18.1%) unique patients, respectively, was significantly higher than those of the other phenotypes. There were consistent phenotype distributions and differences in biomarker patterns by phenotype in the validation cohort, and external verification with MIMIC III further generated supportive results.

CONCLUSIONS : Five clinical phenotypes were correlated with different disease severities and clinical outcomes, which suggested that these phenotypes may help in understanding heterogeneity in MV treatment effects.

Su Longxiang, Zhang Zhongheng, Zheng Fanglan, Pan Pan, Hong Na, Liu Chun, He Jie, Zhu Weiguo, Long Yun, Liu Dawei

2020-Dec-10

Clinical phenotype, Critically ill patients, Machine learning, Mechanical ventilation

General

In Fortschritte der Neurologie-Psychiatrie

'Precision Psychiatry' as the psychiatric variant of 'Precision Medicine' aims to provide high-level diagnosis and treatment based on robust biomarkers and tailored to the individual clinical, neurobiological, and genetic constitution of the patient. The specific peculiarity of psychiatry, in which disease entities are normatively defined based on clinical experience and are also significantly influenced by contemporary history, society and philosophy, has so far made the search for valid and reliable psychobiological connections difficult. Nevertheless, considerable progress has now been made in all areas of psychiatric research, made possible above all by the critical review and renewal of previous concepts of disease and psychopathology, the increased orientation towards neurobiology and genetics, and in particular the use of machine learning methods. Notably, modern machine learning methods make it possible to integrate high-dimensional and multimodal data sets and generate models which provide new psychobiological insights and offer the possibility of individualized, biomarker-driven single-subject prediction of diagnosis, therapy response and prognosis. The aim of the present review is therefore to introduce the concept of 'Precision Psychiatry' to the interested reader, to concisely present modern, machine learning methods required for this, and to clearly present the current state and future of biomarker-based 'precision psychiatry'.

Popovic David, Schiltz Kolja, Falkai Peter, Koutsouleris Nikolaos

2020-Nov

General

Unsupervised fNIRS feature extraction with CAE and ESN autoencoder for driver cognitive load classification.

In Journal of neural engineering ; h5-index 52.0

OBJECTIVE : Understanding the cognitive load of drivers is crucial for road safety. Brain sensing has the potential to provide an objective measure of driver cognitive load. We aim to develop an advanced machine learning framework for classifying driver cognitive load using functional near-infrared spectroscopy (fNIRS).

APPROACH : We conducted a study using fNIRS in a driving simulator with the n-back task used as a secondary task to impart structured cognitive load on drivers. To classify different driver cognitive load levels, we examined the application of convolutional autoencoder (CAE) and Echo State Network (ESN) autoencoder for extracting features from fNIRS.

MAIN RESULTS : By using CAE, the accuracies for classifying two and four levels of driver cognitive load with the 30s window were 73.25% and 47.21, respectively. The proposed ESN autoencoder achieved state-of-art classification results for group-level models without window selection, with accuracies of 80.61% and 52.45 for classifying two and four levels of driver cognitive load.

SIGNIFICANCE : This work builds a foundation for using fNIRS to measure driver cognitive load in real-world applications. Also, the results suggest that the proposed ESN autoencoder can effectively extract temporal information from fNIRS data and can be useful for other fNIRS data classification tasks.

Liu Ruixue, Reimer Bryan, Song Siyang, Mehler Bruce, Solovey Erin

2020-Dec-11

convolutional autoencoder, driver cognitive load, echo state network, fnirs, functional near-infrared spectroscopy

Oncology

Influence of abiraterone and enzalutamide on body composition in patients with metastatic castration resistant prostate cancer.

In Cancer treatment and research communications

INTRODUCTION : Loss of skeletal muscle (SM) and gain of subcutaneous fat (SCF) are known side-effects of androgen-deprivation in treatment of prostate cancer. Scarce data is available concerning the effects of abiraterone/pred (ABI) on body composition and no published data regarding enzalutamide (ENZA). Our objective was to analyse the effects of ENZA on SM/SCF and to compare the results with ABI in patients with metastatic castration-resistant prostate-cancer (mCRPC).

PATIENTS AND METHODS : 54 patients starting ABI (n = 17) or ENZA (n = 37) at a single-centre between 2012 and 2018 were retrospectively identified. SM and SCF were assessed using CT-scans at baseline and after a median of 10.8 months on treatment. A deep learning image-segmentation software was used to quantify SM and SCF. In a subgroup of patients receiving ENZA within a trial, we investigated change of SM using serial timepoints.

RESULTS : Median time of treatment with ABI/ENZA was 14.6 months. A significant loss of SM compared to baseline was observed for ENZA (mean loss 5.2%, p<0.0001) and ABI (mean loss 3.0%, p = 0.02). SCF was not significantly altered. The effects of both drugs did not differ significantly. Loss of SM occurred early on during treatment with ENZA.

CONCLUSION : Treatment with ENZA seems to lead to a loss of SM which is comparable to that of ABI. Further evaluation in larger patient-cohorts is warranted. In routine care, counselling of patients about side effects of ABI/ENZA should include discussions about SM loss.

Fischer Stefanie, Clements Sebastian, McWilliam Alan, Green Andrew, Descamps Tine, Oing Christoph, Gillessen Silke

2020-Dec-01

Abiraterone, Enzalutamide, Muscle loss, Prostate cancer, Sarcopenia

Oncology

Understanding the organ tropism of metastatic breast cancer through the combination of liquid biopsy tools.

In European journal of cancer (Oxford, England : 1990)

BACKGROUND : Liquid biopsy provides real-time data about prognosis and actionable mutations in metastatic breast cancer (MBC). The aim of this study was to explore the combination of circulating tumour DNA (ctDNA) analysis and circulating tumour cells (CTCs) enumeration in estimating target organs more susceptible to MBC involvement.

METHODS : This retrospective study analysed 88 MBC patients characterised for both CTCs and ctDNA at baseline. CTCs were isolated through the CellSearch kit, while ctDNA was analysed using the Guardant360 NGS-based assay. Sites of disease were collected on the basis of imaging. Associations were explored both through uni- and multivariate logistic regression and Fisher's exact test and the random forest machine learning algorithm.

RESULTS : After multivariate logistic regression, ESR1 mutation was the only significant factor associated with liver metastases (OR 8.10), while PIK3CA was associated with lung localisations (OR 3.74). CTC enumeration was independently associated with bone metastases (OR 10.41) and TP53 was associated with lymph node localisations (OR 2.98). The metastatic behaviour was further investigated through a random forest machine learning algorithm. Bone involvement was described by CTC enumeration and alterations in ESR1, GATA3, KIT, CDK4 and ERBB2, while subtype, CTC enumeration, inflammatory BC diagnosis, ESR1 and KIT aberrations described liver metastases. PIK3CA, MET, AR, CTC enumeration and TP53 were associated with lung organotropism. The model, moreover, showed that AR, CCNE1, ESR1, MYC and CTC enumeration were the main drivers in HR positive MBC metastatic pattern.

CONCLUSIONS : These results indicate that ctDNA and CTCs enumeration could provide useful insights regarding MBC organotropism, suggesting a possible role for future monitoring strategies that dynamically focus on high-risk organs defined by tumourbiology.

Gerratana Lorenzo, Davis Andrew A, Polano Maurizio, Zhang Qiang, Shah Ami N, Lin Chenyu, Basile Debora, Toffoli Giuseppe, Wehbe Firas, Puglisi Fabio, Behdad Amir, Platanias Leonidas C, Gradishar William J, Cristofanilli Massimo

2020-Dec-08

Circulating tumour DNA, Circulating tumour cell, Liquid biopsy, Metastatic breast cancer, Organotropism, Precision medicine

Radiology

Identification of breast cancer patients with pathologic complete response in the breast after neoadjuvant systemic treatment by an intelligent vacuum-assisted biopsy.

In European journal of cancer (Oxford, England : 1990)

BACKGROUND : Neoadjuvant systemic treatment elicits a pathologic complete response (pCR) in about 35% of women with breast cancer. In such cases, breast surgery may be considered overtreatment. We evaluated multivariate algorithms using patient, tumor, and vacuum-assisted biopsy (VAB) variables to identify patients with breast pCR.

METHODS : We developed and tested four multivariate algorithms: a logistic regression with elastic net penalty, an Extreme Gradient Boosting (XGBoost) tree, Support Vector Machines (SVM), and neural network. We used data from 457 women, randomly partitioned into training and test set (2:1), enrolled in three trials with stage 1-3 breast cancer, undergoing VAB before surgery. False-negative rate (FNR) and specificity were the main outcome measures. The best performing algorithm was validated in an independent fourth trial.

RESULTS : In the test set (n = 152), the logistic regression with elastic net penalty, XGboost tree, SVM, and neural network revealed an FNR of 1.2% (1 of 85 patients with missed residual cancer). Specificity of the logistic regression with elastic net penalty was 52.2% (35 of 67 women with surgically confirmed breast pCR identified), of the XGBoost tree 55.2% (37 of 67), of SVM 62.7% (42 of 67), and of the neural network 67.2% (45 of 67). External validation (n = 50) of the neural network showed an FNR of 0% (0 of 27) and a specificity of 65.2% (15 of 23). Area under the ROC curve for the neural network was 0.97 (95% CI, 0.94-1.00).

CONCLUSION : A multivariate algorithm can accurately select breast cancer patients without residual cancer after neoadjuvant treatment.

Pfob André, Sidey-Gibbons Chris, Lee Han-Byoel, Tasoulis Marios Konstantinos, Koelbel Vivian, Golatta Michael, Rauch Gaiane M, Smith Benjamin D, Valero Vicente, Han Wonshik, MacNeill Fiona, Weber Walter Paul, Rauch Geraldine, Kuerer Henry M, Heil Joerg

2020-Dec-08

Artificial intelligence, Breast cancer, Individualized treatment, Machine learning, Neoadjuvant systemic treatment, Pathologic complete response, Surgical oncology, Vacuum-assisted biopsy

Oncology

Artificial intelligence applications for oncological positron emission tomography imaging.

In European journal of radiology ; h5-index 47.0

Positron emission tomography (PET), a functional and dynamic molecular imaging technique, is generally used to reveal tumors' biological behavior. Radiomics allows a high-throughput extraction of multiple features from images with artificial intelligence (AI) approaches and develops rapidly worldwide. Quantitative and objective features of medical images have been explored to recognize reliable biomarkers, with the development of PET radiomics. This paper will review the current clinical exploration of PET-based classical machine learning and deep learning methods, including disease diagnosis, the prediction of histological subtype, gene mutation status, tumor metastasis, tumor relapse, therapeutic side effects, therapeutic intervention and evaluation of prognosis. The applications of AI in oncology will be mainly discussed. The image-guided biopsy or surgery assisted by PET-based AI will be introduced as well. This paper aims to present the applications and methods of AI for PET imaging, which may offer important details for further clinical studies. Relevant precautions are put forward and future research directions are suggested.

Li Wanting, Liu Haiyan, Cheng Feng, Li Yanhua, Li Sijin, Yan Jiangwei

2020-Nov-30

Artificial intelligence, Deep learning, Machine learning, Positron emission tomography, Radiomics

Oncology

Predict multicategory causes of death in lung cancer patients using clinicopathologic factors.

In Computers in biology and medicine

BACKGROUND : Random forests (RF) is a widely used machine-learning algorithm, and outperforms many other machine learning algorithms in prediction-accuracy. But it is rarely used for predicting causes of death (COD) in cancer patients. On the other hand, multicategory COD are difficult to classify in lung cancer patients, largely because they have multiple labels (versus binary labels).

METHODS : We tuned RF algorithms to classify 5-category COD among the lung cancer patients in the surveillance, epidemiology and end results-18, whose lung cancers were diagnosed in 2004, for the completeness in their follow-up. The patients were randomly divided into training and validation sets (1:1 and 4:1 sample-splits). We compared the prediction accuracy of the tuned RF and multinomial logistic regression (MLR) models.

RESULTS : We included 42,257 qualified lung cancers in the database. The COD were lung cancer (72.41%), other causes or alive (14.43%), non-lung cancer (6.85%), cardiovascular disease (5.35%), and infection (0.96%). The tuned RF model with 300 iterations and 10 variables outperformed the MLR model (accuracy = 69.8% vs 64.6%, 1:1 sample-split), while 4:1 sample-split produced lower prediction-accuracy than 1:1 sample-split. The top-10 important factors in the RF model were sex, chemotherapy status, age (65+ vs < 65 years), radiotherapy status, nodal status, T category, histology type and laterality, all of which except T category and laterality were also important in MLR model.

CONCLUSION : We tuned RF models to predict 5-category CODs in lung cancer patients, and show RF outperforms MLR in prediction accuracy. We also identified the factors associated with these COD.

Deng Fei, Zhou Haijun, Lin Yong, Heim John A, Shen Lanlan, Li Yuan, Zhang Lanjing

2020-Dec-01

Lung cancer, Machine learning, Multi-label classification, Multinomial logistic regression

General

Biomarker discovery by feature ranking: Evaluation on a case study of embryonal tumors.

In Computers in biology and medicine

The task of biomarker discovery is best translated to the machine learning task of feature ranking. Namely, the goal of biomarker discovery is to identify a set of potentially viable targets for addressing a given biological status. This is aligned with the definition of feature ranking and its goal - to produce a list of features ordered by their importance for the target concept. This differs from the task of feature selection (typically used for biomarker discovery) in that it catches viable biomarkers that have redundant or overlapping information with often highly important biomarkers, while with feature selection this is not the case. We propose to use a methodology for evaluating feature rankings to assess the quality of a given feature ranking and to discover the best cut-off point. We demonstrate the effectiveness of the proposed methodology on 10 datasets containing data about embryonal tumors. We evaluate two most commonly used feature ranking algorithms (Random forests and RReliefF) and using the evaluation methodology identifies a set of viable biomarkers that have been confirmed to be related to cancer.

Petković Matej, Slavkov Ivica, Kocev Dragi, Džeroski Sašo

2020-Nov-28

Biomedicine application, Feature ranking evaluation, Tumor data

General

ScGSLC: An unsupervised graph similarity learning framework for single-cell RNA-seq data clustering.

In Computational biology and chemistry

Accurate clustering of cells from single-cell RNA sequencing (scRNA-seq) data is an essential step for biological analysis such as putative cell type identification. However, scRNA-seq data has high dimension and high sparsity, which makes traditional clustering methods less effective to reflect the similarity between cells. Since genetic network fundamentally defines the functions of cell and deep learning shows strong advantages in network representation learning, we propose a novel scRNA-seq clustering framework ScGSLC based on graph similarity learning. ScGSLC effectively integrates scRNA-seq data and protein-protein interaction network to a graph. Then graph convolution network is employed by ScGSLC to embedding graph and clustering the cells by the calculated similarity between graphs. Unsupervised clustering results of nine public data sets demonstrate that ScGSLC shows better performance than the state-of-the-art methods.

Li Junyi, Jiang Wei, Han Henry, Liu Jing, Liu Bo, Wang Yadong

2020-Nov-18

Graph convolution network, Graph embedding, Graph similarity, Single-cell RNA sequencing data, Unsupervised clustering

General

Variable selection in social-environmental data: sparse regression and tree ensemble machine learning approaches.

In BMC medical research methodology

BACKGROUND : Social-environmental data obtained from the US Census is an important resource for understanding health disparities, but rarely is the full dataset utilized for analysis. A barrier to incorporating the full data is a lack of solid recommendations for variable selection, with researchers often hand-selecting a few variables. Thus, we evaluated the ability of empirical machine learning approaches to identify social-environmental factors having a true association with a health outcome.

METHODS : We compared several popular machine learning methods, including penalized regressions (e.g. lasso, elastic net), and tree ensemble methods. Via simulation, we assessed the methods' ability to identify census variables truly associated with binary and continuous outcomes while minimizing false positive results (10 true associations, 1000 total variables). We applied the most promising method to the full census data (p = 14,663 variables) linked to prostate cancer registry data (n = 76,186 cases) to identify social-environmental factors associated with advanced prostate cancer.

RESULTS : In simulations, we found that elastic net identified many true-positive variables, while lasso provided good control of false positives. Using a combined measure of accuracy, hierarchical clustering based on Spearman's correlation with sparse group lasso regression performed the best overall. Bayesian Adaptive Regression Trees outperformed other tree ensemble methods, but not the sparse group lasso. In the full dataset, the sparse group lasso successfully identified a subset of variables, three of which replicated earlier findings.

CONCLUSIONS : This analysis demonstrated the potential of empirical machine learning approaches to identify a small subset of census variables having a true association with the outcome, and that replicate across empiric methods. Sparse clustered regression models performed best, as they identified many true positive variables while controlling false positive discoveries.

Handorf Elizabeth, Yin Yinuo, Slifker Michael, Lynch Shannon

2020-Dec-10

Social environment, Variable selection

Pathology

Identification of Aspergillus species in human blood plasma by infrared spectroscopy and machine learning.

In Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy

Invasive Aspergillosis is a challenging infection that requires convenient, efficient, and cost-effective diagnostics. This study addresses the potential of infrared spectroscopy to satisfy this clinical need with the aid of machine learning. Two models, based on Partial Least Squares-Discriminant Analysis (PLS-DA), have been trained by a set of infrared spectral data of 9 Aspergillus-spiked and 7 Aspergillus-free plasma samples, and a set of 200 spectral data simulated by oversampling these 16 samples. Two further models have also been trained by the same sets but with auto-scaling performed prior to PLS-DA. These models were assessed using 45 mock samples, simulating the challenging samples of patients at risk of Invasive Aspergillosis, including the presence of drugs (9 tested) and other common pathogens (5 tested) as potential confounders. The simple model shows good prediction performance, yielding a total accuracy of 84.4%, while oversampling and autoscaling improved this accuracy to 93.3%. The results of this study have shown that infrared spectroscopy can identify Aspergillus species in blood plasma even in presence of potential confounders commonly present in blood of patients at risk of Invasive Aspergillosis.

Elkadi Omar Anwar, Hassan Reem, Elanany Mervat, Byrne Hugh J, Ramadan Mohammed A

2020-Nov-30

Aspergillosis, Infrared spectroscopy, Laboratory diagnosis, Machine learning, Plasma

General

Artificial intelligence (AI) impacting diagnosis of glaucoma and understanding the regulatory aspects of AI-based software as medical device.

In Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society

Glaucoma, the group of eye diseases is characterized by increased intraocular pressure, optic neuropathy and visual field defect patterns. Early and correct diagnosis of glaucoma can prevent irreversible vision loss and glaucomatous structural damages to the eye. However, greater chances of misdiagnosis by the currently used conventional methods for diagnosis open up ways for more advanced techniques like the use of artificial intelligence (AI). Artificial intelligence coupled with optical coherence tomography imaging creates an algorithm that can be effectively used to make a model of complex data for detection as well as diagnosis of glaucoma. The present review is an attempt to provide state-of-the-art information on various AI techniques used in the diagnosis and assessment of glaucoma. The second part of the review is focused on understanding how the AI along with machine learning (ML) can be potentially used to be subjected for software as a medical device (SaMD) in precise diagnosis or early detection of disease conditions.

Prabhakar Bala, Singh Rishi Kumar, Yadav Khushwant S

2020-Nov-24

Artificial intelligence, Convolutional neural networks, Deep learning, Glaucoma diagnosis, Machine learning, Software as a medical device (SaMD), Support vector machine

General

Atlas-based score for automatic glaucoma risk stratification.

In Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society

Glaucoma is a disease that affects the optic nerve and can lead to blindness. The cup-to-disc ratio (CDR) measurement is one of the key clinical indicators for glaucoma assessment. However, the CDR only evaluates the relative sizes of the cup and optic disc (OD) via their diameters, and does not characterize local morphological changes that can inform clinicians on early signs of glaucoma. In this work, we propose a novel glaucoma score based on a statistical atlas framework that automatically quantifies the deformations of the OD region induced by glaucoma. A deep-learning approach is first used to segment the optic cup with a dedicated atlas-based data augmentation strategy. The segmented OD region (disc, cup and vessels) is then registered to the statistical OD atlas and the deformation is projected onto the atlas eigenvectors. The atlas glaucoma score (AGS) is then obtained by a linear combination of the principal modes of deformation of the atlas with linear discriminant analysis. The AGS performs better than the CDR on the three datasets used for evaluation, including RIM-ONE and ORIGA650. Compared to the CDR measurement, which yields an area under the ROC curve (AUC) of 91.4% using the expert segmentations, the AGS achieves an AUC of 98.2%. Our novel glaucoma score captures more complex deformations within the optic disc region than the CDR can. Such morphological changes are the first cue of glaucoma onset, before the visual field is affected. The proposed approach can thus significantly improve early detection of glaucoma.

Girard Fantin, Hurtut Thomas, Kavalec Conrad, Cheriet Farida

2020-Oct-16

Deep learning segmentation, Glaucoma detection, Retina, Statistical atlas

General

Deep-learned spike representations and sorting via an ensemble of auto-encoders.

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

Spike sorting refers to the technique of detecting signals generated by single neurons from multi-neuron recordings and is a valuable tool for analyzing the relationships between individual neuronal activity patterns and specific behaviors. Since the precision of spike sorting affects all subsequent analyses, sorting accuracy is critical. Many semi-automatic to fully-automatic spike sorting algorithms have been developed. However, due to unsatisfactory classification accuracy, manual sorting is preferred by investigators despite the intensive time and labor costs. Thus, there still is a strong need for fully automatic spike sorting methods with high accuracy. Various machine learning algorithms have been developed for feature extraction but have yet to show sufficient accuracy for spike sorting. Here we describe a deep learning-based method for extracting features from spike signals using an ensemble of auto-encoders, each with a distinct architecture for distinguishing signals at different levels of resolution. By utilizing ensemble of auto-encoder ensemble, where shallow networks better represent overall signal structure and deep networks better represent signal details, extraction of high-dimensional representative features for improved spike sorting performance is achieved. The model was evaluated on publicly available simulated datasets and single-channel and 4-channel tetrode in vivo datasets. Our model not only classified single-channel spikes with varying degrees of feature similarities and signal to noise levels with higher accuracy, but also more precisely determined the number of source neurons compared to other machine learning methods. The model also demonstrated greater overall accuracy for spike sorting 4-channel tetrode recordings compared to single-channel recordings.

Eom Junsik, Park In Yong, Kim Sewon, Jang Hanbyol, Park Sanggeon, Huh Yeowool, Hwang Dosik

2020-Nov-27

Clustering, Deep learning-based auto-encoder, Feature extraction, Unsupervised spike sorting

Pathology

Changing the landscape of tumor immunology: novel tools to examine T cell specificity.

In Current opinion in immunology ; h5-index 59.0

Immunotherapy has established itself as a stalwart arm in patient care and with precision medicine forms the new paradigm in cancer treatment. T cells are an important group of immune cells capable of potent cancer immune surveillance and immunity. The advent of bioinformatics, particularly more recent advances incorporating algorithms employing machine learning, provide a seemingly limitless ability for T cell analysis and hypothesis generation. Such endeavors have become indispensable to research efforts accelerating and evolving to such an extent that there exists an appreciable gap between knowledge and proof of function and application. Exciting new technologies such as DNA barcoding, cytometry by time-of-flight (CyTOF), and peptide-exchangeable pHLA multimers inclusive of rare and difficult HLA alleles offer high-throughput cell-by-cell analytical capabilities. These outstanding recent contributions to T cell research will help close this gap and potentially bring practical benefit to patients.

Rahman Muhammed A, Murata Kenji, Burt Brian D, Hirano Naoto

2020-Dec-08

Dermatology

Deep learning embedder method and tool for spectral similarity search.

In Journal of proteomics

Spectral similarity calculation is widely used in protein identification tools and mass spectra clustering algorithms while comparing theoretical or experimental spectra. The performance of the spectral similarity calculation plays an important role in these tools and algorithms especially in the analysis of large-scale datasets. Recently, deep learning methods have been proposed to improve the performance of clustering algorithms and protein identification by training the algorithms with existing data and the use of multiple spectra and identified peptide features. While the efficiency of these algorithms is still under study in comparison with traditional approaches, their application in proteomics data analysis is becoming more common. Here, we propose the use of deep learning to improve spectral similarity comparison. We assessed the performance of deep learning for spectral similarity, with GLEAMS and a newly trained embedder model (DLEAMSE), which uses high-quality spectra from PRIDE Cluster. Also, we developed a new bioinformatics tool (mslookup - https://github.com/bigbio/DLEAMSE/) that allows users to quickly search for spectra in previously identified mass spectra publish in public repositories and spectral libraries. Finally, we released a human database to enable bioinformaticians and biologists to search for identified spectra in their machines. SIGNIFICANCE STATEMENT: Spectral similarity calculation plays an important role in proteomics data analysis. With deep learning's ability to learn the implicit and effective features from large-scale training datasets, deep learning-based MS/MS spectra embedding models has emerged as a solution to improve mass spectral clustering similarity calculation algorithms. We compare multiple similarity scoring and deep learning methods in terms of accuracy (compute the similarity for a pair of the mass spectrum) and computing-time performance. The benchmark results showed no major differences in accuracy between DLEAMSE and normalized dot product for spectrum similarity calculations. The DLEAMSE GPU implementation is faster than NDP in preprocessing on the GPU server and the similarity calculation of DLEAMSE (Euclidean distance on 32-D vectors) takes about 1/3 of dot product calculations. The deep learning model (DLEAMSE) encoding and embedding steps needed to run once for each spectrum and the embedded 32-D points can be persisted in the repository for future comparison, which is faster for future comparisons and large-scale data. Based on these, we proposed a new tool mslookup that enables the researcher to find spectra previously identified in public data. The tool can be also used to generate in-house databases of previously identified spectra to share with other laboratories and consortiums.

Qin Chunyuan, Luo Xiyang, Deng Chuan, Shu Kunxian, Zhu Weimin, Griss Johannes, Hermjakob Henning, Bai Mingze, Perez-Riverol Yasset

2020-Dec-08

Deep learning, Mass spectra embedder, Scoring function, Spectral similarity

Oncology

Stereotactic ablative radiation therapy to all lesions in patients with oligometastatic cancers: a phase I dose-escalation trial.

In International journal of radiation oncology, biology, physics

PURPOSE : Increasing evidence suggests that patients with a limited number of metastases benefit from stereotactic ablative radiation therapy (SABR) to all lesions. However, the optimal dose and fractionation remain unknown. This is particularly true for bone and lymph node metastases. Therefore, a prospective single-center dose-escalation trial was initiated.

METHODS : XXXX was an open-label phase I trial evaluating SABR to non-spine bone and lymph node lesions in patients with up to three metastases. Patients with European Cooperative Oncology Group performance status ≤ 1, an estimated life expectancy of at least 6 months, and histologically confirmed non-hematological malignancy were eligible. Three SABR fractionation regimens, i.e. 5 fractions of 7.0 Gy vs. 3 fractions of 10.0 Gy vs. a single fraction of 20.0 Gy, were applied in three consecutive patient cohorts. The rate of ≥ grade 3 toxicity, scored according to the Common Toxicity Criteria for Adverse Events v. 4.03, up to 6 months after SABR, was the primary endpoint. The trial was registered on clinicaltrials.gov (NCTXXXX).

RESULTS : Between July 2017 and December 2018, 90 patients were enrolled. In total 101 metastases were treated. No ≥ grade 3 toxicity was observed in any of the enrolled patients (95% CI 0.0-12.3% for the first cohort with 28 analyzable patients, 95% CI 0.0-11.6% for the second and third cohort with 30 analyzable patients each). Treatment-related grade 2 toxicities occurred in 4/30 vs. 2/30 vs. 2/30 patients for the five, three and one fraction schedule, respectively. Actuarial local control rate at 12 months was 94.5%.

CONCLUSION : All three treatment schedules were feasible and effective with remarkably low toxicity rates and high local control rates. From a patient and resource point of view, the single-fraction schedule is undoubtedly most convenient.

Mercier Carole, Claessens Michaël, Buys Andy, Gryshkevych Sergii, Billiet Charlotte, Joye Ines, Van Laere Steven, Vermeulen Peter, Meijnders Paul, Löfman Fredrik, Poortmans Philip, Dirix Luc, Verellen Dirk, Dirix Piet

2020-Dec-08

Pathology

Image-based machine learning algorithms for disease characterization in the human type 1 diabetes pancreas.

In The American journal of pathology ; h5-index 54.0

Emerging data suggest that type 1 diabetes affects not only the β-cell-containing islets of Langerhans, but also the surrounding exocrine compartment. Using digital pathology, machine learning algorithms were applied to provide high-resolution, whole-slide images of human pancreata to determine if the tissue composition in individuals with or at-risk for type 1 diabetes differs from those without diabetes. Transplant grade pancreata from organ donors were evaluated from 16 non-diabetic autoantibody negative controls, 8 non-diabetic autoantibody positive subjects who have increased-type 1 diabetes risk, and 19 persons with type 1 diabetes (0-12 years duration). HALO image analysis algorithms were implemented to compare architecture of the main pancreatic duct as well as cell size, density, and area of acinar, endocrine, ductal, and other non-endocrine, non-exocrine tissues. Type 1 diabetes was found to affect exocrine area, acinar cell density, and size while the type of difference correlated with the presence or absence of insulin-positive cells remaining in the pancreas. These changes were not observed before disease onset, as indicated by modeling cross-sectional data from pancreata of autoantibody positive subjects and those diagnosed with type 1 diabetes. These data provide novel insights into anatomical differences in type 1 diabetes pancreata and demonstrate that machine learning can be adapted for the evaluation of disease processes from cross-sectional datasets.

Tang Xiaohan, Kusmartseva Irina, Kulkarni Shweta, Posgai Amanda, Speier Stephan, Schatz Desmond A, Haller Michael J, Campbell-Thompson Martha, Wasserfall Clive H, Roep Bart O, Kaddis John S, Atkinson Mark A

2020-Dec-08

Algorithm, cell, endocrine, exocrine, human, modeling, pancreas, tissue, type 1 diabetes

General

How can technology assist occupational voice users?

In Disability and rehabilitation. Assistive technology

SUMMARY : The voice is an important tool for people who use it daily in their occupations. However, what technological options are available to such individuals to allow them to monitor or take care of their voices?

OBJECTIVE : The purpose of this study is to answer two research questions: (1) What technologies exist to monitor or take care of the voice in occupational voice users? (2) What is the technology readiness level (TRL) of the technologies used to monitor or take care of the voice in occupational voice users?

DATA SOURCES : Embase, IEEE, Medline, Proquest, PubMed, Scopus, and Web of Science.

METHODS : A systematic literature review was conducted. Articles that reported results regarding technologies (hardware, software, or mobile apps) that were used to monitor or take care of the voice in occupational voice users were included.

RESULTS : After reviewing 4581 abstracts, 10 full text studies were included in the literature review. The technologies found include 30% hardware, 30% hardware plus software, and 50% mobile apps, with an overall TRL mean of 5.3 (SD = 2.3).

CONCLUSION : Further research is necessary for higher validity in the studies and to increase the readiness in the development of current technologies to offer more options for this population. Implications for Rehabilitation The evidence for the impact of the use of the technologies for occupational voice users is still low There is emerging evidence that mobile apps and artificial intelligence algorithms can be used to investigate vocal disorders or potential risks in occupational voice users More research is required to increase the readiness developmental stage of current technologies for occupational voice users.

Rodríguez-Dueñas William R, Sarmiento-Rojas Jefferson, Gómez-Medina María F, Espitia-Rojas Gleidy Vanessa

2020-Dec-11

Occupational voice, systematic review, technology, voice care

Surgery

Using Predictive Modeling and Machine Learning to Identify Patients Appropriate for Outpatient ACDF.

In Spine ; h5-index 57.0

STUDY DESIGN : Retrospective, case-control.

OBJECTIVE : The aim of this study was to use predictive modeling and machine learning to develop novel tools for identifying patients who may be appropriate for single-level outpatient anterior cervical fusion and discectomy (ACDF), and to compare these to legacy metrics.

SUMMARY OF BACKGROUND DATA : ACDF performed in an ambulatory surgical setting has started to gain popularity in recent years. Currently there are no standardized risk-stratification tools for determining which patients may be safe candidates for outpatient ACDF.

METHODS : Adult patients with American Society of Anesthesiologists (ASA) Class 1, 2, or 3 undergoing one-level ACDF in inpatient or outpatient settings were identified in the National Surgical Quality Improvement Program database. Patients were deemed as "unsafe" for outpatient surgery if they suffered any complication within a week of the index operation. Two different methodologies were used to identify unsafe candidates: a novel predictive model derived from multivariable logistic regression of significant risk factors, and an artificial neural network (ANN) using preoperative variables. Both methods were trained using randomly split 70% of the dataset and validated on the remaining 30%. The methods were compared against legacy risk-stratification measures: ASA and Charlson Comorbidity Index (CCI) using area under the curve (AUC) statistic.

RESULTS : A total of 12,492 patients who underwent single-level ACDF met the study criteria. Of these, 9.79% (1223) were deemed unsafe for outpatient ACDF given development of a complication within 1 week of the index operation. The five clinical variables that were found to be significant in the multivariable predictive model were: advanced age, low hemoglobin, high international normalized ratio, low albumin, and poor functional status. The predictive model had an AUC of 0.757, which was significantly higher than the AUC of both ASA (0.66; P < 0.001) and CCI (0.60; P < 0.001). The ANN exhibited an AUC of 0.740, which was significantly higher than the AUCs of ASA and CCI (all, P < 0.05), and comparable to that of the predictive model (P > 0.05).

CONCLUSION : Predictive analytics and machine learning can be leveraged to aid in identification of patients who may be safe candidates for single-level outpatient ACDF. Surgeons and perioperative teams may find these tools useful to augment clinical decision-making.

LEVEL OF EVIDENCE : 3.

Wang Kevin Y, Suresh Krishna V, Puvanesarajah Varun, Raad Micheal, Margalit Adam, Jain Amit

2020-Dec-08

Surgery

Blood eosinophil count combined with asthma history could predict chronic rhinosinusitis with nasal polyp recurrence.

In Acta oto-laryngologica

BACKGROUND : The use of non-invasive clinical markers for predicting CRS recurrence is still not well investigated.

OBJECTIVE : The aim of this study was to investigate the comprehensive effects of non-invasive clinical markers on the recurrence of CRS with nasal polyps (CRSwNP).

MATERIALS AND METHODS : A total of 346 consecutive CRSwNP patients undergoing endoscopic functional sinus surgery were recruited. The demographic characteristics and clinical parameters were recorded. Machine learning algorithm were used for evaluating the predictive value of asthma history and blood eosinophils percentage.

RESULTS : Finally, 313/346 patients completed the study. The average follow-up time was 24 months after the first surgery. For the CRSwNP with asthma patients, the blood eosinophils percentage cut-off value was 3.7%. However, for the CRSwNP without asthma patients, the blood eosinophils percentage cut-off value was high, at 6.9%.

CONCLUSION : Combined asthma history and blood eosinophils percentage can predict CRSwNP recurrence, while asthma history can reduce the threshold of blood eosinophils percentage to predict CRSwNP recurrence.

SIGNIFICANCE : For the CRS patients, combined asthma history and blood eosinophils percentage can predict recurrence, while asthma history can reduce the threshold of blood eosinophils percentage to predict recurrence.

Wang Xiaoyan, Meng Yifan, Lou Hongfei, Wang Kuiji, Wang Chengshuo, Zhang Luo

2020-Dec-10

Asthma, blood eosinophils percentage, chronic rhinosinusitis with nasal polyps, endoscopic sinus surgery, machine learning algorithm, prediction, recurrence

Public Health

Identifying influential neighbors in social networks and venue affiliations among young MSM: A data science approach to predict HIV infection.

In AIDS (London, England)

OBJECTIVE : Young men who have sex with men (YMSM) bear a disproportionate burden of HIV infection in the United States and their risks of acquiring HIV may be shaped by complex multi-layer social networks. These networks are formed through not only direct contact with social/sex partners but also indirect anonymous contacts encountered when attending social venues. We introduced a new application of a state-of-the-art graph-based deep learning method to predict HIV infection that can identify influential neighbors within these multiple network contexts.

DESIGN AND METHODS : We used empirical network data among YMSM aged 16-29 years old collected from Houston and Chicago in the U.S. between 2014 and 2016. A computational framework GAT-HIV (Graph Attention Networks for HIV) was proposed to predict HIV infections by identifying influential neighbors within social networks. These networks were formed by multiple relations comprised of social/sex partners and shared venue attendances, and using individual-level variables. Further, GAT-HIV was extended to combine multiple social networks using multi-graph GAT methods. A visualization tool was also developed to highlight influential network members for each individual within the multiple social networks.

RESULTS : The multi-graph GAT-HIV models obtained average AUC values of 0.776 and 0.824 for Chicago and Houston respectively, performing better than empirical predictive models (e.g. AUCs of random forest: 0.758 and 0.798). GAT-HIV on single networks also delivered promising prediction performances.

CONCLUSIONS : The proposed methods provide a comprehensive and interpretable framework for graph-based modeling that may inform effective HIV prevention intervention strategies among populations most vulnerable to HIV.

Xiang Yang, Fujimoto Kayo, Li Fang, Wang Qing, Del Vecchio Natascha, Schneider John, Zhi Degui, Tao Cui

2020-Dec-09

General

Multitask, Multilabel, and Multidomain Learning With Convolutional Networks for Emotion Recognition.

In IEEE transactions on cybernetics

Automated emotion recognition in the wild from facial images remains a challenging problem. Although recent advances in deep learning have assumed a significant breakthrough in this topic, strong changes in pose, orientation, and point of view severely harm current approaches. In addition, the acquisition of labeled datasets is costly and the current state-of-the-art deep learning algorithms cannot model all the aforementioned difficulties. In this article, we propose applying a multitask learning loss function to share a common feature representation with other related tasks. Particularly, we show that emotion recognition benefits from jointly learning a model with a detector of facial action units (collective muscle movements). The proposed loss function addresses the problem of learning multiple tasks with heterogeneously labeled data, improving previous multitask approaches. We validate the proposal using three datasets acquired in noncontrolled environments, and an application to predict compound facial emotion expressions.

Pons Gerard, Masip David

2020-Dec-11

General

Healthy Aging Within an Image: Using Muscle Radiodensitometry and Lifestyle Factors to Predict Diabetes and Hypertension.

In IEEE journal of biomedical and health informatics

The strong age dependency of many deleterious health outcomes likely reflects the cumulative effects from a variety of risk and protective factors that occur over one's life course. This notion has become increasingly explored in the etiology of chronic disease and associated comorbidities in aging. Our recent work has shown the robust classification of individuals at risk for cardiovascular pathophysiology using CT-based soft tissue radiodensity parameters obtained from nonlinear trimodal regression analysis (NTRA). Past and present lifestyle influences the incidence of comorbidities like hypertension (HTN), diabetes (DM) and cardiac diseases. 2,943 elderly subjects from the AGES-Reykjavik study were sorted into a three-level binary-tree structure defined by: 1) lifestyle factors (smoking and self-reported physical activity level), 2) comorbid HTN or DM, and 3) cardiac pathophysiology. NTRA parameters were extracted from mid-thigh CT cross-sections to quantify radiodensitometric changes in three tissue types: lean muscle, fat, and loose-connective tissue. Between-group differences were assessed at each binary-tree level, which were then used in tree-based machine learning (ML) models to classify subjects with DM or HTN. Classification scores for detecting HTN or DM based on lifestyle factors were excellent (AUCROC: 0.978 and 0.990, respectively). Finally, tissue importance analysis underlined the comparatively-high significance of connective tissue parameters in ML classification, while predictive models of DM onset from five-year longitudinal data gave a classification accuracy of 94.9%. Altogether, this work serves as an important milestone toward the construction of predictive tools for assessing the impact of lifestyle factors and healthy aging based on a single image.

Recenti Marco, Ricciardi Carlo, Edmunds Kyle, Gislason Magnus K, Sigurdsson Sigurdur, Carraro Ugo, Gargiulo Paolo

2020-Dec-11

General

Neural Network and Random Forest Models in Protein Function Prediction.

In IEEE/ACM transactions on computational biology and bioinformatics

Over the past decade, the demand for automated protein function prediction has increased due to the volume of newly sequenced proteins. In this paper, we address the function prediction task by developing an ensemble system automatically assigning Gene Ontology (GO) terms to the given input protein sequence. We develop an ensemble system which combines the GO predictions made by random forest (RF) and neural network (NN) classifiers. Both RF and NN models rely on features derived from BLAST sequence alignments, taxonomy and protein signature analysis tools. In addition, we report on experiments with a NN model that directly analyzes the amino acid sequence as its sole input, using a convolutional layer. The Swiss-Prot database is used as the training and evaluation data. In the CAFA3 evaluation, which relies on experimental verification of the functional predictions, our submitted ensemble model demonstrates competitive performance ranking among top-10 best-performing systems out of over 100 submitted systems. In this paper, we evaluate and further improve the CAFA3-submitted system. Our machine learning models together with the data pre-processing and feature generation tools are publicly available as an open source software at https://github.com/TurkuNLP/CAFA3.

Hakala Kai, Kaewphan Suwisa, Bjorne Jari, Mehryary Farrokh, Moen Hans, Tolvanen Martti, Salakoski Tapio, Ginter Filip

2020-Dec-11

General

A novel Point-in-Polygon-based sEMG classifier for Hand Exoskeleton Systems.

In IEEE transactions on neural systems and rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society

In the early 2000s, data from the latest World Health Organization estimates paint a picture where one-seventh of the world population needs at least one assistive device. Fortunately, these years are also characterized by a marked technological drive which takes the name of the Fourth Industrial Revolution. In this terrain, robotics is making its way through more and more aspects of everyday life, and robotics-based assistance/rehabilitation is considered one of the most encouraging applications. Providing high-intensity rehabilitation sessions or home assistance through low-cost robotic devices can be indeed an effective solution to democratize services otherwise not accessible to everyone. However, the identification of an intuitive and reliable real-time control system does arise as one of the critical issues to unravel for this technology in order to land in homes or clinics. Intention recognition techniques from surface ElectroMyoGraphic (sEMG) signals are referred to as one of the main ways-to-go in literature. Nevertheless, even if widely studied, the implementation of such procedures to real-case scenarios is still rarely addressed. In a previous work, the development and implementation of a novel sEMG-based classification strategy to control a fully-wearable Hand Exoskeleton System (HES) have been qualitatively assessed by the authors. This paper aims to furtherly demonstrate the validity of such a classification strategy by giving quantitative evidence about the favourable comparison to some of the standard machine-learning-based methods. Real-time action, computational lightness, and suitability to embedded electronics will emerge as the major characteristics of all the investigated techniques.

Secciani Nicola, Topini Alberto, Ridolfi Alessandro, Meli Enrico, Allotta Benedetto

2020-Dec-11

General

The complex relationship between female age and embryo euploidy.

In Minerva ginecologica

BACKGROUND : Female age is the strongest predictor of embryo chromosomal abnormalities and has a non linear relationship with the blastocyst euploidy rate: with advancing age there is an acceleration in the reduction of blastocyst euploidy. Aneuploidy was found to significantly increase with maternal age from 30% in embryos from young women to 70% in women older than 40 years old. The association seems mainly due to chromosomal abnormalities occurring in the oocyte.We aimed to elaborate a model for the blastocyst euploid rate for patients undergoing IVF/ICSI cycles using advanced machine learning techniques.

METHODS : This was a retrospective analysis of IVF/ICSI cycles performed from 2014 to 2016. In total, data of 3879 blastocysts were collected for the analysis. Patients underwent PGT-Aneuploidy analysis (PGT-A) at the Center for Reproductive Medicine of European Hospital, Rome, Italy have been included in the analysis. The method involved whole-genome amplification followed by array comparative genome hybridization. To model the rate of euploid blastocysts, the data were split into a train set (used to fit and calibrate the models) and a test set (used to assess models' predictive performance). Three different models were calibrated: a classical linear regression; a Gradient Boosted Tree (GBT) machine learning model; a model belonging to the Generalized Additive Models (GAM).

RESULTS : The present study confirms that female age, which is the strongest predictor of embryo chromosomal abnormalities, and blastocyst euploidy rate have a non-linear relationship, well depicted by the GBT and the GAM models. According to this model, the rate of reduction in the percentage of euploid blastocysts increases with age: the yearly relative variation is -10% at the age of 37 and -30% at the age of 45. Other factors including male age, female and male body mass index, fertilization rate and ovarian reserve may only marginally impact on embryo euploidy rate.

CONCLUSIONS : Female age is the strongest predictor of embryo chromosomal abnormalities and has a non-linear relationship with the blastocyst euploidy rate. Other factors related to both the male and female subjects may only minimally affect this outcome.

La Marca Antonio, Capuzzo Martina, Imbrogno Maria G, Donno Valeria, Spedicato Giorgio, Sacchi Sandro, Minasi Maria Giulia, Spinella Francesca, Greco Pierfrancesco, Fiorentino Francesco, Greco Ermanno

2020-Dec-11

Radiology

Machine-Learning Radiomics to Predict Early Recurrence in Perihilar Cholangiocarcinoma after Curative Resection.

In Liver international : official journal of the International Association for the Study of the Liver

BACKGROUND AND AIMS : Up to 40%-65% of patients with perihilar cholangiocarcinoma (PHC) rapidly progress to early recurrence (ER) even after curative resection. Quantification of ER risk is difficult and a reliable prognostic prediction tool is absent. We developed and validated a multilevel model, integrating clinicopathology, molecular pathology and radiology, especially radiomics coupled with machine-learning algorithms, to predict the ER of patients after curative resection in PHC.

METHODS : In total, 274 patients who underwent contrast-enhanced CT (CECT) and curative resection at 2-institutions were retrospectively identified and randomly divided into training (n=167), internal validation (n=70), and external validation (n=37) sets. A machine-learning analysis of 18,120 radiomic features based on multi-phase CECT and 48 clinico-radiologic characteristics was performed for the multilevel model.

RESULTS : Comprehensively, 7 independent factors (tumor differentiation, lymph node metastasis, preoperative CA19-9 level, enhancement pattern, A-Shrink score, V-Shrink score, P-Shrink score) were built to the multilevel model and quantified the risk of ER. We benchmarked the gain in discrimination with the area under the curve (AUC) of 0.883, superior to the rival clinical and radiomic models (AUCs 0.792-0.805). The accuracy (ACC) of the multilevel model was 0.826, which was significantly higher than those of the conventional staging systems (AJCC 8th (0.641), MSKCC (0.617), and Gazzaniga (0.581)).

CONCLUSION : The radiomics-based multilevel model demonstrated superior performance to rival models and conventional staging systems, and could serve as a visual prognostic tool to plan surveillance of ER and guide postoperative individualized management in PHC.

Qin Huan, Hu Xianling, Zhang Junfeng, Dai Haisu, He Yonggang, Zhao Zhiping, Yang Jiali, Xu Zhengrong, Hu Xiaofei, Chen Zhiyu

2020-Dec-11

early recurrence, machine learning, multilevel model, perihilar cholangiocarcinoma, radiomics

Public Health

Comparison of Use of Health Care Services and Spending for Unauthorized Immigrants vs Authorized Immigrants or US Citizens Using a Machine Learning Model.

In JAMA network open

Importance : Knowledge about use of health care services (health care utilization) and expenditures among unauthorized immigrant populations is uncertain because of limitations in ascertaining legal status in population data.

Objective : To examine health care utilization and expenditures that are attributable to unauthorized and authorized immigrants vs US-born individuals.

Design, Setting, and Participants : This cross-sectional study used the data on documentation status from the Los Angeles Family and Neighborhood Survey (LAFANS) to develop a random forest classifier machine learning model. K-fold cross-validation was used to test model performance. The LAFANS is a randomized, multilevel, in-person survey of households residing in Los Angeles County, California, consisting of 2 waves. Wave 1 began in April 2000 and ended in January 2002, and wave 2 began in August 2006 and ended in December 2008. The machine learning model was then applied to a nationally representative database, the 2016-2017 Medical Expenditure Panel Survey (MEPS), to predict health care expenditures and utilization among unauthorized and authorized immigrants and US-born individuals. A generalized linear model analyzed health care expenditures. Logistic regression modeling estimated dichotomous use of emergency department (ED), inpatient, outpatient, and office-based physician visits by immigrant groups with adjusting for confounding factors. Data were analyzed from May 1, 2019, to October 14, 2020.

Exposures : Self-reported immigration status (US-born, authorized, and unauthorized status).

Main Outcomes and Measures : Annual health care expenditures per capita and use of ED, outpatient, inpatient, and office-based physician care.

Results : Of 47 199 MEPS respondents with nonmissing data, 35 079 (74.3%) were US born, 10 816 (22.9%) were authorized immigrants, and 1304 (2.8%) were unauthorized immigrants (51.7% female; mean age, 47.6 [95% CI, 47.4-47.8] years). Compared with authorized immigrants and US-born individuals, unauthorized immigrants were more likely to be aged 18 to 44 years (80.8%), Latino (96.3%), and Spanish speaking (95.2%) and to have less than 12 years of education (53.7%). Half of unauthorized immigrants (47.1%) were uninsured compared with 15.9% of authorized immigrants and 6.0% of US-born individuals. Mean annual health care expenditures per person were $1629 (95% CI, $1330-$1928) for unauthorized immigrants, $3795 (95% CI, $3555-$4035) for authorized immigrants, and $6088 (95% CI, $5935-$6242) for US-born individuals.

Conclusions and Relevance : Contrary to much political discourse in the US, this cross-sectional study found no evidence that unauthorized immigrants are a substantial economic burden on safety net facilities such as EDs. This study illustrates the value of machine learning in the study of unauthorized immigrants using large-scale, secondary databases.

Wilson Fernando A, Zallman Leah, Pagán José A, Ortega Alexander N, Wang Yang, Tatar Moosa, Stimpson Jim P

2020-Dec-01

Public Health

Access denied: the shortage of digitized fitness resources for people with disabilities.

In Disability and rehabilitation ; h5-index 45.0

PURPOSE : The COVID-19 pandemic has drastically impacted every aspect of life, including how people exercise and access fitness resources. Prior to COVID-19, the global burden of disease attributable to sedentary behavior disproportionately affected the health of people with disabilities (PWD). This pre-existing gap has only widened during COVID-19 due to limited disability-friendly digital exercise resources. The purpose of this work is to examine this gap in accessibility to digital fitness resources, and re-frame the notion of accessibility to suit the contemporary context.

MATERIALS AND METHODS : Using machine learning, video titles/descriptions about home exercise ordered by relevance populated on YouTube between 1 January 2020 and 30 June 2020 were examined.

RESULTS : Using the search terms, "home exercise," "home-based exercise," "exercise no equipment," "workout no equipment," "exercise at home," or "at-home exercise," 700 videos ordered by relevance included 28 (4%) that were inclusive of participants with disabilities. Unfortunately, most digital fitness resources are therefore inaccessible to PWD. The global pause the pandemic has induced may be the right moment to construct a comprehensive, indexed digital library of home-based fitness video content for the disabled. There is a further need for more nuanced understandings of accessibility as technological advancements continue. Implications for Rehabilitation Physical activity is incredibly important to the quality of life and health of all people. Physical activity levels, however, remain lower among persons with disabilities. Access to disability-friendly resources remains a challenge and worsened by the circumstances of COVID-19 due to an apparent lack of digital fitness resources for persons with disabilities. A broader and comprehensive definition of accessibility must recognize digital advances and access to physical activity for persons with disabilities must feature digital resources.

Stratton Catherine, Kadakia Shevali, Balikuddembe Joseph K, Peterson Mark, Hajjioui Abderrazak, Cooper Rory, Hong Bo-Young, Pandiyan Uma, Muñoz-Velasco Laura Paulina, Joseph James, Krassioukov Andrei, Tripathi Deo Rishi, Tuakli-Wosornu Yetsa A

2020-Dec-11

Accessibility, digital resources, home exercise, inclusive, people with disabilities

General

Quantitative analysis of excipient dominated drug formulations by Raman spectroscopy combined with deep learning.

In Analytical methods : advancing methods and applications

Owing to the growing interest in the application of Raman spectroscopy for quantitative purposes in solid pharmaceutical preparations, an article on the identification of compositions in excipient dominated drugs based on Raman spectra is presented. We proposed label-free Raman spectroscopy in conjunction with deep learning (DL) and non-negative least squares (NNLS) as a solution to overcome the drug fast screening bottleneck, which is not only a great challenge to drug administration, but also a major scientific challenge linked to falsified and/or substandard medicines. The result showed that Raman spectroscopy remains a cost effective, rapid, and user-friendly method, which if combined with DL and NNLS leads to fast implantation in the identification of lactose dominated drug (LDD) formulations. Meanwhile, Raman spectroscopy with the peak matching method allows a visual interpretation of the spectral signature (presence or absence of active pharmaceutical ingredients (APIs) and low content APIs).

Fu Xiang, Zhong Li-Min, Cao Yong-Bing, Chen Hui, Lu Feng

2020-Dec-11

General

An Aggregated-Based Deep Learning Method for Leukemic B-lymphoblast Classification.

In Diagnostics (Basel, Switzerland)

Leukemia is a cancer of blood cells in the bone marrow that affects both children and adolescents. The rapid growth of unusual lymphocyte cells leads to bone marrow failure, which may slow down the production of new blood cells, and hence increases patient morbidity and mortality. Age is a crucial clinical factor in leukemia diagnosis, since if leukemia is diagnosed in the early stages, it is highly curable. Incidence is increasing globally, as around 412,000 people worldwide are likely to be diagnosed with some type of leukemia, of which acute lymphoblastic leukemia accounts for approximately 12% of all leukemia cases worldwide. Thus, the reliable and accurate detection of normal and malignant cells is of major interest. Automatic detection with computer-aided diagnosis (CAD) models can assist medics, and can be beneficial for the early detection of leukemia. In this paper, a single center study, we aimed to build an aggregated deep learning model for Leukemic B-lymphoblast classification. To make a reliable and accurate deep learner, data augmentation techniques were applied to tackle the limited dataset size, and a transfer learning strategy was employed to accelerate the learning process, and further improve the performance of the proposed network. The results show that our proposed approach was able to fuse features extracted from the best deep learning models, and outperformed individual networks with a test accuracy of 96.58% in Leukemic B-lymphoblast diagnosis.

Kasani Payam Hosseinzadeh, Park Sang-Won, Jang Jae-Won

2020-Dec-08

acute lymphoblastic leukemia, computer-aided diagnosis, deep learning, transfer learning

Oncology

A review on medical imaging synthesis using deep learning and its clinical applications.

In Journal of applied clinical medical physics ; h5-index 28.0

This paper reviewed the deep learning-based studies for medical imaging synthesis and its clinical application. Specifically, we summarized the recent developments of deep learning-based methods in inter- and intra-modality image synthesis by listing and highlighting the proposed methods, study designs, and reported performances with related clinical applications on representative studies. The challenges among the reviewed studies were then summarized with discussion.

Wang Tonghe, Lei Yang, Fu Yabo, Wynne Jacob F, Curran Walter J, Liu Tian, Yang Xiaofeng

2020-Dec-11

CT, MRI, PET, deep learning, image synthesis, radiation therapy

Radiology

Sub-2 mm Depth of Interaction Localization in PET Detectors with Prismatoid Light Guide Arrays and Single-Ended Readout using Convolutional Neural Networks.

In Medical physics ; h5-index 59.0

PURPOSE : Depth of interaction (DOI) readout in PET imaging has been researched in efforts to mitigate parallax error, which would enable the development of small diameter, high resolution PET scanners. However, DOI PET hasn't yet been commercialized due to the lack of practical, cost-effective and data efficient DOI readout methods. The rationale for this study was to develop a supervised machine learning algorithm for DOI estimation in PET that can be trained and deployed on unique sets of crystals.

METHODS : Depth collimated flood data was experimentally acquired using a Na-22 source with a depth-encoding single-ended readout Prism-PET module consisting of lutetium yttrium orthosilicate (LYSO) crystals coupled 4-to-1 to 3 x 3 mm² silicon photomultiplier (SiPM) pixels on one end and a prismatoid light guide array on the other end. A convolutional neural network (CNN) was trained to perform DOI estimation on data from center, edge and corner crystals in the Prism-PET module using (a) all 64 readout pixels and (b) only the 4 highest readout signals per event. CNN testing was performed on data from crystals not included in CNN training.

RESULTS : An average DOI resolution of 1.84 mm full width at half maximum (FWHM) across all crystals was achieved when using all 64 readout signals per event with the CNN compared to 3.04 mm FWHM DOI resolution using classical estimation. When using only the 4 highest signals per event, an average DOI resolution of 1.92 mm FWHM was achieved, representing only a 4% dropoff in CNN performance compared to using all 64 pixels per event.

CONCLUSIONS : Our CNN-based DOI estimation algorithm provides the best reported DOI resolution in a single-ended readout module and can be readily deployed on crystals not used for model training.

LaBella Andy, Cao Xinjie, Zeng Xinjie, Zhao Wei, Goldan Amir H

2020-Dec-11

CNN, DOI, Machine Learning, PET, Prism-PET

General

Leveraging heterogeneous network embedding for metabolic pathway prediction.

In Bioinformatics (Oxford, England)

MOTIVATION : Metabolic pathway reconstruction from genomic sequence information is a key step in predicting regulatory and functional potential of cells at the individual, population and community levels of organization. Although the most common methods for metabolic pathway reconstruction are gene-centric e.g. mapping annotated proteins onto known pathways using a reference database, pathway-centric methods based on heuristics or machine learning to infer pathway presence provide a powerful engine for hypothesis generation in biological systems. Such methods rely on rule sets or rich feature information that may not be known or readily accessible.

RESULTS : Here, we present pathway2vec, a software package consisting of six representational learning modules used to automatically generate features for pathway inference. Specifically, we build a three-layered network composed of compounds, enzymes and pathways, where nodes within a layer manifest inter-interactions and nodes between layers manifest betweenness interactions. This layered architecture captures relevant relationships used to learn a neural embedding-based low-dimensional space of metabolic features. We benchmark pathway2vec performance based on node-clustering, embedding visualization and pathway prediction using MetaCyc as a trusted source. In the pathway prediction task, results indicate that it is possible to leverage embeddings to improve prediction outcomes.

AVAILABILITY AND IMPLEMENTATION : The software package and installation instructions are published on http://github.com/pathway2vec.

CONTACT : shallam@mail.ubc.ca.

SUPPLEMENTARY INFORMATION : Supplementary data are available at Bioinformatics online.

M A Basher Abdur Rahman, Hallam Steven J

2020-Oct-20

Pathology

ECMarker: interpretable machine learning model identifies gene expression biomarkers predicting clinical outcomes and reveals molecular mechanisms of human disease in early stages.

In Bioinformatics (Oxford, England)

MOTIVATION : Gene expression and regulation, a key molecular mechanism driving human disease development, remains elusive, especially at early stages. Integrating the increasing amount of population-level genomic data and understanding gene regulatory mechanisms in disease development are still challenging. Machine learning has emerged to solve this, but many machine learning methods were typically limited to building an accurate prediction model as a 'black box', barely providing biological and clinical interpretability from the box.

RESULTS : To address these challenges, we developed an interpretable and scalable machine learning model, ECMarker, to predict gene expression biomarkers for disease phenotypes and simultaneously reveal underlying regulatory mechanisms. Particularly, ECMarker is built on the integration of semi- and discriminative-restricted Boltzmann machines, a neural network model for classification allowing lateral connections at the input gene layer. This interpretable model is scalable without needing any prior feature selection and enables directly modeling and prioritizing genes and revealing potential gene networks (from lateral connections) for the phenotypes. With application to the gene expression data of non-small-cell lung cancer patients, we found that ECMarker not only achieved a relatively high accuracy for predicting cancer stages but also identified the biomarker genes and gene networks implying the regulatory mechanisms in the lung cancer development. In addition, ECMarker demonstrates clinical interpretability as its prioritized biomarker genes can predict survival rates of early lung cancer patients (P-value < 0.005). Finally, we identified a number of drugs currently in clinical use for late stages or other cancers with effects on these early lung cancer biomarkers, suggesting potential novel candidates on early cancer medicine.

AVAILABILITYAND IMPLEMENTATION : ECMarker is open source as a general-purpose tool at https://github.com/daifengwanglab/ECMarker.

CONTACT : daifeng.wang@wisc.edu.

SUPPLEMENTARY INFORMATION : Supplementary data are available at Bioinformatics online.

Jin Ting, Nguyen Nam D, Talos Flaminia, Wang Daifeng

2020-Nov-06

General

Integration and Co-design of Memristive Devices and Algorithms for Artificial Intelligence.

In iScience

Memristive devices share remarkable similarities to biological synapses, dendrites, and neurons at both the physical mechanism level and unit functionality level, making the memristive approach to neuromorphic computing a promising technology for future artificial intelligence. However, these similarities do not directly transfer to the success of efficient computation without device and algorithm co-designs and optimizations. Contemporary deep learning algorithms demand the memristive artificial synapses to ideally possess analog weighting and linear weight-update behavior, requiring substantial device-level and circuit-level optimization. Such co-design and optimization have been the main focus of memristive neuromorphic engineering, which often abandons the "non-ideal" behaviors of memristive devices, although many of them resemble what have been observed in biological components. Novel brain-inspired algorithms are being proposed to utilize such behaviors as unique features to further enhance the efficiency and intelligence of neuromorphic computing, which calls for collaborations among electrical engineers, computing scientists, and neuroscientists.

Wang Wei, Song Wenhao, Yao Peng, Li Yang, Van Nostrand Joseph, Qiu Qinru, Ielmini Daniele, Yang J Joshua

2020-Dec-18

Computer Architecture, Hardware Co-design, Materials Science

Oncology

Combination of a Big Data Analytics Resource System With an Artificial Intelligence Algorithm to Identify Clinically Actionable Radiation Dose Thresholds for Dysphagia in Head and Neck Patients.

In Advances in radiation oncology

Purpose : We combined clinical practice changes, standardizations, and technology to automate aggregation, integration, and harmonization of comprehensive patient data from the multiple source systems used in clinical practice into a big data analytics resource system (BDARS). We then developed novel artificial intelligence algorithms, coupled with the BDARS, to identify structure dose volume histograms (DVH) metrics associated with dysphagia.

Methods and Materials : From the BDARS harmonized data of ≥22,000 patients, we identified 132 patients recently treated for head and neck cancer who also demonstrated dysphagia scores that worsened from base line to a maximum grade ≥2. We developed a method that used both physical and biologically corrected (α/β = 2.5) DVH curves to test both absolute and percentage volume based DVH metrics. Combining a statistical categorization algorithm with machine learning (SCA-ML) provided more extensive detailing of response threshold evidence than either approach alone. A sensitivity guided, minimum input, machine learning (ML) model was iteratively constructed to identify the key structure DVH metric thresholds.

Results : Seven swallowing structures producing 738 candidate DVH metrics were ranked for association with dysphagia using SCA-ML scoring. Structures included superior pharyngeal constrictor (SPC), inferior pharyngeal constrictor (IPC), larynx, and esophagus. Bilateral parotid and submandibular gland (SG) structures were categorized by relative mean dose (eg, SG_high, SG_low) as a dose versus tumor centric analog to contra and ipsilateral designations. Structure DVH metrics with high SCA-ML scores included the following: SPC: D20% (equivalent dose [EQD2] Gy) ≥47.7; SPC: D25% (Gy) ≥50.4; IPC: D35% (Gy) ≥61.7; parotid_low: D60% (Gy) ≥13.2; and SG_high: D35% (Gy) ≥61.7. Larynx: D25% (Gy) ≥21.2 and SG_low: D45% ≥28.2 had high SCA-ML scores but were segmented on less than 90% of plans. A model based on SPC: D20% (EQD2 Gy) alone had sensitivity and area under the curve of 0.88 ± 0.13 and 0.74 ± 0.17, respectively.

Conclusions : This study provides practical demonstration of combining big data with artificial intelligence to increase volume of evidence in clinical learning paradigms.

Mayo Charles S, Mierzwa Michelle, Moran Jean M, Matuszak Martha M, Wilkie Joel, Sun Grace, Yao John, Weyburn Grant, Anderson Carlos J, Owen Dawn, Rao Arvind

Oncology

Artificial Intelligence Research: The Utility and Design of a Relational Database System.

In Advances in radiation oncology

Although many researchers talk about a "patient database," they typically are not referring to a database at all, but instead to a spreadsheet of curated facts about a cohort of patients. This article describes relational database systems and how they differ from spreadsheets. At their core, spreadsheets are only capable of describing one-to-one (1:1) relationships. However, this article demonstrates that clinical medical data encapsulate numerous one-to-many relationships. Consequently, spreadsheets are very inefficient relative to relational database systems, which gracefully manage such data. Databases provide other advantages, in that the data fields are "typed" (that is, they contain specific kinds of data). This prevents users from entering spurious data during data import. Because each record contains a "key," it becomes impossible to add duplicate information (ie, add the same patient twice). Databases store data in very efficient ways, minimizing space and memory requirements on the host system. Likewise, databases can be queried or manipulated using a highly complex language called SQL. Consequently, it becomes trivial to cull large amounts of data from a vast number of data fields on very precise subsets of patients. Databases can be quite large (terabytes or more in size), yet still are highly efficient to query. Consequently, with the explosion of data available in electronic health records and other data sources, databases become increasingly important to contain or order these data. Ultimately, this will enable the clinical researcher to perform artificial intelligence analyses across vast amounts of clinical data in a way heretofore impossible. This article provides initial guidance in terms of creating a relational database system.

Dilling Thomas J

Oncology

Prediction of Survival and Recurrence Patterns by Machine Learning in Gastric Cancer Cases Undergoing Radiation Therapy and Chemotherapy.

In Advances in radiation oncology

Purpose : Radical surgery is the most important treatment modality in gastric cancer. Preoperative or postoperative radiation therapy (RT) and perioperative chemotherapy are the treatment options that should be added to surgery. This study aimed to evaluate the overall survival (OS) and recurrence patterns by machine learning in gastric cancer cases undergoing RT.

Methods and Materials : Between 2012 and 2019, the OS and recurrence patterns of 75 gastric cancer cases receiving RT ± chemotherapy at the Department of Radiation Oncology were evaluated by machine learning. Logistic regression, multilayer perceptron, XGBoost, support vector classification, random forest, and Gaussian Naive Bayes (GNB) algorithms were used to predict OS, hematogenous distant metastases, and peritoneal metastases. After the correlation analysis, the backward feature selection was performed as the variable selection method, and the variables with P values less than .005 were selected.

Results : Over the median 23-month follow-up, recurrence was seen in 33 cases, and 36 patients died. The median OS was 23 (min: 7; max: 82) months, and the disease-free survival was 18 (min: 5, max: 80) months. The most common recurrence pattern was hematogenous distant metastasis, followed by peritoneal metastasis. In this study, the most successful algorithms in the prediction of OS, distant metastases, and peritoneal metastases were found to be GNB with an accuracy of 81% (95% confidence interval [CI], 0.65-0.97, area under the curve [AUC]: 0.89), XGBoost with 86% accuracy (95% CI, 0.74-0.97, AUC: 0.86), and random forest with 97% accuracy (95% CI, 0.92-1.00, AUC: 0.97), respectively.

Conclusions : In gastric cancer, GNB, XGBoost, and random forest algorithms were determined to be the most successful algorithms for predicting OS, distant metastases, and peritoneal metastases, respectively. To determine the most accurate algorithm and perhaps make personalized treatments applicable, more precise machine learning studies are needed with an increased number of cases in the coming years.

Akcay Melek, Etiz Durmus, Celik Ozer

General

A machine learning-based model to estimate PM2.5 concentration levels in Delhi's atmosphere.

In Heliyon

During the last many years, the air quality of the capital city of India, Delhi had been hazardous. A large number of people have been diagnosed with Asthma and other breathing-related problems. The basic reason behind this has been the high concentration of life-threatening PM2.5 particles dissolved in its atmosphere. A good model, to forecast the concentration level of these dissolved particles, may help to prepare the residents with better prevention and safety strategies in order to save them from many health-related diseases. This work aims to forecast the PM2.5 concentration levels in various regions of Delhi on an hourly basis, by applying time series analysis and regression, based on various atmospheric and surface factors such as wind speed, atmospheric temperature, pressure, etc. The data for the analysis is obtained from various weather monitoring sites, set-up in the city, by the Indian Meteorological Department (IMD). A regression model is proposed, which uses Extra-Trees regression and AdaBoost, for further boosting. Experimentation for comparative study with the recent works is done and results indicate the efficacy of the proposed model.

Kumar Saurabh, Mishra Shweta, Singh Sunil Kumar

2020-Nov

Atmospheric pollution, Computer science, Machine learning, PM2.5 prediction, Regression, Time series analysis

General

A hotel's customers personal, behavioral, demographic, and geographic dataset from Lisbon, Portugal (2015-2018).

In Data in brief

This data article describes a hotel customer dataset with 31 variables describing a total of 83,590 instances (customers). It comprehends three full years of customer behavioral data. In addition to personal and behavioral information, the dataset also contains demographic and geographical information. This dataset contributes to reducing the lack of real-world business data that can be used for educational and research purposes. The dataset can be used in data mining, machine learning, and other analytical field problems in the scope of data science. Due to its unit of analysis, it is a dataset especially suitable for building customer segmentation models, including clustering and RFM (Recency, Frequency, and Monetary value) models, but also be used in classification and regression problems.

Antonio Nuno, de Almeida Ana, Nunes Luís

2020-Dec

Classification, Clustering, Data mining, Data science, Hospitality, Machine learning, RFM modeling, Regression

General

Prediction of Autoimmune Diseases by Targeted Metabolomic Assay of Urinary Organic Acids.

In Metabolites

Autoimmune diseases (ADs) are chronic disorders characterized by the loss of self-tolerance, and although being heterogeneous, they share common pathogenic mechanisms. Self-antigens and inflammation markers are established diagnostic tools; however, the metabolic imbalances that underlie ADs are poorly described. The study aimed to employ metabolomics for the detection of disease-related changes in autoimmune diseases that could have predictive value. Quantitative analysis of 28 urine organic acids was performed using Gas Chromatography-Mass Spectrometry in a group of 392 participants. Autoimmune thyroiditis, inflammatory bowel disease, psoriasis and rheumatoid arthritis were the most prevalent autoimmune diseases of the study. Statistically significant differences were observed in the tricarboxylate cycle metabolites, succinate, methylcitrate and malate, the pyroglutamate and 2-hydroxybutyrate from the glutathione cycle and the metabolites methylmalonate, 4-hydroxyphenylpyruvate, 2-hydroxyglutarate and 2-hydroxyisobutyrate between the AD group and the control. Artificial neural networks and Binary logistic regression resulted in the highest predictive accuracy scores (66.7% and 74.9%, respectively), while Methylmalonate, 2-Hydroxyglutarate and 2-hydroxybutyrate were proposed as potential biomarkers for autoimmune diseases. Urine organic acid levels related to the mechanisms of energy production and detoxification were associated with the presence of autoimmune diseases and could be an adjunct tool for early diagnosis and prediction.

Tsoukalas Dimitris, Fragoulakis Vassileios, Papakonstantinou Evangelos, Antonaki Maria, Vozikis Athanassios, Tsatsakis Aristidis, Buga Ana Maria, Mitroi Mihaela, Calina Daniela

2020-Dec-08

artificial intelligence, autoimmune diseases, disease prediction, glutathione cycle, metabolomics, organic acids, tricarboxylate cycle

Cardiology

Neural-Network-Based Diagnosis Using 3-Dimensional Myocardial Architecture and Deformation: Demonstration for the Differentiation of Hypertrophic Cardiomyopathy.

In Frontiers in cardiovascular medicine

The diagnosis of cardiomyopathy states may benefit from machine-learning (ML) based approaches, particularly to distinguish those states with similar phenotypic characteristics. Three-dimensional myocardial deformation analysis (3D-MDA) has been validated to provide standardized descriptors of myocardial architecture and deformation, and may therefore offer appropriate features for the training of ML-based diagnostic tools. We aimed to assess the feasibility of automated disease diagnosis using a neural network trained using 3D-MDA to discriminate hypertrophic cardiomyopathy (HCM) from its mimic states: cardiac amyloidosis (CA), Anderson-Fabry disease (AFD), and hypertensive cardiomyopathy (HTNcm). 3D-MDA data from 163 patients (mean age 53.1 ± 14.8 years; 68 females) with left ventricular hypertrophy (LVH) of known etiology was provided. Source imaging data was from cardiac magnetic resonance (CMR). Clinical diagnoses were as follows: 85 HCM, 30 HTNcm, 30 AFD, and 18 CA. A fully-connected-layer feed-forward neural was trained to distinguish HCM vs. other mimic states. Diagnostic performance was compared to threshold-based assessments of volumetric and strain-based CMR markers, in addition to baseline clinical patient characteristics. Threshold-based measures provided modest performance, the greatest area under the curve (AUC) being 0.70. Global strain parameters exhibited reduced performance, with AUC under 0.64. A neural network trained exclusively from 3D-MDA data achieved an AUC of 0.94 (sensitivity 0.92, specificity 0.90) when performing the same task. This study demonstrates that ML-based diagnosis of cardiomyopathy states performed exclusively from 3D-MDA is feasible and can distinguish HCM from mimic disease states. These findings suggest strong potential for computer-assisted diagnosis in clinical practice.

Satriano Alessandro, Afzal Yarmaghan, Sarim Afzal Muhammad, Fatehi Hassanabad Ali, Wu Cody, Dykstra Steven, Flewitt Jacqueline, Feuchter Patricia, Sandonato Rosa, Heydari Bobak, Merchant Naeem, Howarth Andrew G, Lydell Carmen P, Khan Aneal, Fine Nowell M, Greiner Russell, White James A

2020

cardiomyopathy, hypertrophic, machine learning, magnetic resonance, neural network, strain analysis

General

A Machine Learning Based Dose Prediction of Lutein Supplements for Individuals With Eye Fatigue.

In Frontiers in nutrition

Purpose: Nutritional intervention was always implemented based on "one-size-fits-all" recommendation instead of personalized strategy. We aimed to develop a machine learning based model to predict the optimal dose of a botanical combination of lutein ester, zeaxanthin, extracts of black currant, chrysanthemum, and goji berry for individuals with eye fatigue. Methods: 504 features, including demographic, anthropometrics, eye-related indexes, blood biomarkers, and dietary habits, were collected at baseline from 303 subjects in a randomized controlled trial. An aggregated score of visual health (VHS) was developed from total score of eye fatigue symptoms, visuognosis persistence, macular pigment optical density, and Schirmer test to represent an overall eye fatigue level. VHS at 45 days after intervention was predicted by XGBoost algorithm using all features at baseline to show the eye fatigue improvement. Optimal dose of the combination was chosen based on the predicted VHS. Results: After feature selection and parameter optimization, a model was trained and optimized with a Pearson's correlation coefficient of 0.649, 0.638, and 0.685 in training, test and validation set, respectively. After removing the features collected by invasive blood test and costly optical coherence tomography, the model remained good performance. Among 58 subjects in test and validation sets, 39 should take the highest dose as the optimal option, 17 might take a lower dose, while 2 could not benefit from the combination. Conclusion: We applied XGBoost algorithm to develop a model which could predict optimized dose of the combination to provide personalized nutrition solution for individuals with eye fatigue.

Kan Juntao, Li Ao, Zou Hong, Chen Liang, Du Jun

2020

XGBoost, dose prediction, eye fatigue, lutein supplements, machine learning

General

Prediction of Chronic Periodontitis Severity Using Machine Learning Models Based On Salivary Bacterial Copy Number.

In Frontiers in cellular and infection microbiology ; h5-index 53.0

Periodontitis is a widespread chronic inflammatory disease caused by interactions between periodontal bacteria and homeostasis in the host. We aimed to investigate the performance and reliability of machine learning models in predicting the severity of chronic periodontitis. Mouthwash samples from 692 subjects (144 healthy controls and 548 generalized chronic periodontitis patients) were collected, the genomic DNA was isolated, and the copy numbers of nine pathogens were measured using multiplex qPCR. The nine pathogens are as follows: Porphyromonas gingivalis (Pg), Tannerella forsythia (Tf), Treponema denticola (Td), Prevotella intermedia (Pi), Fusobacterium nucleatum (Fn), Campylobacter rectus (Cr), Aggregatibacter actinomycetemcomitans (Aa), Peptostreptococcus anaerobius (Pa), and Eikenella corrodens (Ec). By adding the species one by one in order of high accuracy to find the optimal combination of input features, we developed an algorithm that predicts the severity of periodontitis using four machine learning techniques. The accuracy was the highest when the models classified "healthy" and "moderate or severe" periodontitis (H vs. M-S, average accuracy of four models: 0.93, AUC = 0.96, sensitivity of 0.96, specificity of 0.81, and diagnostic odds ratio = 112.75). One or two red complex pathogens were used in three models to distinguish slight chronic periodontitis patients from healthy controls (average accuracy of 0.78, AUC = 0.82, sensitivity of 0.71, and specificity of 0.84, diagnostic odds ratio = 12.85). Although the overall accuracy was slightly reduced, the models showed reliability in predicting the severity of chronic periodontitis from 45 newly obtained samples. Our results suggest that a well-designed combination of salivary bacteria can be used as a biomarker for classifying between a periodontally healthy group and a chronic periodontitis group.

Kim Eun-Hye, Kim Seunghoon, Kim Hyun-Joo, Jeong Hyoung-Oh, Lee Jaewoong, Jang Jinho, Joo Ji-Young, Shin Yerang, Kang Jihoon, Park Ae Kyung, Lee Ju-Youn, Lee Semin

2020

chronic periodontitis, machine learning, multiplex qPCR, salivary bacterial copy number, severity prediction, slight periodontitis

Oncology

Dose Super-Resolution in Prostate Volumetric Modulated Arc Therapy Using Cascaded Deep Learning Networks.

In Frontiers in oncology

Purpose : This study proposes a cascaded network model for generating high-resolution doses (i.e., a 1 mm grid) from low-resolution doses (i.e., ≥3 mm grids) with reduced computation time.

Methods : Using the anisotropic analytical algorithm with three grid sizes (1, 3, and 5 mm) and the Acuros XB algorithm with two grid sizes (1 and 3 mm), dose distributions were calculated for volumetric modulated arc therapy plans for 73 prostate cancer patients. Our cascaded network model consisted of a hierarchically densely connected U-net (HD U-net) and a residual dense network (RDN), which were trained separately following a two-dimensional slice-by-slice procedure. The first network (HD U-net) predicted the downsampled high-resolution dose (generated through bicubic downsampling of the baseline high-resolution dose) using the low-resolution dose; subsequently, the second network (RDN) predicted the high-resolution dose from the output of the first network. Further, the predicted high-resolution dose was converted to its absolute value. We quantified the network performance using the spatial/dosimetric parameters (dice similarity coefficient, mean dose, maximum dose, minimum dose, homogeneity index, conformity index, and V_95%, V_70%, V_50%, and V_30%) for the low-resolution and predicted high-resolution doses relative to the baseline high-resolution dose. Gamma analysis (between the baseline dose and the low-resolution dose/predicted high-resolution dose) was performed with a 2%/2 mm criterion and 10% threshold.

Results : The average computation time to predict a high-resolution axial dose plane was <0.02 s. The dice similarity coefficient values for the predicted doses were closer to 1 when compared to those for the low-resolution doses. Most of the dosimetric parameters for the predicted doses agreed more closely with those for the baseline than for the low-resolution doses. In most of the parameters, no significant differences (p-value of >0.05) between the baseline and predicted doses were observed. The gamma passing rates for the predicted high-resolution does were higher than those for the low-resolution doses.

Conclusion : The proposed model accurately predicted high-resolution doses for the same dose calculation algorithm. Our model uses only dose data as the input without additional data, which provides advantages of convenience to user over other dose super-resolution methods.

Shin Dong-Seok, Kim Kyeong-Hyeon, Kang Sang-Won, Kang Seong-Hee, Kim Jae-Sung, Kim Tae-Ho, Kim Dong-Su, Cho Woong, Suh Tae Suk, Chung Jin-Beom

2020

cascaded networks, deep learning, dose grid size, dose super-resolution, prostate volumetric modulated arc therapy

General

How to apply the novel dynamic ARDL simulations (dynardl) and Kernel-based regularized least squares (krls).

In MethodsX

The application of dynamic Autoregressive Distributed Lag (dynardl) simulations and Kernel-based Regularized Least Squares (krls) to time series data is gradually gaining recognition in energy, environmental and health economics. The Kernel-based Regularized Least Squares technique is a simplified machine learning-based algorithm with strength in its interpretation and accounting for heterogeneity, additivity and nonlinear effects. The novel dynamic ARDL Simulations algorithm is useful for testing cointegration, long and short-run equilibrium relationships in both levels and differences. Advantageously, the novel dynamic ARDL Simulations has visualization interface to examine the possible counterfactual change in the desired variable based on the notion of ceteris paribus. Thus, the novel dynamic ARDL Simulations and Kernel-based Regularized Least Squares techniques are useful and improved time series techniques for policy formulation.•We customize ARDL and dynamic simulated ARDL by adding plot estimates with confidence intervals.•A step-by-step procedure of applying ARDL, dynamic ARDL Simulations and Kernel-based Regularized Least Squares is provided.•All techniques are applied to examine the economic effect of denuclearization in Switzerland by 2034.

Sarkodie Samuel Asumadu, Owusu Phebe Asantewaa

2020

Average marginal effects, Counterfactual change, Dynamic autoregressive distributed lag simulations, Dynardl, Impulse-Response, Kernel-based regularized least squares, Krls, Pointwise derivatives, Response surface regressions, time series techniques

General

Ultraelastic Yarns from Curcumin-Assisted ELD toward Wearable Human-Machine Interface Textiles.

In Advanced science (Weinheim, Baden-Wurttemberg, Germany)

Intelligent human-machine interfaces (HMIs) integrated wearable electronics are essential to promote the Internet of Things (IoT). Herein, a curcumin-assisted electroless deposition technology is developed for the first time to achieve stretchable strain sensing yarns (SSSYs) with high conductivity (0.2 Ω cm^-1) and ultralight weight (1.5 mg cm^-1). The isotropically deposited structural yarns can bear high uniaxial elongation (>1100%) and still retain low resistivity after 5000 continuous stretching-releasing cycles under 50% strain. Apart from the high flexibility enabled by helical loaded structure, a precise strain sensing function can be facilitated under external forces with metal-coated conductive layers. Based on the mechanics analysis, the strain sensing responses are scaled with the dependences on structural variables and show good agreements with the experimental results. The application of interfacial enhanced yarns as wearable logic HMIs to remotely control the robotic hand and manipulate the color switching of light on the basis of gesture recognition is demonstrated. It is hoped that the SSSYs strategy can shed an extra light in future HMIs development and incoming IoT and artificial intelligence technologies.

Zhu Chuang, Li Ruohao, Chen Xue, Chalmers Evelyn, Liu Xiaoteng, Wang Yuqi, Xu Ben Bin, Liu Xuqing

2020-Dec

curcumin, electroless deposition, human–machine interfaces, textiles, wearable electronics

General

A priori estimation of sequencing effort in complex microbial metatranscriptomes.

In Ecology and evolution

Metatranscriptome analysis or the analysis of the expression profiles of whole microbial communities has the additional challenge of dealing with a complex system with dozens of different organisms expressing genes simultaneously. An underlying issue for virtually all metatranscriptomic sequencing experiments is how to allocate the limited sequencing budget while guaranteeing that the libraries have sufficient depth to cover the breadth of expression of the community. Estimating the required sequencing depth to effectively sample the target metatranscriptome using RNA-seq is an essential first step to obtain robust results in subsequent analysis and to avoid overexpansion, once the information contained in the library reaches saturation. Here, we present a method to calculate the sequencing effort using a simulated series of metatranscriptomic/metagenomic matrices. This method is based on an extrapolation rarefaction curve using a Weibull growth model to estimate the maximum number of observed genes as a function of sequencing depth. This approach allowed us to compute the effort at different confidence intervals and to obtain an approximate a priori effort based on an initial fraction of sequences. The analytical pipeline presented here may be successfully used for the in-depth and time-effective characterization of complex microbial communities, representing a useful tool for the microbiome research community.

Monleon-Getino Toni, Frias-Lopez Jorge

2020-Dec

NGS, machine learning, metagenomics, metatranscriptomics, rarefaction curve, sample size, sequencing effort, simulation

General

Robust and simplified machine learning identification of pitfall trap-collected ground beetles at the continental scale.

In Ecology and evolution

Insect populations are changing rapidly, and monitoring these changes is essential for understanding the causes and consequences of such shifts. However, large-scale insect identification projects are time-consuming and expensive when done solely by human identifiers. Machine learning offers a possible solution to help collect insect data quickly and efficiently.Here, we outline a methodology for training classification models to identify pitfall trap-collected insects from image data and then apply the method to identify ground beetles (Carabidae). All beetles were collected by the National Ecological Observatory Network (NEON), a continental scale ecological monitoring project with sites across the United States. We describe the procedures for image collection, image data extraction, data preparation, and model training, and compare the performance of five machine learning algorithms and two classification methods (hierarchical vs. single-level) identifying ground beetles from the species to subfamily level. All models were trained using pre-extracted feature vectors, not raw image data. Our methodology allows for data to be extracted from multiple individuals within the same image thus enhancing time efficiency, utilizes relatively simple models that allow for direct assessment of model performance, and can be performed on relatively small datasets.The best performing algorithm, linear discriminant analysis (LDA), reached an accuracy of 84.6% at the species level when naively identifying species, which was further increased to >95% when classifications were limited by known local species pools. Model performance was negatively correlated with taxonomic specificity, with the LDA model reaching an accuracy of ~99% at the subfamily level. When classifying carabid species not included in the training dataset at higher taxonomic levels species, the models performed significantly better than if classifications were made randomly. We also observed greater performance when classifications were made using the hierarchical classification method compared to the single-level classification method at higher taxonomic levels.The general methodology outlined here serves as a proof-of-concept for classifying pitfall trap-collected organisms using machine learning algorithms, and the image data extraction methodology may be used for nonmachine learning uses. We propose that integration of machine learning in large-scale identification pipelines will increase efficiency and lead to a greater flow of insect macroecological data, with the potential to be expanded for use with other noninsect taxa.

Blair Jarrett, Weiser Michael D, Kaspari Michael, Miller Matthew, Siler Cameron, Marshall Katie E

2020-Dec

Carabidae, computer vision, insect sampling, machine learning, macroecology

General

Automated facial recognition for wildlife that lack unique markings: A deep learning approach for brown bears.

In Ecology and evolution

Emerging technologies support a new era of applied wildlife research, generating data on scales from individuals to populations. Computer vision methods can process large datasets generated through image-based techniques by automating the detection and identification of species and individuals. With the exception of primates, however, there are no objective visual methods of individual identification for species that lack unique and consistent body markings. We apply deep learning approaches of facial recognition using object detection, landmark detection, a similarity comparison network, and an support vector machine-based classifier to identify individuals in a representative species, the brown bear Ursus arctos. Our open-source application, BearID, detects a bear's face in an image, rotates and extracts the face, creates an "embedding" for the face, and uses the embedding to classify the individual. We trained and tested the application using labeled images of 132 known individuals collected from British Columbia, Canada, and Alaska, USA. Based on 4,674 images, with an 80/20% split for training and testing, respectively, we achieved a facial detection (ability to find a face) average precision of 0.98 and an individual classification (ability to identify the individual) accuracy of 83.9%. BearID and its annotated source code provide a replicable methodology for applying deep learning methods of facial recognition applicable to many other species that lack distinguishing markings. Further analyses of performance should focus on the influence of certain parameters on recognition accuracy, such as age and body size. Combining BearID with camera trapping could facilitate fine-scale behavioral research such as individual spatiotemporal activity patterns, and a cost-effective method of population monitoring through mark-recapture studies, with implications for species and landscape conservation and management. Applications to practical conservation include identifying problem individuals in human-wildlife conflicts, and evaluating the intrapopulation variation in efficacy of conservation strategies, such as wildlife crossings.

Clapham Melanie, Miller Ed, Nguyen Mary, Darimont Chris T

2020-Dec

deep learning, face recognition, grizzly bear, individual ID, machine learning, wildlife monitoring

General

The computational approaches of lncRNA identification based on coding potential: Status quo and challenges.

In Computational and structural biotechnology journal

Long noncoding RNAs (lncRNAs) make up a large proportion of transcriptome in eukaryotes, and have been revealed with many regulatory functions in various biological processes. When studying lncRNAs, the first step is to accurately and specifically distinguish them from the colossal transcriptome data with complicated composition, which contains mRNAs, lncRNAs, small RNAs and their primary transcripts. In the face of such a huge and progressively expanding transcriptome data, the in-silico approaches provide a practicable scheme for effectively and rapidly filtering out lncRNA targets, using machine learning and probability statistics. In this review, we mainly discussed the characteristics of algorithms and features on currently developed approaches. We also outlined the traits of some state-of-the-art tools for ease of operation. Finally, we pointed out the underlying challenges in lncRNA identification with the advent of new experimental data.

Li Jing, Zhang Xuan, Liu Changning

2020

Algorithm, Coding potential, Feature, In sillico, LncRNA identification, sORF

General

New Design Method for Fabricating Multilayer Membranes Using CO2-Assisted Polymer Compression Process.

In Molecules (Basel, Switzerland)

It was verified that deep learning can be used in creating multilayer membranes with multiple porosities using the CO₂-assisted polymer compression (CAPC) method. To perform training while reducing the number of experimental data as much as possible, the experimental data of the compression behavior of two layers were expanded to three layers for training, but sufficient accuracy could not be obtained. However, the accuracy was dramatically improved by adding the experimental data of the three layers. The possibility of only simulating process results without the necessity for a model is a merit unique to deep learning. Overall, in this study, the results show that by devising learning data, deep learning is extremely effective in designing multilayer membranes using the CAPC method.

Aizawa Takafumi

2020-Dec-08

CO2-assisted polymer compression, carbon dioxide, deep learning, multilayer porous membrane, process simulation

General

Engagement Enhancement Based on Human-in-the-Loop Optimization for Neural Rehabilitation.

In Frontiers in neurorobotics

Enhancing patients' engagement is of great benefit for neural rehabilitation. However, physiological and neurological differences among individuals can cause divergent responses to the same task, and the responses can further change considerably during training; both of these factors make engagement enhancement a challenge. This challenge can be overcome by training task optimization based on subjects' responses. To this end, an engagement enhancement method based on human-in-the-loop optimization is proposed in this paper. Firstly, an interactive speed-tracking riding game is designed as the training task in which four reference speed curves (RSCs) are designed to construct the reference trajectory in each generation. Each RSC is modeled using a piecewise function, which is determined by the starting velocity, transient time, and end velocity. Based on the parameterized model, the difficulty of the training task, which is a key factor affecting the engagement, can be optimized. Then, the objective function is designed with consideration to the tracking accuracy and the surface electromyogram (sEMG)-based muscle activation, and the physical and physiological responses of the subjects can consequently be evaluated simultaneously. Moreover, a covariance matrix adaption evolution strategy, which is relatively tolerant of both measurement noises and human adaptation, is used to generate the optimal parameters of the RSCs periodically. By optimization of the RSCs persistently, the objective function can be maximized, and the subjects' engagement can be enhanced. Finally, the performance of the proposed method is demonstrated by the validation and comparison experiments. The results show that both subjects' sEMG-based motor engagement and electroencephalography based neural engagement can be improved significantly and maintained at a high level.

Wang Jiaxing, Wang Weiqun, Ren Shixin, Shi Weiguo, Hou Zeng-Guang

2020

EEG based neural engagement, human-in-the-loop optimization, neural rehabilitation, sEMG based muscle activation, tracking accuracy

General

Learning Generative State Space Models for Active Inference.

In Frontiers in computational neuroscience

In this paper we investigate the active inference framework as a means to enable autonomous behavior in artificial agents. Active inference is a theoretical framework underpinning the way organisms act and observe in the real world. In active inference, agents act in order to minimize their so called free energy, or prediction error. Besides being biologically plausible, active inference has been shown to solve hard exploration problems in various simulated environments. However, these simulations typically require handcrafting a generative model for the agent. Therefore we propose to use recent advances in deep artificial neural networks to learn generative state space models from scratch, using only observation-action sequences. This way we are able to scale active inference to new and challenging problem domains, whilst still building on the theoretical backing of the free energy principle. We validate our approach on the mountain car problem to illustrate that our learnt models can indeed trade-off instrumental value and ambiguity. Furthermore, we show that generative models can also be learnt using high-dimensional pixel observations, both in the OpenAI Gym car racing environment and a real-world robotic navigation task. Finally we show that active inference based policies are an order of magnitude more sample efficient than Deep Q Networks on RL tasks.

Çatal Ozan, Wauthier Samuel, De Boom Cedric, Verbelen Tim, Dhoedt Bart

2020

active inference, deep learning, free energy, generative modeling, robotics

General

Blood Pressure Modulation With Low-Intensity Focused Ultrasound Stimulation to the Vagus Nerve: A Pilot Animal Study.

In Frontiers in neuroscience ; h5-index 72.0

Objective : For hypertensive individuals, their blood pressure (BP) is often managed by taking medications. However, antihypertensive drugs might cause adverse effects such as congestive heart failure and are ineffective in significant numbers of the hypertensive population. As an alternative method for hypertension management, non-drug devices-based neuromodulation approaches such as functional electrical stimulation (FES) have been proposed. The FES approach requires the implantation of a stimulator into the body. One recently emerging technique, called low-intensity focused ultrasound stimulation (FUS), has been proposed to non-invasively modulate neural activities. In this pilot study, the feasibility of adopting low-intensity FUS neuromodulation for BP regulation was investigated using animal models.

Methods : A FUS system was developed for BP modulation in rabbits. For each rabbit, the low-intensity FUS with different acoustic intensities was used to stimulate its exposed left vagus nerve, and the BP waveform was synchronously recorded in its right common carotid artery. The effects of the different FUS intensities on systolic blood pressure (SBP), diastolic blood pressure (DBP), mean blood pressure (MAP), and heart rate (HR) were extensively examined from the BP recordings.

Results : The results demonstrated that the proposed FUS method could successfully induce changes in SBP, DBP, MAP, and HR values. When increasing acoustic intensities, the values of SBP, DBP, and MAP would tend to decrease more substantially.

Conclusion : The findings of this study suggested that BP could be modulated through the FUS, which might provide a new way for non-invasive and non-drug management of hypertension.

Ji Ning, Lin Wan-Hua, Chen Fei, Xu Lisheng, Huang Jianping, Li Guanglin

2020

blood pressure management, hypertension animal study, low-intensity focused ultrasound stimulation, neuromodulation, vagus nerve

Radiology

Current and Potential Applications of Artificial Intelligence in Gastrointestinal Stromal Tumor Imaging.

In Contrast media & molecular imaging

The most common mesenchymal tumors are gastrointestinal stromal tumors (GISTs), which have malignant potential and can occur anywhere along the gastrointestinal system. Imaging methods are important and indispensable of GISTs in diagnosis, risk staging, therapy, and follow-up. The recommended imaging method for staging and follow-up is computed tomography (CT) according to current guidelines. Artificial intelligence (AI) applies and elaborates theses, procedures, modes, and utilization systems for simulating, enlarging, and stretching the intellectual capacity of humans. Recently, researchers have done a few studies to explore AI applications in GIST imaging. This article reviews the present AI studies in GISTs imaging, including preoperative diagnosis, risk stratification and prediction of prognosis, gene mutation, and targeted therapy response.

Yang Cai-Wei, Liu Xi-Jiao, Liu Si-Yun, Wan Shang, Ye Zheng, Song Bin

2020

Surgery

Identifying Diabetic Macular Edema and Other Retinal Diseases by Optical Coherence Tomography Image and Multiscale Deep Learning.

In Diabetes, metabolic syndrome and obesity : targets and therapy

Purpose : Diabetic Macular Edema has been one of the research hotspots all over the world. But as the global population continues to grow, the number of OCT images requiring manual analysis is becoming increasingly unaffordable. Medical images are often fuzzy due to the inherent physical processes of acquiring them. It is difficult for traditional algorithms to use low-quality data. And traditional algorithms usually only provide diagnostic results, which makes the reliability and interpretability of the model face challenges. To solve problem above, we proposed a more intuitive and robust diagnosis model with self-enhancement ability and clinical triage patients' ability.

Methods : We used 38,057 OCT images (Drusen, DME, CNV and Normal) to establish and evaluate the model. All data are OCT images of fundus retina. There were 37,457 samples in the training dataset and 600 samples in the validation dataset. In order to diagnose these images accurately, we propose a multiscale transfer learning algorithm. Firstly, the sample is sent to the automatic self-enhancement module for edge detection and enhancement. Then, the processed data are sent to the image diagnosis module to determine the disease type. This process makes more data more effective and can be accurately classified. Finally, we calculated the accuracy, precision, sensitivity and specificity of the model, and verified the performance of the model from the perspective of clinical application.

Results : The model proposed in this paper can provide the diagnosis results and display the detection targets more intuitively. The model reached 94.5% accuracy, 97.2% precision, 97.7% sensitivity and 97% specificity in the independent testing dataset.

Conclusion : Comparing the performance of relevant work and ablation test, our model achieved relatively good performance. It is proved that the model proposed in this paper has a stronger ability to recognize diseases even in the face of low-quality images. Experiment results also demonstrate its clinical referral capability. It can reduce the workload of medical staff and save the precious time of patients.

Zhang Quan, Liu Zhiang, Li Jiaxu, Liu Guohua

2020

assisted diagnostics, clinical triage, deep learning, diabetic macular edema, optical coherence tomography, retina diseases, self-reinforcing

General

Non-contact physiological monitoring of preterm infants in the Neonatal Intensive Care Unit.

In NPJ digital medicine

The implementation of video-based non-contact technologies to monitor the vital signs of preterm infants in the hospital presents several challenges, such as the detection of the presence or the absence of a patient in the video frame, robustness to changes in lighting conditions, automated identification of suitable time periods and regions of interest from which vital signs can be estimated. We carried out a clinical study to evaluate the accuracy and the proportion of time that heart rate and respiratory rate can be estimated from preterm infants using only a video camera in a clinical environment, without interfering with regular patient care. A total of 426.6 h of video and reference vital signs were recorded for 90 sessions from 30 preterm infants in the Neonatal Intensive Care Unit (NICU) of the John Radcliffe Hospital in Oxford. Each preterm infant was recorded under regular ambient light during daytime for up to four consecutive days. We developed multi-task deep learning algorithms to automatically segment skin areas and to estimate vital signs only when the infant was present in the field of view of the video camera and no clinical interventions were undertaken. We propose signal quality assessment algorithms for both heart rate and respiratory rate to discriminate between clinically acceptable and noisy signals. The mean absolute error between the reference and camera-derived heart rates was 2.3 beats/min for over 76% of the time for which the reference and camera data were valid. The mean absolute error between the reference and camera-derived respiratory rate was 3.5 breaths/min for over 82% of the time. Accurate estimates of heart rate and respiratory rate could be derived for at least 90% of the time, if gaps of up to 30 seconds with no estimates were allowed.

Villarroel Mauricio, Chaichulee Sitthichok, Jorge João, Davis Sara, Green Gabrielle, Arteta Carlos, Zisserman Andrew, McCormick Kenny, Watkinson Peter, Tarassenko Lionel

2019-Dec-12

General

Preventing inpatient falls with injuries using integrative machine learning prediction: a cohort study.

In NPJ digital medicine

Patient falls during hospitalization can lead to severe injuries and remain one of the most vexing patient-safety problems facing hospitals. They lead to increased medical care costs, lengthened hospital stays, more litigation, and even death. Existing methods and technology to address this problem mostly focus on stratifying inpatients at risk, without predicting fall severity or injuries. Here, a retrospective cohort study was designed and performed to predict the severity of inpatient falls, based on a machine learning classifier integrating multi-view ensemble learning and model-based missing data imputation method. As input, over two thousand inpatient fall patients' demographic characteristics, diagnoses, procedural data, and bone density measurements were retrieved from the HMH clinical data warehouse from two separate time periods. The predictive classifier developed based on multi-view ensemble learning with missing values (MELMV) outperformed other three baseline models; achieved a cross-validated AUC of 0.713 (95% CI, 0.701-0.725), an AUC of 0.808 (95% CI, 0.740-0.876) on the separate testing set. Our studies show the efficacy of integrative machine-learning based classifier model in dealing with multi-source patient data, which in this case delivers robust predictive performance on the severity of patient falls. The severe fall index provided by the MELMV classifier is calculated to identify inpatients who are at risk of having severe injuries if they fall, thus triggering additional steps of intervention to prevent a harmful fall, beyond the standard-of-care procedure for all high-risk fall patients.

Wang Lin, Xue Zhong, Ezeana Chika F, Puppala Mamta, Chen Shenyi, Danforth Rebecca L, Yu Xiaohui, He Tiancheng, Vassallo Mark L, Wong Stephen T C

2019-Dec-12

General

Machine learning to reveal hidden risk combinations for the trajectory of posttraumatic stress disorder symptoms.

In Scientific reports ; h5-index 158.0

The nature of the recovery process of posttraumatic stress disorder (PTSD) symptoms is multifactorial. The Massive Parallel Limitless-Arity Multiple-testing Procedure (MP-LAMP), which was developed to detect significant combinational risk factors comprehensively, was utilized to reveal hidden combinational risk factors to explain the long-term trajectory of the PTSD symptoms. In 624 population-based subjects severely affected by the Great East Japan Earthquake, 61 potential risk factors encompassing sociodemographics, lifestyle, and traumatic experiences were analyzed by MP-LAMP regarding combinational associations with the trajectory of PTSD symptoms, as evaluated by the Impact of Event Scale-Revised score after eight years adjusted by the baseline score. The comprehensive combinational analysis detected 56 significant combinational risk factors, including 15 independent variables, although the conventional bivariate analysis between single risk factors and the trajectory detected no significant risk factors. The strongest association was observed with the combination of short resting time, short walking time, unemployment, and evacuation without preparation (adjusted P value = 2.2 × 10^-4, and raw P value = 3.1 × 10^-9). Although short resting time had no association with the poor trajectory, it had a significant interaction with short walking time (P value = 1.2 × 10^-3), which was further strengthened by the other two components (P value = 9.7 × 10^-5). Likewise, components that were not associated with a poor trajectory in bivariate analysis were included in every observed significant risk combination due to their interactions with other components. Comprehensive combination detection by MP-LAMP is essential for explaining multifactorial psychiatric symptoms by revealing the hidden combinations of risk factors.

Takahashi Yuta, Yoshizoe Kazuki, Ueki Masao, Tamiya Gen, Zhiqian Yu, Utsumi Yusuke, Sakuma Atsushi, Tsuda Koji, Hozawa Atsushi, Tsuji Ichiro, Tomita Hiroaki

2020-Dec-10

General

PsychoAge and SubjAge: development of deep markers of psychological and subjective age using artificial intelligence.

In Aging ; h5-index 49.0

Aging clocks that accurately predict human age based on various biodata types are among the most important recent advances in biogerontology. Since 2016 multiple deep learning solutions have been created to interpret facial photos, omics data, and clinical blood parameters in the context of aging. Some of them have been patented to be used in commercial settings. However, psychological changes occurring throughout the human lifespan have been overlooked in the field of "deep aging clocks". In this paper, we present two deep learning predictors trained on social and behavioral data from Midlife in the United States (MIDUS) study: (a) PsychoAge, which predicts chronological age, and (b) SubjAge, which describes personal aging rate perception. Using 50 distinct features from the MIDUS dataset these models have achieved a mean absolute error of 6.7 years for chronological age and 7.3 years for subjective age. We also show that both PsychoAge and SubjAge are predictive of all-cause mortality risk, with SubjAge being a more significant risk factor. Both clocks contain actionable features that can be modified using social and behavioral interventions, which enables a variety of aging-related psychology experiment designs. The features used in these clocks are interpretable by human experts and may prove to be useful in shifting personal perception of aging towards a mindset that promotes productive and healthy behaviors.

Zhavoronkov Alex, Kochetov Kirill, Diamandis Peter, Mitina Maria

2020-Dec-08

aging clock, artificial intelligence, deep learning, psychology of aging, subjective age

General

Transforming task representations to perform novel tasks.

In Proceedings of the National Academy of Sciences of the United States of America

An important aspect of intelligence is the ability to adapt to a novel task without any direct experience (zero shot), based on its relationship to previous tasks. Humans can exhibit this cognitive flexibility. By contrast, models that achieve superhuman performance in specific tasks often fail to adapt to even slight task alterations. To address this, we propose a general computational framework for adapting to novel tasks based on their relationship to prior tasks. We begin by learning vector representations of tasks. To adapt to new tasks, we propose metamappings, higher-order tasks that transform basic task representations. We demonstrate the effectiveness of this framework across a wide variety of tasks and computational paradigms, ranging from regression to image classification and reinforcement learning. We compare to both human adaptability and language-based approaches to zero-shot learning. Across these domains, metamapping is successful, often achieving 80 to 90% performance, without any data, on a novel task, even when the new task directly contradicts prior experience. We further show that metamapping can not only generalize to new tasks via learned relationships, but can also generalize using novel relationships unseen during training. Finally, using metamapping as a starting point can dramatically accelerate later learning on a new task and reduce learning time and cumulative error substantially. Our results provide insight into a possible computational basis of intelligent adaptability and offer a possible framework for modeling cognitive flexibility and building more flexible artificial intelligence systems.

Lampinen Andrew K, McClelland James L

2020-Dec-10

artificial intelligence, cognitive science, transfer, zero-shot

Oncology

Investigating Serum and Tissue Expression Identified a Cytokine/Chemokine Signature as a Highly Effective Melanoma Marker.

In Cancers

The identification of reliable and quantitative melanoma biomarkers may help an early diagnosis and may directly affect melanoma mortality and morbidity. The aim of the present study was to identify effective biomarkers by investigating the expression of 27 cytokines/chemokines in melanoma compared to healthy controls, both in serum and in tissue samples. Serum samples were from 232 patients recruited at the IDI-IRCCS hospital. Expression was quantified by xMAP technology, on 27 cytokines/chemokines, compared to the control sera. RNA expression data of the same 27 molecules were obtained from 511 melanoma- and healthy-tissue samples, from the GENT2 database. Statistical analysis involved a 3-step approach: analysis of the single-molecules by Mann-Whitney analysis; analysis of paired-molecules by Pearson correlation; and profile analysis by the machine learning algorithm Support Vector Machine (SVM). Single-molecule analysis of serum expression identified IL-1b, IL-6, IP-10, PDGF-BB, and RANTES differently expressed in melanoma (p < 0.05). Expression of IL-8, GM-CSF, MCP-1, and TNF-α was found to be significantly correlated with Breslow thickness. Eotaxin and MCP-1 were found differentially expressed in male vs. female patients. Tissue expression analysis identified very effective marker/predictor genes, namely, IL-1Ra, IL-7, MIP-1a, and MIP-1b, with individual AUC values of 0.88, 0.86, 0.93, 0.87, respectively. SVM analysis of the tissue expression data identified the combination of these four molecules as the most effective signature to discriminate melanoma patients (AUC = 0.98). Validation, using the GEPIA2 database on an additional 1019 independent samples, fully confirmed these observations. The present study demonstrates, for the first time, that the IL-1Ra, IL-7, MIP-1a, and MIP-1b gene signature discriminates melanoma from control tissues with extremely high efficacy. We therefore propose this 4-molecule combination as an effective melanoma marker.

Cesati Marco, Scatozza Francesca, D’Arcangelo Daniela, Antonini-Cappellini Gian Carlo, Rossi Stefania, Tabolacci Claudio, Nudo Maurizio, Palese Enzo, Lembo Luigi, Di Lella Giovanni, Facchiano Francesco, Facchiano Antonio

2020-Dec-08

Support Vector Machine, cytokines, machine learning, melanoma markers, principal component analysis

Public Health

Access denied: the shortage of digitized fitness resources for people with disabilities.

In Disability and rehabilitation ; h5-index 45.0

PURPOSE : The COVID-19 pandemic has drastically impacted every aspect of life, including how people exercise and access fitness resources. Prior to COVID-19, the global burden of disease attributable to sedentary behavior disproportionately affected the health of people with disabilities (PWD). This pre-existing gap has only widened during COVID-19 due to limited disability-friendly digital exercise resources. The purpose of this work is to examine this gap in accessibility to digital fitness resources, and re-frame the notion of accessibility to suit the contemporary context.

MATERIALS AND METHODS : Using machine learning, video titles/descriptions about home exercise ordered by relevance populated on YouTube between 1 January 2020 and 30 June 2020 were examined.

RESULTS : Using the search terms, "home exercise," "home-based exercise," "exercise no equipment," "workout no equipment," "exercise at home," or "at-home exercise," 700 videos ordered by relevance included 28 (4%) that were inclusive of participants with disabilities. Unfortunately, most digital fitness resources are therefore inaccessible to PWD. The global pause the pandemic has induced may be the right moment to construct a comprehensive, indexed digital library of home-based fitness video content for the disabled. There is a further need for more nuanced understandings of accessibility as technological advancements continue. Implications for Rehabilitation Physical activity is incredibly important to the quality of life and health of all people. Physical activity levels, however, remain lower among persons with disabilities. Access to disability-friendly resources remains a challenge and worsened by the circumstances of COVID-19 due to an apparent lack of digital fitness resources for persons with disabilities. A broader and comprehensive definition of accessibility must recognize digital advances and access to physical activity for persons with disabilities must feature digital resources.

Stratton Catherine, Kadakia Shevali, Balikuddembe Joseph K, Peterson Mark, Hajjioui Abderrazak, Cooper Rory, Hong Bo-Young, Pandiyan Uma, Muñoz-Velasco Laura Paulina, Joseph James, Krassioukov Andrei, Tripathi Deo Rishi, Tuakli-Wosornu Yetsa A

2020-Dec-11

Accessibility, digital resources, home exercise, inclusive, people with disabilities

Pathology

Machine learning can predict disease manifestations and outcomes in lymphangioleiomyomatosis.

In The European respiratory journal

BACKGROUND : LAM is a rare multisystem disease with variable clinical manifestations and differing rates of progression that make management decisions and giving prognostic advice difficult. We used machine learning to identify clusters of associated features which could be used to stratify patients and predict outcomes in individuals.

PATIENTS AND METHODS : Using unsupervised machine learning we generated patient clusters using data from 173 women with LAM from the UK and 186 replication subjects from the NHLBI LAM registry. Prospective outcomes were associated with cluster results.

RESULTS : Two and three-cluster models were developed. A three-cluster model separated a large group of subjects presenting with dyspnoea or pneumothorax from a second cluster with a high prevalence of angiomyolipoma symptoms (p=0.0001) and TSC (p=0.041). The third cluster were older, never presented with dyspnoea or pneumothorax (p=0.0001) and had better lung function. Similar clusters were reproduced in the NHLBI cohort. Assigning patients to clusters predicted prospective outcomes: in a two-cluster model future risk of pneumothorax was 3.3 fold (95% C.I. 1.7-5.6) greater in cluster one than two (p=0.0002). Using the three-cluster model, the need for intervention for angiomyolipoma was lower in clusters two and three than cluster one (p<0.00001). In the NHLBI cohort, the incidence of death or lung transplant was much lower in clusters two and three (p=0.0045).

CONCLUSIONS : Machine learning has identified clinically relevant clusters associated with complications and outcome. Assigning individuals to clusters could improve decision making and prognostic information for patients.

Chernbumroong Saisakul, Johnson Janice, Gupta Nishant, Miller Suzanne, McCormack Francis X, Garibaldi Jonathan M, Johnson Simon R

2020-Dec-10

Radiology

Improved Glioma Grading Using Deep Convolutional Neural Networks.

In AJNR. American journal of neuroradiology

BACKGROUND AND PURPOSE : Accurate determination of glioma grade leads to improved treatment planning. The criterion standard for glioma grading is invasive tissue sampling. Recently, radiomic features have shown excellent potential in glioma-grade prediction. These features may not fully exploit the underlying information in MR images. The objective of this study was to investigate the performance of features learned by a convolutional neural network compared with standard radiomic features for grade prediction.

MATERIALS AND METHODS : A total of 237 patients with gliomas were included in this study. All images were resampled, registered, skull-stripped, and segmented to extract the tumors. The learned features from the trained convolutional neural network were used for grade prediction. The performance of the proposed method was compared with standard machine learning approaches, support vector machine, random forests, and gradient boosting trained with radiomic features.

RESULTS : The experimental results demonstrate that using learned features extracted from the convolutional neural network achieves an average accuracy of 87%, outperforming the methods considering radiomic features alone. The top-performing machine learning model is gradient boosting with an average accuracy of 64%. Thus, there is a 23% improvement in accuracy, and it is an efficient technique for grade prediction.

CONCLUSIONS : Convolutional neural networks are able to learn discriminating features automatically, and these features provide added value for grading gliomas. The proposed framework may provide substantial improvement in glioma-grade prediction; however, further validation is needed.

Gutta S, Acharya J, Shiroishi M S, Hwang D, Nayak K S

2020-Dec-10

General

Evidence of widespread, independent sequence signature for transcription factor cobinding.

In Genome research ; h5-index 99.0

Transcription factors (TFs) are the vocabulary that genomes use to regulate gene expression and phenotypes. The interactions among TFs enrich this vocabulary and orchestrate diverse biological processes. While simple models identify open chromatin and the presence of TF motifs as the two major contributors to TF binding patterns, it remains elusive what contributes to the in vivo TF cobinding landscape. In this study, we developed a machine learning algorithm to explore the contributors of the cobinding patterns. The algorithm substantially outperforms the state-of-the-field models for TF cobinding prediction. Game theory-based feature importance analysis reveals that, for most of the TF pairs we studied, independent motif sequences contribute more than at least one of the two TFs under investigation to their cobinding patterns. Such independent motif sequences include, but are not limited to, transcription initiation-related proteins and known TF complexes. We found the motif sequence signatures and the TFs are rarely mutual, corroborating a hierarchical and directional organization of the regulatory network and refuting the possibility of artifacts caused by shared sequence similarity with the TFs under investigation. We modeled such regulatory language with directed graphs, which reveal shared, global factors that are related to many binding and cobinding patterns.

Zhou Manqi, Li Hongyang, Wang Xueqing, Guan Yuanfang

2020-Dec-10

Surgery

Applications of deep learning in dentistry.

In Oral surgery, oral medicine, oral pathology and oral radiology ; h5-index 33.0

Over the last few years, translational applications of so-called artificial intelligence in the field of medicine have garnered a significant amount of interest. The present article aims to review existing dental literature that has examined deep learning, a subset of machine learning that has demonstrated the highest performance when applied to image processing and that has been tested as a formidable diagnostic support tool through its automated analysis of radiographic/photographic images. Furthermore, the article will critically evaluate the literature to describe potential methodological weaknesses of the studies and the need for further development. This review includes 28 studies that have described the applications of deep learning in various fields of dentistry. Research into the applications of deep learning in dentistry contains claims of its high accuracy. Nonetheless, many of these studies have substantial limitations and methodological issues (e.g., examiner reliability, the number of images used for training/testing, the methods used for validation) that have significantly limited the external validity of their results. Therefore, future studies that acknowledge the methodological limitations of existing literature will help to establish a better understanding of the usefulness of applying deep learning in dentistry.

Corbella Stefano, Srinivas Shanmukh, Cabitza Federico

2020-Nov-18

Pathology

An Artificial Intelligence-based Support Tool for Automation and Standardisation of Gleason Grading in Prostate Biopsies.

In European urology focus

BACKGROUND : Gleason grading is the standard diagnostic method for prostate cancer and is essential for determining prognosis and treatment. The dearth of expert pathologists, the inter- and intraobserver variability, as well as the labour intensity of Gleason grading all necessitate the development of a user-friendly tool for robust standardisation.

OBJECTIVE : To develop an artificial intelligence (AI) algorithm, based on machine learning and convolutional neural networks, as a tool for improved standardisation in Gleason grading in prostate cancer biopsies.

DESIGN, SETTING, AND PARTICIPANTS : A total of 698 prostate biopsy sections from 174 patients were used for training. The training sections were annotated by two senior consultant pathologists. The final algorithm was tested on 37 biopsy sections from 21 patients, with digitised slide images from two different scanners.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS : Correlation, sensitivity, and specificity parameters were calculated.

RESULTS AND LIMITATIONS : The algorithm shows high accuracy in detecting cancer areas (sensitivity: 100%, specificity: 68%). Compared with the pathologists, the algorithm also performed well in detecting cancer areas (intraclass correlation coefficient [ICC]: 0.99) and assigning the Gleason patterns correctly: Gleason patterns 3 and 4 (ICC: 0.96 and 0.94, respectively), and to a lesser extent, Gleason pattern 5 (ICC: 0.82). Similar results were obtained using two different scanners.

CONCLUSIONS : Our AI-based algorithm can reliably detect prostate cancer and quantify the Gleason patterns in core needle biopsies, with similar accuracy as pathologists. The results are reproducible on images from different scanners with a proven low level of intraobserver variability. We believe that this AI tool could be regarded as an efficient and interactive tool for pathologists.

PATIENT SUMMARY : We developed a sensitive artificial intelligence tool for prostate biopsies, which detects and grades cancer with similar accuracy to pathologists. This tool holds promise to improve the diagnosis of prostate cancer.

Marginean Felicia, Arvidsson Ida, Simoulis Athanasios, Christian Overgaard Niels, Åström Kalle, Heyden Anders, Bjartell Anders, Krzyzanowska Agnieszka

2020-Dec-07

Convolutional neural network, Deep learning, Prostate cancer, Machine learning

Radiology

Prediction of Microvascular Invasion of Hepatocellular Carcinoma Based on Preoperative Diffusion-Weighted MR Using Deep Learning.

In Academic radiology

RATIONALE AND OBJECTIVES : To investigate the value of diffusion-weighted magnetic resonance imaging for the prediction of microvascular invasion (MVI) of Hepatocellular Carcinoma (HCC) using Convolutional Neural Networks (CNN).

MATERIAL AND METHODS : This study was approved by the local institutional review board and the patients' informed consent was waived. Consecutive 97 subjects with 100 HCCs from July 2012 to October 2018 with surgical resection were retrieved. All subjects with diffusion-weighted imaging (DWI) examinations were performed with single-shot echo-planar imaging in a breath-hold routine. DWI parameters were three b values of 0,100,600 sec/mm². First, apparent diffusion coefficients (ADC) images were computed by mono-exponentially fitting the three b-value points. Then, multiple 2D axial patches (28 × 28) of HCCs from b0, b100, b600, and ADC images were extracted to increase the dataset for training the CNN model. Finally, the fusion of deep features derived from three b value images and ADC was conducted based on the CNN model for MVI prediction. The data set was split into the training set (60 HCCs) and the independent test set (40 HCCs). The output probability of the deep learning model in the MVI prediction of HCCs was assessed by the independent student's t-test for data following a normal distribution and Mann-Whitney U test for data violating the normal distribution. Receiver operating characteristic curve and area under the curve (AUC) were also used to assess the performance for MVI prediction of HCCs in the fixed test set.

RESULTS : Deep features in b600 images yielded better performance (AUC = 0.74, p = 0.004) for MVI prediction than b0 (AUC = 0.69, p = 0.023) and b100 (AUC = 0.734, p = 0.011). Comparatively, deep features in the ADC map obtained lower performance (AUC = 0.71, p = 0.012) than that of the higher b value images (b600) for MVI prediction. Furthermore, the fusion of deep features from the b0, b100, b600, and ADC images yielded the best results (AUC = 0.79, p = 0.002) for MVI prediction.

CONCLUSION : Fusion of deep features derived from DWI images concerning the three b-value images and the ADC image yields better performance for MVI prediction.

Wang Guangyi, Jian Wanwei, Cen Xiaoping, Zhang Lijuan, Guo Hui, Liu Zaiyi, Liang Changhong, Zhou Wu

2020-Dec-07

Deep learning, Diffusion-weighted MR, Hepatocellular carcinoma, Microvascular invasion

General

Glaucoma Detection from Raw SD-OCT Volumes: A Novel Approach Focused on Spatial Dependencies.

In Computer methods and programs in biomedicine

BACKGROUND AND OBJECTIVE : Glaucoma is the leading cause of blindness worldwide. Many studies based on fundus image and optical coherence tomography (OCT) imaging have been developed in the literature to help ophthalmologists through artificial-intelligence techniques. Currently, 3D spectral-domain optical coherence tomography (SD-OCT) samples have become more important since they could enclose promising information for glaucoma detection. To analyse the hidden knowledge of the 3D scans for glaucoma detection, we have proposed, for the first time, a deep-learning methodology based on leveraging the spatial dependencies of the features extracted from the B-scans.

METHODS : The experiments were performed on a database composed of 176 healthy and 144 glaucomatous SD-OCT volumes centred on the optic nerve head (ONH). The proposed methodology consists of two well-differentiated training stages: a slide-level feature extractor and a volume-based predictive model. The slide-level discriminator is characterised by two new, residual and attention, convolutional modules which are combined via skip-connections with other fine-tuned architectures. Regarding the second stage, we first carried out a data-volume conditioning before extracting the features from the slides of the SD-OCT volumes. Then, Long Short-Term Memory (LSTM) networks were used to combine the recurrent dependencies embedded in the latent space to provide a holistic feature vector, which was generated by the proposed sequential-weighting module (SWM).

RESULTS : The feature extractor reports AUC values higher than 0.93 both in the primary and external test sets. Otherwise, the proposed end-to-end system based on a combination of CNN and LSTM networks achieves an AUC of 0.8847 in the prediction stage, which outperforms other state-of-the-art approaches intended for glaucoma detection. Additionally, Class Activation Maps (CAMs) were computed to highlight the most interesting regions per B-scan when discerning between healthy and glaucomatous eyes from raw SD-OCT volumes.

CONCLUSIONS : The proposed model is able to extract the features from the B-scans of the volumes and combine the information of the latent space to perform a volume-level glaucoma prediction. Our model, which combines residual and attention blocks with a sequential weighting module to refine the LSTM outputs, surpass the results achieved from current state-of-the-art methods focused on 3D deep-learning architectures.

García Gabriel, Colomer Adrián, Naranjo Valery

2020-Nov-24

Convolutional attention blocks, Glaucoma detection, LSTM networks, Residual connections, SD-OCT volumes, Sequential-weighting module

General

Digital Gaming Interventions in Psychiatry: Evidence, Applications and Challenges.

In Psychiatry research ; h5-index 64.0

Human evolution has regularly intersected with technology. Digitalization of various services has brought a paradigm shift in consumerism. Treading this path, mental health practice has gradually moved to Digital Mental Health Interventions (DMHI), to improve service access and delivery. Applied games are one such innovation that has gained recent popularity in psychiatry. Based on the principles of gamification, they target psychosocial and cognitive domains, according to the deficits in various psychiatric disorders. They have been used to deliver cognitive behaviour therapy, cognitive training and rehabilitation, behavioural modification, social motivation, attention enhancement, and biofeedback. Research shows their utility in ADHD, autistic spectrum disorders, eating disorders, post-traumatic stress, impulse control disorders, depression, schizophrenia, dementia, and even healthy aging. Virtual reality and artificial intelligence have been used in conjunction with gaming interventions to improvise their scope. Even though these interventions hold promise in engagement, ease of use, reduction of stigma, and bridging the mental-health gap, there are pragmatic challenges, especially in developing countries. These include network quality, infrastructure, feasibility, socio-cultural adaptability, and potential for abuse. Keeping this in the background, this review summarizes the scope, promise, and evidence of digital gaming in psychiatric practice, and highlights the potential caveats in their implementation.

Vajawat Bhavika, Varshney Prateek, Banerjee Debanjan

2020-Nov-24

Gamification, digital games, gaming interventions, mental health, psychiatry, review

General

GCNG: graph convolutional networks for inferring gene interaction from spatial transcriptomics data.

In Genome biology ; h5-index 114.0

Most methods for inferring gene-gene interactions from expression data focus on intracellular interactions. The availability of high-throughput spatial expression data opens the door to methods that can infer such interactions both within and between cells. To achieve this, we developed Graph Convolutional Neural networks for Genes (GCNG). GCNG encodes the spatial information as a graph and combines it with expression data using supervised training. GCNG improves upon prior methods used to analyze spatial transcriptomics data and can propose novel pairs of extracellular interacting genes. The output of GCNG can also be used for downstream analysis including functional gene assignment.Supporting website with software and data: https://github.com/xiaoyeye/GCNG .

Yuan Ye, Bar-Joseph Ziv

2020-Dec-10

Extracellular gene interactions, Graph convolutional networks, Spatial transcriptomics

Public Health

Systematic evaluation of supervised machine learning for sample origin prediction using metagenomic sequencing data.

In Biology direct

BACKGROUND : The advent of metagenomic sequencing provides microbial abundance patterns that can be leveraged for sample origin prediction. Supervised machine learning classification approaches have been reported to predict sample origin accurately when the origin has been previously sampled. Using metagenomic datasets provided by the 2019 CAMDA challenge, we evaluated the influence of variable technical, analytical and machine learning approaches for result interpretation and novel source prediction.

RESULTS : Comparison between 16S rRNA amplicon and shotgun sequencing approaches as well as metagenomic analytical tools showed differences in normalized microbial abundance, especially for organisms present at low abundance. Shotgun sequence data analyzed using Kraken2 and Bracken, for taxonomic annotation, had higher detection sensitivity. As classification models are limited to labeling pre-trained origins, we took an alternative approach using Lasso-regularized multivariate regression to predict geographic coordinates for comparison. In both models, the prediction errors were much higher in Leave-1-city-out than in 10-fold cross validation, of which the former realistically forecasted the increased difficulty in accurately predicting samples from new origins. This challenge was further confirmed when applying the model to a set of samples obtained from new origins. Overall, the prediction performance of the regression and classification models, as measured by mean squared error, were comparable on mystery samples. Due to higher prediction error rates for samples from new origins, we provided an additional strategy based on prediction ambiguity to infer whether a sample is from a new origin. Lastly, we report increased prediction error when data from different sequencing protocols were included as training data.

CONCLUSIONS : Herein, we highlight the capacity of predicting sample origin accurately with pre-trained origins and the challenge of predicting new origins through both regression and classification models. Overall, this work provides a summary of the impact of sequencing technique, protocol, taxonomic analytical approaches, and machine learning approaches on the use of metagenomics for prediction of sample origin.

Chen Julie Chih-Yu, Tyler Andrea D

2020-Dec-10

CAMDA, Lasso regularization, Machine learning, MetaSUB, Metagenomics, Microbiome, Multiclass classification, Multivariate regression

General

New Interfaces and Approaches to Machine Learning When Classifying Gestures within Music.

In Entropy (Basel, Switzerland)

Interactive music uses wearable sensors (i.e., gestural interfaces-GIs) and biometric datasets to reinvent traditional human-computer interaction and enhance music composition. In recent years, machine learning (ML) has been important for the artform. This is because ML helps process complex biometric datasets from GIs when predicting musical actions (termed performance gestures). ML allows musicians to create novel interactions with digital media. Wekinator is a popular ML software amongst artists, allowing users to train models through demonstration. It is built on the Waikato Environment for Knowledge Analysis (WEKA) framework, which is used to build supervised predictive models. Previous research has used biometric data from GIs to train specific ML models. However, previous research does not inform optimum ML model choice, within music, or compare model performance. Wekinator offers several ML models. Thus, we used Wekinator and the Myo armband GI and study three performance gestures for piano practice to solve this problem. Using these, we trained all models in Wekinator and investigated their accuracy, how gesture representation affects model accuracy and if optimisation can arise. Results show that neural networks are the strongest continuous classifiers, mapping behaviour differs amongst continuous models, optimisation can occur and gesture representation disparately affects model mapping behaviour; impacting music practice.

Rhodes Chris, Allmendinger Richard, Climent Ricardo

2020-Dec-07

HCI, Myo, Wekinator, gestural interfaces, gesture representation, interactive machine learning, interactive music, music composition, optimisation, performance gestures

General

Assessing Mediterranean Diet Adherence with the Smartphone: The Medipiatto Project.

In Nutrients ; h5-index 86.0

The Mediterranean diet (MD) is regarded as a healthy eating pattern with beneficial effects both for the decrease of the risk for non-communicable diseases and also for body weight reduction. In the current manuscript, we propose an automated smartphone application which monitors and evaluates the user's adherence to MD using images of the food and drinks that they consume. We define a set of rules for automatic adherence estimation, which focuses on the main MD food groups. We use a combination of a convolutional neural network (CNN) and a graph convolutional network to detect the types of foods and quantities from the users' food images and the defined set of rules to evaluate the adherence to MD. Our experiments show that our system outperforms a basic CNN in terms of recognizing food items and estimating quantity and yields comparable results as experienced dietitians when it comes to overall MD adherence estimation. As the system is novel, these results are promising; however, there is room for improvement of the accuracy by gathering and training with more data and certain refinements can be performed such as re-defining the set of rules to also be able to be used for sub-groups of MD (e.g., vegetarian type of MD).

Vasiloglou Maria F, Lu Ya, Stathopoulou Thomai, Papathanail Ioannis, Fäh David, Ghosh Arindam, Baumann Manuel, Mougiakakou Stavroula

2020-Dec-07

Mediterranean diet, Mediterranean diet adherence, Mediterranean diet score, artificial intelligence, computer vision, machine learning, smartphone

General

A Multiscale Topographical Analysis Based on Morphological Information: The HEVC Multiscale Decomposition.

In Materials (Basel, Switzerland)

In this paper, we evaluate the effect of scale analysis as well as the filtering process on the performances of an original compressed-domain classifier in the field of material surface topographies classification. Each surface profile is multiscale analyzed by using a Gaussian Filter analyzing method to be decomposed into three multiscale filtered image types: Low-pass (LP), Band-pass (BP), and High-pass (HP) filtered versions, respectively. The complete set of filtered image data constitutes the collected database. First, the images are lossless compressed using the state-of-the art High-efficiency video coding (HEVC) video coding standard. Then, the Intra-Prediction Modes Histogram (IPHM) feature descriptor is computed directly in the compressed domain from each HEVC compressed image. Finally, we apply the IPHM feature descriptors as an input of a Support Vector Machine (SVM) classifier. SVM is introduced here to strengthen the performances of the proposed classification system thanks to the powerful properties of machine learning tools. We evaluate the proposed solution we called "HEVC Multiscale Decomposition" (HEVC-MD) on a huge database of nearly 42,000 multiscale topographic images. A simple preliminary version of the algorithm reaches an accuracy of 52%. We increase this accuracy to 70% by using the multiscale analysis of the high-frequency range HP filtered image data sets. Finally, we verify that considering only the highest-scale analysis of low-frequency range LP was more appropriate for classifying our six surface topographies with an accuracy of up to 81%. To compare these new topographical descriptors to those conventionally used, SVM is applied on a set of 34 roughness parameters defined on the International Standard GPS ISO 25178 (Geometrical Product Specification), and one obtains accuracies of 38%, 52%, 65%, and 57% respectively for Sa, multiscale Sa, 34 roughness parameters, and multiscale ones. Compared to conventional roughness descriptors, the HEVC-MD descriptors increase surfaces discrimination from 65% to 81%.

Eseholi Tarek, Coudoux François-Xavier, Corlay Patrick, Sadli Rahmad, Bigerelle Maxence

2020-Dec-07

high-efficiency video coding (HEVC), mechanical engineering, roughness analysis, support vector machine (SVM), surface roughness, texture feature descriptors, texture image classification

Public Health

Ruling In and Ruling Out COVID-19: Computing SARS-CoV-2 Infection Risk From Symptoms, Imaging and Test Data.

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : Assigning meaningful probabilities of SARS-CoV-2 infection risk presents a diagnostic challenge across the continuum of care.

OBJECTIVE : Develop and clinically validate an adaptable, personalized diagnostic model to assist clinicians in ruling-in and ruling-out COVID-19 in potential patients. This study compares the diagnostic performance of probabilistic, graphical, and machine-learning models against a previously published benchmark model.

METHODS : We integrated patient symptom and test data using machine learning and Bayesian inference to quantify individual patient risk of SARS-CoV-2 infection. We trained models with 100,000 simulated patient profiles based on thirteen symptoms, estimated local prevalence, imaging, and molecular diagnostic performance from published reports. We tested these models with consecutive patients who presented with a COVID-19 compatible illness at the University of California San Diego Medical Center over 14 days starting in March 2020.

RESULTS : We included 55 consecutive patients with fever (78%) or cough (77%) presenting for ambulatory (n=11) or hospital care (n=44). 51% (n=28) were female, 49% were age <60. Common comorbidities included diabetes (22%), hypertension (27%), cancer (16%) and cardiovascular disease (13%). 69% of these (n=38) were RT-PCR confirmed positive for SARS-CoV-2 infection, 11 had repeated negative nucleic acid testing and an alternate diagnosis. Bayesian inference network, distance metric-learning, and ensemble models discriminated between patients with SARS-CoV-2 infection and alternate diagnoses with sensitivities of 81.6 - 84.2%, specificities of 58.8 - 70.6%, and accuracies of 61.4 - 71.8%. After integrating imaging and laboratory test statistics with the predictions of the Bayesian inference network, changes in diagnostic uncertainty at each step in the simulated clinical evaluation process were highly sensitive to location, symptom, and diagnostic test choices.

CONCLUSIONS : Decision support models that incorporate symptoms and available test results can help providers diagnose SARS-CoV-2 infection in real world settings.

CLINICALTRIAL :

D’Ambrosia Christopher, Christensen Henrik, Aronoff-Spencer Eliah

2020-Nov-02

Radiology

Deep Learning CT Image Reconstruction in Clinical Practice.

In RoFo : Fortschritte auf dem Gebiete der Rontgenstrahlen und der Nuklearmedizin

BACKGROUND :  Computed tomography (CT) is a central modality in modern radiology contributing to diagnostic medicine in almost every medical subspecialty, but particularly in emergency services. To solve the inverse problem of reconstructing anatomical slice images from the raw output the scanner measures, several methods have been developed, with filtered back projection (FBP) and iterative reconstruction (IR) subsequently providing criterion standards. Currently there are new approaches to reconstruction in the field of artificial intelligence utilizing the upcoming possibilities of machine learning (ML), or more specifically, deep learning (DL).

METHOD :  This review covers the principles of present CT image reconstruction as well as the basic concepts of DL and its implementation in reconstruction. Subsequently commercially available algorithms and current limitations are being discussed.

RESULTS AND CONCLUSION :  DL is an ML method that utilizes a trained artificial neural network to solve specific problems. Currently two vendors are providing DL image reconstruction algorithms for the clinical routine. For these algorithms, a decrease in image noise and an increase in overall image quality that could potentially facilitate the diagnostic confidence in lesion conspicuity or may translate to dose reduction for given clinical tasks have been shown. One study showed equal diagnostic accuracy in the detection of coronary artery stenosis for DL reconstructed images compared to IR at higher image quality levels. Consequently, a lot more research is necessary and should aim at diagnostic superiority in the clinical context covering a broadness of pathologies to demonstrate the reliability of such DL approaches.

KEY POINTS :   · Following iterative reconstruction, there is a new approach to CT image reconstruction in the clinical routine using deep learning (DL) as a method of artificial intelligence.. · DL image reconstruction algorithms decrease image noise, improve image quality, and have potential to reduce radiation dose.. · Diagnostic superiority in the clinical context should be demonstrated in future trials..

CITATION FORMAT : · Arndt C, Güttler F, Heinrich A et al. Deep Learning CT Image Reconstruction in Clinical Practice. Fortschr Röntgenstr 2020; DOI: 10.1055/a-1248-2556.

Arndt Clemens, Güttler Felix, Heinrich Andreas, Bürckenmeyer Florian, Diamantis Ioannis, Teichgräber Ulf

2020-Dec-10

General

Classification of malignant tumors in breast ultrasound using a pretrained deep residual network model and support vector machine.

In Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society

In this study, a transfer learning method was utilized to recognize and classify benign and malignant breast tumors, using two-dimensional breast ultrasound (US) images, to decrease the effort expended by physicians and improve the quality of clinical diagnosis. The pretrained deep residual network model was utilized for image feature extraction from the convolutional layer of the trained network; whereas, the linear support vector machine (SVM), with a sequential minimal optimization solver, was used to classify the extracted feature. We used an image dataset with 2099 unlabeled two-dimensional breast US images, collected from 543 patients (benign: 302, malignant: 241). The classification performance yielded a sensitivity of 94.34 % and a specificity of 93.22 % for malignant images (Area under curve = 0.938). The positive and negative predictive values were 92.6 and 94.8, respectively. A comparison between the diagnosis made by the physician and the automated classification by a trained classifier, showed that the latter had significantly better outcomes. This indicates the potential applicability of the proposed approach that incorporates both the pretrained deep learning network and a well-trained classifier, to improve the quality and efficacy of clinical diagnosis.

Shia Wei-Chung, Chen Dar-Ren

2020-Nov-27

Computer-aided diagnosis, Deep residual network, Sequential minimal optimization, Support vector machine, Ultrasound imaging

Pathology

Machine learning techniques for mitoses classification.

In Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society

BACKGROUND : Pathologists analyze biopsy material at both the cellular and structural level to determine diagnosis and cancer stage. Mitotic figures are surrogate biomarkers of cellular proliferation that can provide prognostic information; thus, their precise detection is an important factor for clinical care. Convolutional Neural Networks (CNNs) have shown remarkable performance on several recognition tasks. Utilizing CNNs for mitosis classification may aid pathologists to improve the detection accuracy.

METHODS : We studied two state-of-the-art CNN-based models, ESPNet and DenseNet, for mitosis classification on six whole slide images of skin biopsies and compared their quantitative performance in terms of sensitivity, specificity, and F-score. We used raw RGB images of mitosis and non-mitosis samples with their corresponding labels as training input. In order to compare with other work, we studied the performance of these classifiers and two other architectures, ResNet and ShuffleNet, on the publicly available MITOS breast biopsy dataset and compared the performance of all four in terms of precision, recall, and F-score (which are standard for this data set), architecture, training time and inference time.

RESULTS : The ESPNet and DenseNet results on our primary melanoma dataset had a sensitivity of 0.976 and 0.968, and a specificity of 0.987 and 0.995, respectively, with F-scores of .968 and .976, respectively. On the MITOS dataset, ESPNet and DenseNet showed a sensitivity of 0.866 and 0.916, and a specificity of 0.973 and 0.980, respectively. The MITOS results using DenseNet had a precision of 0.939, recall of 0.916, and F-score of 0.927. The best published result on MITOS (Saha et al. 2018) reported precision of 0.92, recall of 0.88, and F-score of 0.90. In our architecture comparisons on MITOS, we found that DenseNet beats the others in terms of F-Score (DenseNet 0.927, ESPNet 0.890, ResNet 0.865, ShuffleNet 0.847) and especially Recall (DenseNet 0.916, ESPNet 0.866, ResNet 0.807, ShuffleNet 0.753), while ResNet and ESPNet have much faster inference times (ResNet 6 s, ESPNet 8 s, DenseNet 31 s). ResNet is faster than ESPNet, but ESPNet has a higher F-Score and Recall than ResNet, making it a good compromise solution.

CONCLUSION : We studied several state-of-the-art CNNs for detecting mitotic figures in whole slide biopsy images. We evaluated two CNNs on a melanoma cancer dataset and then compared four CNNs on a public breast cancer data set, using the same methodology on both. Our methodology and architecture for mitosis finding in both melanoma and breast cancer whole slide images has been thoroughly tested and is likely to be useful for finding mitoses in any whole slide biopsy images.

Nofallah Shima, Mehta Sachin, Mercan Ezgi, Knezevich Stevan, May Caitlin J, Weaver Donald, Witten Daniela, Elmore Joann G, Shapiro Linda

2020-Nov-27

Convolutional neural networks, Machine learning, Melanoma, Mitoses, Pathology

General

Mid-term prediction of at-fault crash driver frequency using fusion deep learning with city-level traffic violation data.

In Accident; analysis and prevention

Traffic violations and improper driving are behaviors that primarily contribute to traffic crashes. This study aimed to develop effective approaches for predicting at-fault crash driver frequency using only city-level traffic enforcement predictors. A fusion deep learning approach combining a convolution neural network (CNN) and gated recurrent units (GRU) was developed to compare predictive performance with one econometric approach, two machine learning approaches, and another deep learning approach. The performance comparison was conducted for (1) at-fault crash driver frequency prediction tasks and (2) city-level crash risk prediction tasks. The proposed CNN-GRU achieved remarkable prediction accuracy and outperformed other approaches, while the other approaches also exhibited excellent performances. The results suggest that effective prediction approaches and appropriate traffic safety measures can be developed by considering both crash frequency and crash risk prediction tasks. In addition, the accumulated local effects (ALE) plot was utilized to investigate the contribution of each traffic enforcement activity on traffic safety in a scenario of multicollinearity among predictors. The ALE plot illustrated a complex nonlinear relationship between traffic enforcement predictors and the response variable. These findings can facilitate the development of traffic safety measures and serve as a good foundation for further investigations and utilization of traffic violation data.

Wu Yuan-Wei, Hsu Tien-Pen

2020-Dec-07

At-fault crash driver, Deep learning, Traffic enforcement, Traffic violation

Radiology

Fully automatic framework for comprehensive coronary artery calcium scores analysis on non-contrast cardiac-gated CT scan: Total and vessel-specific quantifications.

In European journal of radiology ; h5-index 47.0

OBJECTIVES : To develop a fully automatic multiview shape constraint framework for comprehensive coronary artery calcium scores (CACS) quantification via deep learning on nonenhanced cardiac CT images.

METHODS : In this retrospective single-centre study, a multi-task deep learning framework was proposed to detect and quantify coronary artery calcification from CT images collected between October 2018 and March 2019. A total of 232 non-contrast cardiac-gated CT scans were retrieved and studied (80 % for model training and 20 % for testing). CACS results of testing datasets (n = 46), including Agatston score, calcium volume score, calcium mass score, were calculated fully automatically and manually at total and vessel-specific levels, respectively.

RESULTS : No significant differences were found in CACS quantification obtained using automatic or manual methods at total and vessel-specific levels (Agatston score: automatic 535.3 vs. manual 542.0, P = 0.993; calcium volume score: automatic 454.2 vs. manual 460.6, P = 0.990; calcium mass score: automatic 128.9 vs. manual 129.5, P = 0.992). Compared to the ground truth, the number of calcified vessels can be accurate recognized automatically (total: automatic 107 vs. manual 102, P = 0.125; left main artery: automatic 15 vs. manual 14, P = 1.000 ; left ascending artery: automatic 37 vs. manual 37, P = 1.000; left circumflex artery: automatic 22 vs. manual 20, P = 0.625; right coronary artery: automatic 33 vs. manual 31, P = 0.500). At the patient's level, there was no statistic difference existed in the classification of Agatston scoring (P = 0.317) and the number of calcified vessels (P = 0.102) between the automatic and manual results.

CONCLUSIONS : The proposed framework can achieve reliable and comprehensive quantification for the CACS, including the calcified extent and distribution indicators at both total and vessel-specific levels.

Zhang Nan, Yang Guang, Zhang Weiwei, Wang Wenjing, Zhou Zhen, Zhang Heye, Xu Lei, Chen Yundai

2020-Nov-24

Calcium, Coronary artery disease, Deep learning, Tomography, X-ray computed

Radiology

Estimating myelin-water content from anatomical and diffusion images using spatially undersampled myelin-water imaging through machine learning.

In NeuroImage ; h5-index 117.0

Myelin is vital for healthy neuronal development, and can therefore provide valuable information regarding neuronal maturation. Anatomical and diffusion weighted images (DWI) possess information related to the myelin content and the current study investigates whether quantitative myelin markers can be extracted from anatomical and DWI using neural networks. Thirteen volunteers (mean age 29y) are included, and for each subject, a residual neural network was trained using spatially undersampled reference myelin-water markers. The network is trained on a voxel-by-voxel basis, resulting in a large amount of training data for each volunteer. The inputs used are the anatomical contrasts (cT1w, cT2w), the standardized T1w/T2w ratio, estimates of the relaxation times (T1, T2) and their ratio (T1/T2), and common DWI metrics (FA, RD, MD, λ₁, λ₂, λ₃). Furthermore, to estimate the added value of the DWI metrics, neural networks were trained using either the combined set (DWI, T1w and T2w) or only the anatomical (T1w and T2w) images. The reconstructed myelin-water maps are in good agreement with the reference myelin-water content in terms of the coefficient of variation (CoV) and the intraclass correlation coefficient (ICC). A 6-fold undersampling using both anatomical and DWI metrics resulted in ICC = .68 and CoV = 5.9%. Moreover, using twice the training data (3-fold undersampling) resulted in an ICC that is comparable to the reproducibility of the myelin-water imaging itself (CoV = 5.5% vs. CoV = 6.7% and ICC = .74 vs ICC = .80). To achieve this, beside the T1w, T2w images, DWI is required. This preliminary study shows the potential of machine learning approaches to extract specific myelin-content from anatomical and diffusion-weighted scans.

Drenthen Gerhard S, Backes Walter H, Jansen Jacobus F A

2020-Dec-07

Neural networks, artificial intelligence, magnetic resonance imaging, myelin-water fraction

General

An integrated computational methodology with data-driven machine learning, molecular modeling and PBPK modeling to accelerate solid dispersion formulation design.

In European journal of pharmaceutics and biopharmaceutics : official journal of Arbeitsgemeinschaft fur Pharmazeutische Verfahrenstechnik e.V

Drug in solid dispersion (SD) takes advantage of fast and extended dissolution thus attains a higher bioavailability than the crystal form. However, current development of SD relies on a random large-scale formulation screening method with low efficiency. Current research aims to integrate various computational tools, including machine learning (ML), molecular dynamic (MD) simulation and physiologically based pharmacokinetic (PBPK) modeling, to accelerate the development of SD formulations. Firstly, based on a dataset consisting of 674 dissolution profiles of SD, the random forest algorithm was used to construct a classification model to distinguish two types of dissolution profiles "spring-and-parachute" and "maintain supersaturation", and a regression model to predict the time-dependent dissolution profiles. Both of the two prediction models showed good prediction performance. Moreover, feature importance was performed to help understand the key information that contributes to the model. After that, the vemurafenib (VEM) SD formulation in previous report were used as an example to validate the models. MD simulation was used to investigate the dissolution behavior of two SD formulations with two polymers (HPMCAS and Eudragit) at the molecular level. The results showed that HPMCAS-based formulation resulted in faster dissolution than the Eudragit formulation, which agreed with the reported experimental results. Finally, a PBPK model was constructed to accurately predict the human pharmacokinetic profile of the VEM-HPMCAS SD formulation. In conclusion, an combined computational tools have been developed to in silico predict formulation composition, in vitro release and in vivo absorption behavior of SD formulations. The integrated computational methodology will significantly facilitate pharmaceutical formulation development than the traditional trial-and-error approach in the laboratory.

Gao Hanlu, Wang Wei, Dong Jie, Ye Zhuyifan, Ouyang Defang

2020-Dec-07

dissolution profile, machine learning, molecular dynamics (MD) simulations, pharmacokinetic modeling, solid dispersion

Surgery

Circulating tumor cell detection and single-cell analysis using an integrated workflow based on ChimeraX® -i120 Platform: A prospective study.

In Molecular oncology

Circulating tumor cell (CTC) analysis holds great potential to be a non-invasive solution for clinical cancer management. A complete workflow that combined CTC detection and single-cell molecular analysis is required. We developed the ChimeraX^® -i120 platform to facilitate negative enrichment, immunofluorescent labeling, and machine learning-based identification of CTCs. Analytical performances were evaluated and a total of 477 participants were enrolled to validate the clinical feasibility of ChimeraX^® -i120 CTC detection. We analyzed copy number alteration profiles of isolated single cells. The ChimeraX^® -i120 platform had high sensitivity, accuracy, and reproducibility for CTC detection. In clinical samples, an average value of > 60% CTC-positive rate was found for five cancer types (i.e., liver, biliary duct, breast, colorectal, and lung), while CTCs were rarely identified in blood from healthy donors. In hepatocellular carcinoma patients treated with curative resection, CTC status was significantly associated with tumor characteristics, prognosis, and treatment response (all P<0.05). Single-cell sequencing analysis revealed that heterogeneous genomic alteration patterns resided in different cells, patients, and cancers. Our results suggest that the use of this ChimeraX^® -i120 platform and the integrated workflow has validity as a tool for CTC detection and downstream genomic profiling in the clinical setting.

Wang Peng-Xiang, Sun Yun-Fan, Jin Wei-Xiang, Cheng Jian-Wen, Peng Hai-Xiang, Xu Yang, Zhou Kai-Qian, Chen Li-Meng, Huang Kai, Wu Sui-Yi, Hu Bo, Zhang Ze-Fan, Guo Wei, Cao Ya, Zhou Jian, Fan Jia, Yang Xin-Rong

2020-Dec-10

Circulating tumor cell, Enumeration, Integrated platform, Liquid biopsy, Machine learning-based image recognition, Single-cell sequencing

General

Machine Learning-Based Pipeline for High Accuracy Bioparticle Sizing.

In Micromachines

High accuracy measurement of size is essential in physical and biomedical sciences. Various sizing techniques have been widely used in sorting colloidal materials, analyzing bioparticles and monitoring the qualities of food and atmosphere. Most imaging-free methods such as light scattering measure the averaged size of particles and have difficulties in determining non-spherical particles. Imaging acquisition using camera is capable of observing individual nanoparticles in real time, but the accuracy is compromised by the image defocusing and instrumental calibration. In this work, a machine learning-based pipeline is developed to facilitate a high accuracy imaging-based particle sizing. The pipeline consists of an image segmentation module for cell identification and a machine learning model for accurate pixel-to-size conversion. The results manifest a significantly improved accuracy, showing great potential for a wide range of applications in environmental sensing, biomedical diagnostical, and material characterization.

Luo Shaobo, Zhang Yi, Nguyen Kim Truc, Feng Shilun, Shi Yuzhi, Liu Yang, Hutchinson Paul, Chierchia Giovanni, Talbot Hugues, Bourouina Tarik, Jiang Xudong, Liu Ai Qun

2020-Dec-07

CCD, CMOS, machine learning, particle sizing, segmentation

Radiology

Artificial intelligence can predict the mortality of COVID-19 patients at the admission time using routine blood samples.

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : COVID-19, which is accompanied by acute respiratory distress, multiple organ failure, and death, has spread worldwide much faster than previously thought. However, at present, it has limited treatments.

OBJECTIVE : To overcome this issue, we developed an artificial intelligence (AI) model of COVID-19, named EDRnet, to predict in-hospital mortality using a routine blood sample at the time of hospital admission.

METHODS : We selected 28 blood biomarkers and used the age and gender information of patients as model inputs. To improve the mortality prediction, we adopted an ensemble approach combining a deep neural network and random forest model. We trained our model with a database of blood samples from 361 COVID-19 patients in Wuhan, China, and applied it to 106 COVID-19 patients in three Korean medical institutions.

RESULTS : In the testing datasets, EDRnet provided high sensitivity (100%), specificity (91.35%), and accuracy (91.51%). To extend the number of patient data, we developed a web application (http://beatcovid19.ml/), where anyone can access the model to predict the mortality and can register his or her own blood laboratory results.

CONCLUSIONS : Our new AI model, EDRnet, accurately predicts the mortality rate for COVID-19. It is publicly available and aims to help healthcare providers fight COVID1-19 and improve patients' outcome.

CLINICALTRIAL :

Ko Hoon, Chung Heewon, Kang Wu Seong, Park Chul, Kim Do Wan, Kim Seong Eun, Chung Chi Ryang, Ko Ryoung Eun, Lee Hooseok, Seo Jae Ho, Choi Tae-Young, Jaimes Rafael, Kim Kyung Won, Lee Jinseok

2020-Dec-08

Public Health

DrivAER: Identification of driving transcriptional programs in single-cell RNA sequencing data.

In GigaScience

BACKGROUND : Single-cell RNA sequencing (scRNA-seq) unfolds complex transcriptomic datasets into detailed cellular maps. Despite recent success, there is a pressing need for specialized methods tailored towards the functional interpretation of these cellular maps.

FINDINGS : Here, we present DrivAER, a machine learning approach for the identification of driving transcriptional programs using autoencoder-based relevance scores. DrivAER scores annotated gene sets on the basis of their relevance to user-specified outcomes such as pseudotemporal ordering or disease status. DrivAER iteratively evaluates the information content of each gene set with respect to the outcome variable using autoencoders. We benchmark our method using extensive simulation analysis as well as comparison to existing methods for functional interpretation of scRNA-seq data. Furthermore, we demonstrate that DrivAER extracts key pathways and transcription factors that regulate complex biological processes from scRNA-seq data.

CONCLUSIONS : By quantifying the relevance of annotated gene sets with respect to specified outcome variables, DrivAER greatly enhances our ability to understand the underlying molecular mechanisms.

Simon Lukas M, Yan Fangfang, Zhao Zhongming

2020-Dec-10

Autoencoder, machine learning, manifold interpretation, single-cell RNA sequencing, transcription factor

General

Interpretable PID parameter tuning for control engineering using general dynamic neural networks: An extensive comparison.

In PloS one ; h5-index 176.0

Modern automation systems largely rely on closed loop control, wherein a controller interacts with a controlled process via actions, based on observations. These systems are increasingly complex, yet most deployed controllers are linear Proportional-Integral-Derivative (PID) controllers. PID controllers perform well on linear and near-linear systems but their simplicity is at odds with the robustness required to reliably control complex processes. Modern machine learning techniques offer a way to extend PID controllers beyond their linear control capabilities by using neural networks. However, such an extension comes at the cost of losing stability guarantees and controller interpretability. In this paper, we examine the utility of extending PID controllers with recurrent neural networks--namely, General Dynamic Neural Networks (GDNN); we show that GDNN (neural) PID controllers perform well on a range of complex control systems and highlight how they can be a scalable and interpretable option for modern control systems. To do so, we provide an extensive study using four benchmark systems that represent the most common control engineering benchmarks. All control environments are evaluated with and without noise as well as with and without disturbances. The neural PID controller performs better than standard PID control in 15 of 16 tasks and better than model-based control in 13 of 16 tasks. As a second contribution, we address the lack of interpretability that prevents neural networks from being used in real-world control processes. We use bounded-input bounded-output stability analysis to evaluate the parameters suggested by the neural network, making them understandable for engineers. This combination of rigorous evaluation paired with better interpretability is an important step towards the acceptance of neural-network-based control approaches for real-world systems. It is furthermore an important step towards interpretable and safely applied artificial intelligence.

Günther Johannes, Reichensdörfer Elias, Pilarski Patrick M, Diepold Klaus

2020

Surgery

Mapping risk of ischemic heart disease using machine learning in a Brazilian state.

In PloS one ; h5-index 176.0

Cardiovascular diseases are the leading cause of deaths globally. Machine learning studies predicting mortality rates for ischemic heart disease (IHD) at the municipal level are very limited. The goal of this paper was to create and validate a Heart Health Care Index (HHCI) to predict risk of IHD based on location and risk factors. Secondary data, geographical information system (GIS) and machine learning were used to validate the HHCI and stratify the IHD municipality risk in the state of Paraná. A positive spatial autocorrelation was found (Moran's I = 0.6472, p-value = 0.001), showing clusters of high IHD mortality. The Support Vector Machine, which had an RMSE of 0.789 and error proportion close to one (0.867), was the best for prediction among eight machine learning algorithms after validation. In the north and northwest regions of the state, HHCI was low and mortality clusters patterns were high. By creating an HHCI through ML, we can predict IHD mortality rate at municipal level, identifying predictive characteristics that impact health conditions of these localities' guided health management decisions for improvements for IHD within the emergency care network in the state of Paraná.

Bergamini Marcela, Iora Pedro Henrique, Rocha Thiago Augusto Hernandes, Tchuisseu Yolande Pokam, Dutra Amanda de Carvalho, Scheidt João Felipe Herman Costa, Nihei Oscar Kenji, de Barros Carvalho Maria Dalva, Staton Catherine Ann, Vissoci João Ricardo Nickenig, de Andrade Luciano

2020

General

Comprehensive nutrient analysis in agricultural organic amendments through non-destructive assays using machine learning.

In PloS one ; h5-index 176.0

Portable X-ray fluorescence (pXRF) and Diffuse Reflectance Fourier Transformed Mid-Infrared (DRIFT-MIR) spectroscopy are rapid and cost-effective analytical tools for material characterization. Here, we provide an assessment of these methods for the analysis of total Carbon, Nitrogen and total elemental composition of multiple elements in organic amendments. We developed machine learning methods to rapidly quantify the concentrations of macro- and micronutrient elements present in the samples and propose a novel system for the quality assessment of organic amendments. Two types of machine learning methods, forest regression and extreme gradient boosting, were used with data from both pXRF and DRIFT-MIR spectroscopy. Cross-validation trials were run to evaluate generalizability of models produced on each instrument. Both methods demonstrated similar broad capabilities in estimating nutrients using machine learning, with pXRF being suitable for nutrients and contaminants. The results make portable spectrometry in combination with machine learning a scalable solution to provide comprehensive nutrient analysis for organic amendments.

Towett Erick K, Drake Lee B, Acquah Gifty E, Haefele Stephan M, McGrath Steve P, Shepherd Keith D

2020

General

Decoding the neural dynamics of free choice in humans.

In PLoS biology

How do we choose a particular action among equally valid alternatives? Nonhuman primate findings have shown that decision-making implicates modulations in unit firing rates and local field potentials (LFPs) across frontal and parietal cortices. Yet the electrophysiological brain mechanisms that underlie free choice in humans remain ill defined. Here, we address this question using rare intracerebral electroencephalography (EEG) recordings in surgical epilepsy patients performing a delayed oculomotor decision task. We find that the temporal dynamics of high-gamma (HG, 60-140 Hz) neural activity in distinct frontal and parietal brain areas robustly discriminate free choice from instructed saccade planning at the level of single trials. Classification analysis was applied to the LFP signals to isolate decision-related activity from sensory and motor planning processes. Compared with instructed saccades, free-choice trials exhibited delayed and longer-lasting HG activity during the delay period. The temporal dynamics of the decision-specific sustained HG activity indexed the unfolding of a deliberation process, rather than memory maintenance. Taken together, these findings provide the first direct electrophysiological evidence in humans for the role of sustained high-frequency neural activation in frontoparietal cortex in mediating the intrinsically driven process of freely choosing among competing behavioral alternatives.

Thiery Thomas, Saive Anne-Lise, Combrisson Etienne, Dehgan Arthur, Bastin Julien, Kahane Philippe, Berthoz Alain, Lachaux Jean-Philippe, Jerbi Karim

2020-Dec-10

Public Health

Ruling In and Ruling Out COVID-19: Computing SARS-CoV-2 Infection Risk From Symptoms, Imaging and Test Data.

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : Assigning meaningful probabilities of SARS-CoV-2 infection risk presents a diagnostic challenge across the continuum of care.

OBJECTIVE : Develop and clinically validate an adaptable, personalized diagnostic model to assist clinicians in ruling-in and ruling-out COVID-19 in potential patients. This study compares the diagnostic performance of probabilistic, graphical, and machine-learning models against a previously published benchmark model.

METHODS : We integrated patient symptom and test data using machine learning and Bayesian inference to quantify individual patient risk of SARS-CoV-2 infection. We trained models with 100,000 simulated patient profiles based on thirteen symptoms, estimated local prevalence, imaging, and molecular diagnostic performance from published reports. We tested these models with consecutive patients who presented with a COVID-19 compatible illness at the University of California San Diego Medical Center over 14 days starting in March 2020.

RESULTS : We included 55 consecutive patients with fever (78%) or cough (77%) presenting for ambulatory (n=11) or hospital care (n=44). 51% (n=28) were female, 49% were age <60. Common comorbidities included diabetes (22%), hypertension (27%), cancer (16%) and cardiovascular disease (13%). 69% of these (n=38) were RT-PCR confirmed positive for SARS-CoV-2 infection, 11 had repeated negative nucleic acid testing and an alternate diagnosis. Bayesian inference network, distance metric-learning, and ensemble models discriminated between patients with SARS-CoV-2 infection and alternate diagnoses with sensitivities of 81.6 - 84.2%, specificities of 58.8 - 70.6%, and accuracies of 61.4 - 71.8%. After integrating imaging and laboratory test statistics with the predictions of the Bayesian inference network, changes in diagnostic uncertainty at each step in the simulated clinical evaluation process were highly sensitive to location, symptom, and diagnostic test choices.

CONCLUSIONS : Decision support models that incorporate symptoms and available test results can help providers diagnose SARS-CoV-2 infection in real world settings.

CLINICALTRIAL :

D’Ambrosia Christopher, Christensen Henrik, Aronoff-Spencer Eliah

2020-Nov-02

Radiology

Artificial intelligence can predict the mortality of COVID-19 patients at the admission time using routine blood samples.

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : COVID-19, which is accompanied by acute respiratory distress, multiple organ failure, and death, has spread worldwide much faster than previously thought. However, at present, it has limited treatments.

OBJECTIVE : To overcome this issue, we developed an artificial intelligence (AI) model of COVID-19, named EDRnet, to predict in-hospital mortality using a routine blood sample at the time of hospital admission.

METHODS : We selected 28 blood biomarkers and used the age and gender information of patients as model inputs. To improve the mortality prediction, we adopted an ensemble approach combining a deep neural network and random forest model. We trained our model with a database of blood samples from 361 COVID-19 patients in Wuhan, China, and applied it to 106 COVID-19 patients in three Korean medical institutions.

RESULTS : In the testing datasets, EDRnet provided high sensitivity (100%), specificity (91.35%), and accuracy (91.51%). To extend the number of patient data, we developed a web application (http://beatcovid19.ml/), where anyone can access the model to predict the mortality and can register his or her own blood laboratory results.

CONCLUSIONS : Our new AI model, EDRnet, accurately predicts the mortality rate for COVID-19. It is publicly available and aims to help healthcare providers fight COVID1-19 and improve patients' outcome.

CLINICALTRIAL :

Ko Hoon, Chung Heewon, Kang Wu Seong, Park Chul, Kim Do Wan, Kim Seong Eun, Chung Chi Ryang, Ko Ryoung Eun, Lee Hooseok, Seo Jae Ho, Choi Tae-Young, Jaimes Rafael, Kim Kyung Won, Lee Jinseok

2020-Dec-08

General

Hybrid-COVID: a novel hybrid 2D/3D CNN based on cross-domain adaptation approach for COVID-19 screening from chest X-ray images.

In Physical and engineering sciences in medicine

The novel Coronavirus disease (COVID-19), which first appeared at the end of December 2019, continues to spread rapidly in most countries of the world. Respiratory infections occur primarily in the majority of patients treated with COVID-19. In light of the growing number of COVID-19 cases, the need for diagnostic tools to identify COVID-19 infection at early stages is of vital importance. For decades, chest X-ray (CXR) technologies have proven their ability to accurately detect respiratory diseases. More recently, with the availability of COVID-19 CXR scans, deep learning algorithms have played a critical role in the healthcare arena by allowing radiologists to recognize COVID-19 patients from their CXR images. However, the majority of screening methods for COVID-19 reported in recent studies are based on 2D convolutional neural networks (CNNs). Although 3D CNNs are capable of capturing contextual information compared to their 2D counterparts, their use is limited due to their increased computational cost (i.e. requires much extra memory and much more computing power). In this study, a transfer learning-based hybrid 2D/3D CNN architecture for COVID-19 screening using CXRs has been developed. The proposed architecture consists of the incorporation of a pre-trained deep model (VGG16) and a shallow 3D CNN, combined with a depth-wise separable convolution layer and a spatial pyramid pooling module (SPP). Specifically, the depth-wise separable convolution helps to preserve the useful features while reducing the computational burden of the model. The SPP module is designed to extract multi-level representations from intermediate ones. Experimental results show that the proposed framework can achieve reasonable performances when evaluated on a collected dataset (3 classes to be predicted: COVID-19, Pneumonia, and Normal). Notably, it achieved a sensitivity of 98.33%, a specificity of 98.68% and an overall accuracy of 96.91.

Bayoudh Khaled, Hamdaoui Fayçal, Mtibaa Abdellatif

2020-Dec-10

COVID-19, Chest X-ray, Deep learning, Hybrid 2D/3D CNN, Pneumonia

General

[Enhanced imaging in urological endoscopy].

In Der Urologe. Ausg. A

White light cystoscopy and the concise documentation of pathological findings are standard diagnostic procedures in urology. Additional imaging modalities and technical innovations may support clinicians in the detection of bladder tumors. Modern endoscopy systems provide ultra-high-resolution imaging and the option of digital contrast enhancement. Photodynamic diagnostics and narrow band imaging are well-established in clinical routine and have shown significant benefits in the detection of bladder cancer. By means of multispectral imaging, different modalities can now be combined in real-time. Probe-based procedures such as optical coherence tomography (OCT) or Raman spectroscopy can further contribute to advanced imaging through an "optical biopsy" which may primarily improve diagnostics in the upper urinary tract. The aim of all techniques is to optimize the detection rate in order to achieve a more accurate diagnosis, resection and lower recurrence rates. Current research projects aim to digitalize the documentation of endoscopy and also make it more patient- and user-friendly. In the future, the use of image processing and artificial intelligence may automatically support the surgeon during endoscopy.

Kriegmair M C, Hein S, Schoeb D S, Zappe H, Suárez-Ibarrola R, Waldbillig F, Gruene B, Pohlmann P-F, Praus F, Wilhelm K, Gratzke C, Miernik A, Bolenz C

2020-Dec-10

Bladder cancer, Cystscopy, Fluorescence, Photodynamic imaging, Urothelial carcinoma

Public Health

Assessment of Machine Learning to Estimate the Individual Treatment Effect of Corticosteroids in Septic Shock.

In JAMA network open

Importance : The survival benefit of corticosteroids in septic shock remains uncertain.

Objective : To estimate the individual treatment effect (ITE) of corticosteroids in adults with septic shock in intensive care units using machine learning and to evaluate the net benefit of corticosteroids when the decision to treat is based on the individual estimated absolute treatment effect.

Design, Setting, and Participants : This cohort study used individual patient data from 4 trials on steroid supplementation in adults with septic shock as a training cohort to model the ITE using an ensemble machine learning approach. Data from a double-blinded, placebo-controlled randomized clinical trial comparing hydrocortisone with placebo were used for external validation. Data analysis was conducted from September 2019 to February 2020.

Exposures : Intravenous hydrocortisone 50 mg dose every 6 hours for 5 to 7 days with or without enteral 50 μg of fludrocortisone daily for 7 days. The control was either the placebo or usual care.

Main Outcomes and Measures : All-cause 90-day mortality.

Results : A total of 2548 participants were included in the development cohort, with median (interquartile range [IQR]) age of 66 (55-76) years and 1656 (65.0%) men. The median (IQR) Simplified Acute Physiology Score (SAPS II) was 55 [42-69], and median (IQR) Sepsis-related Organ Failure Assessment score on day 1 was 11 (9-13). The crude pooled relative risk (RR) of death at 90 days was 0.89 (95% CI, 0.83 to 0.96) in favor of corticosteroids. According to the optimal individual model, the estimated median absolute risk reduction was of 2.90% (95% CI, 2.79% to 3.01%). In the external validation cohort of 75 patients, the area under the curve of the optimal individual model was 0.77 (95% CI, 0.59 to 0.92). For any number willing to treat (NWT; defined as the acceptable number of people to treat to avoid 1 additional outcome considering the risk of harm associated with the treatment) less than 25, the net benefit of treating all patients vs treating nobody was negative. When the NWT was 25, the net benefit was 0.01 for the treat all with hydrocortisone strategy, -0.01 for treat all with hydrocortisone and fludrocortisone strategy, 0.06 for the treat by SAPS II strategy, and 0.31 for the treat by optimal individual model strategy. The net benefit of the SAPS II and the optimal individual model treatment strategies converged to zero for a smaller number willing to treat, but the individual model was consistently superior than model based on the SAPS II score.

Conclusions and Relevance : These findings suggest that an individualized treatment strategy to decide which patient with septic shock to treat with corticosteroids yielded positive net benefit regardless of potential corticosteroid-associated side effects.

Pirracchio Romain, Hubbard Alan, Sprung Charles L, Chevret Sylvie, Annane Djillali

2020-Dec-01

Surgery

A Multi-mRNA Host-Response Molecular Blood Test for the Diagnosis and Prognosis of Acute Infections and Sepsis: Proceedings from a Clinical Advisory Panel.

In Journal of personalized medicine

Current diagnostics are insufficient for diagnosis and prognosis of acute infections and sepsis. Clinical decisions including prescription and timing of antibiotics, ordering of additional diagnostics and level-of-care decisions rely on understanding etiology and implications of a clinical presentation. Host mRNA signatures can differentiate infectious from noninfectious etiologies, bacterial from viral infections, and predict 30-day mortality. The 29-host-mRNA blood-based InSep^TM test (Inflammatix, Burlingame, CA, formerly known as HostDx^TM Sepsis) combines machine learning algorithms with a rapid point-of-care platform with less than 30 min turnaround time to enable rapid diagnosis of acute infections and sepsis, as well as prediction of disease severity. A scientific advisory panel including emergency medicine, infectious disease, intensive care and clinical pathology physicians discussed technical and clinical requirements in preparation of successful introduction of InSep into the market. Topics included intended use; patient populations of greatest need; patient journey and sample flow in the emergency department (ED) and beyond; clinical and biomarker-based decision algorithms; performance characteristics for clinical utility; assay and instrument requirements; and result readouts. The panel identified clear demand for a solution like InSep, requirements regarding test performance and interpretability, and a need for focused medical education due to the innovative but complex nature of the result readout. Innovative diagnostic solutions such as the InSep test could improve management of patients with suspected acute infections and sepsis in the ED, thereby lessening the overall burden of these conditions on patients and the healthcare system.

Ducharme James, Self Wesley H, Osborn Tiffany M, Ledeboer Nathan A, Romanowsky Jonathan, Sweeney Timothy E, Liesenfeld Oliver, Rothman Richard E

2020-Dec-07

acute infections, bacterial, diagnosis, emergency medicine, point-of-care, prognosis, sepsis, viral

General

Hybrid-COVID: a novel hybrid 2D/3D CNN based on cross-domain adaptation approach for COVID-19 screening from chest X-ray images.

In Physical and engineering sciences in medicine

The novel Coronavirus disease (COVID-19), which first appeared at the end of December 2019, continues to spread rapidly in most countries of the world. Respiratory infections occur primarily in the majority of patients treated with COVID-19. In light of the growing number of COVID-19 cases, the need for diagnostic tools to identify COVID-19 infection at early stages is of vital importance. For decades, chest X-ray (CXR) technologies have proven their ability to accurately detect respiratory diseases. More recently, with the availability of COVID-19 CXR scans, deep learning algorithms have played a critical role in the healthcare arena by allowing radiologists to recognize COVID-19 patients from their CXR images. However, the majority of screening methods for COVID-19 reported in recent studies are based on 2D convolutional neural networks (CNNs). Although 3D CNNs are capable of capturing contextual information compared to their 2D counterparts, their use is limited due to their increased computational cost (i.e. requires much extra memory and much more computing power). In this study, a transfer learning-based hybrid 2D/3D CNN architecture for COVID-19 screening using CXRs has been developed. The proposed architecture consists of the incorporation of a pre-trained deep model (VGG16) and a shallow 3D CNN, combined with a depth-wise separable convolution layer and a spatial pyramid pooling module (SPP). Specifically, the depth-wise separable convolution helps to preserve the useful features while reducing the computational burden of the model. The SPP module is designed to extract multi-level representations from intermediate ones. Experimental results show that the proposed framework can achieve reasonable performances when evaluated on a collected dataset (3 classes to be predicted: COVID-19, Pneumonia, and Normal). Notably, it achieved a sensitivity of 98.33%, a specificity of 98.68% and an overall accuracy of 96.91.

Bayoudh Khaled, Hamdaoui Fayçal, Mtibaa Abdellatif

2020-Dec-10

COVID-19, Chest X-ray, Deep learning, Hybrid 2D/3D CNN, Pneumonia

General

Analyses of Risk, Racial Disparity, and Outcomes Among US Patients With Cancer and COVID-19 Infection.

In JAMA oncology ; h5-index 85.0

Importance : Patients with specific cancers may be at higher risk than those without cancer for coronavirus disease 2019 (COVID-19) and its severe outcomes. At present, limited data are available on the risk, racial disparity, and outcomes for COVID-19 illness in patients with cancer.

Objectives : To investigate how patients with specific types of cancer are at risk for COVID-19 infection and its adverse outcomes and whether there are cancer-specific race disparities for COVID-19 infection.

Design, Setting, and Participants : This retrospective case-control analysis of patient electronic health records included 73.4 million patients from 360 hospitals and 317 000 clinicians across 50 US states to August 14, 2020. The odds of COVID-19 infections for 13 common cancer types and adverse outcomes were assessed.

Exposures : The exposure groups were patients diagnosed with a specific cancer, whereas the unexposed groups were patients without the specific cancer.

Main Outcomes and Measures : The adjusted odds ratio (aOR) and 95% CI were estimated using the Cochran-Mantel-Haenszel test for the risk of COVID-19 infection.

Results : Among the 73.4 million patients included in the analysis (53.6% female), 2 523 920 had at least 1 of the 13 common cancers diagnosed (all cancer diagnosed within or before the last year), and 273 140 had recent cancer (cancer diagnosed within the last year). Among 16 570 patients diagnosed with COVID-19, 1200 had a cancer diagnosis and 690 had a recent cancer diagnosis of at least 1 of the 13 common cancers. Those with recent cancer diagnosis were at significantly increased risk for COVID-19 infection (aOR, 7.14 [95% CI, 6.91-7.39]; P < .001), with the strongest association for recently diagnosed leukemia (aOR, 12.16 [95% CI, 11.03-13.40]; P < .001), non-Hodgkin lymphoma (aOR, 8.54 [95% CI, 7.80-9.36]; P < .001), and lung cancer (aOR, 7.66 [95% CI, 7.07-8.29]; P < .001) and weakest for thyroid cancer (aOR, 3.10 [95% CI, 2.47-3.87]; P < .001). Among patients with recent cancer diagnosis, African Americans had a significantly higher risk for COVID-19 infection than White patients; this racial disparity was largest for breast cancer (aOR, 5.44 [95% CI, 4.69-6.31]; P < .001), followed by prostate cancer (aOR, 5.10 [95% CI, 4.34-5.98]; P < .001), colorectal cancer (aOR, 3.30 [95% CI, 2.55-4.26]; P < .001), and lung cancer (aOR, 2.53 [95% CI, 2.10-3.06]; P < .001). Patients with cancer and COVID-19 had significantly worse outcomes (hospitalization, 47.46%; death, 14.93%) than patients with COVID-19 without cancer (hospitalization, 24.26%; death, 5.26%) (P < .001) and patients with cancer without COVID-19 (hospitalization, 12.39%; death, 4.03%) (P < .001).

Conclusions and Relevance : In this case-control study, patients with cancer were at significantly increased risk for COVID-19 infection and worse outcomes, which was further exacerbated among African Americans. These findings highlight the need to protect and monitor patients with cancer as part of the strategy to control the pandemic.

Wang QuanQiu, Berger Nathan A, Xu Rong

2020-Dec-10

General

Clinical significance of combining salivary mRNAs and carcinoembryonic antigen for ovarian cancer detection.

In Scandinavian journal of clinical and laboratory investigation ; h5-index 22.0

Salivary mRNA biomarkers and serum carcinoembryonic antigen (CEA) have been recognized as promising liquid biopsy methods for detection of multiple cancers. However, current tests normally use solitary type of biomarkers, and are limited by unsatisfactory sensitivity and specificity when applied to differentiate cancer patients from healthy controls. In this study, a combined approach of CEA and salivary mRNA biomarkers was evaluated for discriminatory performance of ovarian cancer patients from healthy controls. We designed our study with two phases: a discovery phase to find and evaluate multiple biomarkers, and an independent validation phase to confirm the applicability of the selected biomarkers. In the discovery phase, a total of 140 ovarian cancer patients and 140 healthy controls were recruited. The CEA level in blood as well as five mRNA biomarkers in saliva (i.e. AGPAT1, B2M, BASP1, IER3 and IL1β) were measured, followed by developing a machine-learning model to differentiate ovarian cancer patients and healthy controls. We found a novel panel of biomarkers, which could differentiate ovarian cancer patients from healthy controls with high sensitivity (89.3%) and high specificity (82.9%). Next, we applied this panel of biomarkers in an independent validation study that consisted of 60 ovarian cancer patients and 60 healthy controls. The ovarian cancer patients were successfully differentiated from healthy controls in the validation phase, with sensitivity reaching 85.0% and specificity reaching 88.3%. To our best knowledge, it is the first time that a combined use of CEA and salivary mRNA biomarkers were applied for non-invasive detection of ovarian cancer.

Yang Jinfang, Xiang Cuiping, Liu Jianmeng

2020-Dec-10

CEA, Liquid biopsy, blood, cancer, mRNA, ovarian, saliva

General

Using supervised machine learning on neuropsychological data to distinguish OCD patients with and without sensory phenomena from healthy controls.

In The British journal of clinical psychology

OBJECTIVES : While theoretical models link obsessive-compulsive disorder (OCD) with executive function deficits, empirical findings from the neuropsychological literature remain mixed. These inconsistencies are likely exacerbated by the challenge of high-dimensional data (i.e., many variables per subject), which is common across neuropsychological paradigms and necessitates analytical advances. More unique to OCD is the heterogeneity of symptom presentations, each of which may relate to distinct neuropsychological features. While researchers have traditionally attempted to account for this heterogeneity using a symptom-based approach, an alternative involves focusing on underlying symptom motivations. Although the most studied symptom motivation involves fear of harmful events, 60-70% of patients also experience sensory phenomena, consisting of uncomfortable sensations or perceptions that drive compulsions. Sensory phenomena have received limited attention in the neuropsychological literature, despite evidence that symptoms motivated by these experiences may relate to distinct cognitive processes.

METHODS : Here, we used a supervised machine learning approach to characterize neuropsychological processes in OCD, accounting for sensory phenomena.

RESULTS : Compared to logistic regression and other algorithms, random forest best differentiated healthy controls (n = 59; balanced accuracy = .70), patients with sensory phenomena (n = 29; balanced accuracy = .59), and patients without sensory phenomena (n = 46; balanced accuracy = .62). Decision-making best distinguished between groups based on sensory phenomena, and among the patient subsample, those without sensory phenomena uniquely displayed greater risk sensitivity compared to healthy controls (d = .07, p = .008).

CONCLUSIONS : Results suggest that different cognitive profiles may characterize patients motivated by distinct drives. The superior performance and generalizability of the newer algorithms highlights the utility of considering multiple analytic approaches when faced with complex data.

PRACTITIONER POINTS : Practitioners should be aware that sensory phenomena are common experiences among patients with OCD. OCD patients with sensory phenomena may be distinguished from those without based on neuropsychological processes.

Stamatis Caitlin A, Batistuzzo Marcelo C, Tanamatis Tais, Miguel Euripedes C, Hoexter Marcelo Q, Timpano Kiara R

2020-Dec-10

executive function, machine learning, neuropsychology, obsessive-compulsive disorder, sensory phenomena

General

Recalibrating the epigenetic clock: implications for assessing biological age in the human cortex.

In Brain : a journal of neurology

Human DNA methylation data have been used to develop biomarkers of ageing, referred to as 'epigenetic clocks', which have been widely used to identify differences between chronological age and biological age in health and disease including neurodegeneration, dementia and other brain phenotypes. Existing DNA methylation clocks have been shown to be highly accurate in blood but are less precise when used in older samples or in tissue types not included in training the model, including brain. We aimed to develop a novel epigenetic clock that performs optimally in human cortex tissue and has the potential to identify phenotypes associated with biological ageing in the brain. We generated an extensive dataset of human cortex DNA methylation data spanning the life course (n = 1397, ages = 1 to 108 years). This dataset was split into 'training' and 'testing' samples (training: n = 1047; testing: n = 350). DNA methylation age estimators were derived using a transformed version of chronological age on DNA methylation at specific sites using elastic net regression, a supervised machine learning method. The cortical clock was subsequently validated in a novel independent human cortex dataset (n = 1221, ages = 41 to 104 years) and tested for specificity in a large whole blood dataset (n = 1175, ages = 28 to 98 years). We identified a set of 347 DNA methylation sites that, in combination, optimally predict age in the human cortex. The sum of DNA methylation levels at these sites weighted by their regression coefficients provide the cortical DNA methylation clock age estimate. The novel clock dramatically outperformed previously reported clocks in additional cortical datasets. Our findings suggest that previous associations between predicted DNA methylation age and neurodegenerative phenotypes might represent false positives resulting from clocks not robustly calibrated to the tissue being tested and for phenotypes that become manifest in older ages. The age distribution and tissue type of samples included in training datasets need to be considered when building and applying epigenetic clock algorithms to human epidemiological or disease cohorts.

Shireby Gemma L, Davies Jonathan P, Francis Paul T, Burrage Joe, Walker Emma M, Neilson Grant W A, Dahir Aisha, Thomas Alan J, Love Seth, Smith Rebecca G, Lunnon Katie, Kumari Meena, Schalkwyk Leonard C, Morgan Kevin, Brookes Keeley, Hannon Eilis, Mill Jonathan

2020-Oct-29

DNA methylation, age, brain, clock, cortex

General

The Impact of Artificial Intelligence on the Chess World.

In JMIR serious games

This paper focuses on key areas in which artificial intelligence has affected the chess world, including cheat detection methods, which are especially necessary recently, as there has been an unexpected rise in the popularity of online chess. Many major chess events that were to take place in 2020 have been canceled, but the global popularity of chess has in fact grown in recent months due to easier conversion of the game from offline to online formats compared with other games. Still, though a game of chess can be easily played online, there are some concerns about the increased chances of cheating. Artificial intelligence can address these concerns.

Duca Iliescu Delia Monica

2020-Dec-10

AlphaZero, MuZero, artificial intelligence, cheat detection, chess, coronavirus, games

General

Emerging role of artificial intelligence in therapeutics for COVID-19: a systematic review.

In Journal of biomolecular structure & dynamics

To elucidate the role of artificial intelligence (AI) in therapeutics for coronavirus disease 2019 (COVID-19). Five databases were searched (December 2019-May 2020). We included both published and pre-print original articles in English that applied AI, machine learning or deep learning in drug repurposing, novel drug discovery, vaccine and antibody development for COVID-19. Out of 31 studies included, 16 studies applied AI for drug repurposing, whereas 10 studies utilized AI for novel drug discovery. Only four studies used AI technology for vaccine development, whereas one study generated stable antibodies against SARS-CoV-2. Approx. 50% of studies exclusively targeted 3CLpro of SARS-CoV-2, and only two studies targeted ACE/TMPSS2 for inhibiting host viral interactions. Around 16% of the identified drugs are in different phases of clinical evaluation against COVID-19. AI has emerged as a promising solution of COVID-19 therapeutics. During this current pandemic, many of the researchers have used AI-based strategies to process large databases in a more customized manner leading to the faster identification of several potential targets, novel/repurposing of drugs and vaccine candidates. A number of these drugs are either approved or are in a late-stage clinical trial and are potentially effective against SARS-CoV2 indicating validity of the methodology. However, as the use of AI-based screening program is currently in budding stage, sole reliance on such algorithms is not advisable at this current point of time and an evidence based approach is warranted to confirm their usefulness against this life-threatening disease. Communicated by Ramaswamy H. Sarma.

Kaushal Karanvir, Sarma Phulan, Rana S V, Medhi Bikash, Naithani Manisha

2020-Dec-10

Artificial intelligence, COVID-19, drug repurposing, novel drug discovery, vaccine development

Surgery

Artificial intelligence and perioperative medicine.

In Minerva anestesiologica ; h5-index 29.0

Perioperative medicine is a patient-centered, multidisciplinary and integrated clinical practice that starts from the moment of contemplation of surgery until full recovery. Every perioperative phase (preoperative, intraoperative and postoperative) has to be studied and planned in order to optimize the entire patient management. Perioperative optimization does not only concern a short-term outcome improvement, but it has also a strong impact on long term survival. Clinical cases variability leads to the collection and analysis of a huge amount of different data, coming from multiple sources, making perioperative management standardization very difficult. Artificial Intelligence (AI) can play a primary role in this challenge, helping human mind in perioperative practice planning and decision-making process. AI refers to the ability of a computer system to perform functions and reasoning typical of the human mind; Machine Learning (ML) could play a fundamental role in pre-surgical planning, during intraoperative phase and postoperative management. Perioperative medicine is the cornerstone of surgical patient management and the tools deriving from the application of AI seem very promising as a support in optimizing the management of each individual patient. Despite the increasing help that will derive from the use of AI tools, the uniqueness of the patient and the particularity of each individual clinical case will always keep the role of the human mind central in clinical and perioperative management. The role of the physician, who must analyse the outputs provided by AI by following his own experience and knowledge, remains and will always be essential.

Bignami Elena G, Cozzani Federico, Del Rio Paolo, Bellini Valentina

2020-Dec-10