bioRxiv Preprint

Sawmya, S.; Saha, A.; Tasnim, S.; Toufikuzzaman, M.; Anjum, N.; Rafid, A. H. M.; Rahman, M. S.; Rahman, M. S.

2021-04-20

bioRxiv Preprint

Rynkiewicz, P.; Babbitt, G. A.; Cui, F.; Hudson, A. O.; Lynch, M. L.

2021-04-20

ArXiv Preprint

Marta Li Wang

2021-04-19

ArXiv Preprint

Bishwas Mandal, Adaeze Okeukwu, Yihong Theis

2021-04-19

ArXiv Preprint

Bishwas Mandal, Adaeze Okeukwu, Yihong Theis

2021-04-19

In Science advances

Kim Ji-Yong, Yun Yong Ju, Jeong Joshua, Kim C-Yoon, Müller Klaus-Robert, Lee Seong-Whan

2021-Apr

In BMJ open

OBJECTIVES : There are increasing requirements to make research data, especially clinical trial data, more broadly available for secondary analyses. However, data availability remains a challenge due to complex privacy requirements. This challenge can potentially be addressed using synthetic data.

SETTING : Replication of a published stage III colon cancer trial secondary analysis using synthetic data generated by a machine learning method.

PARTICIPANTS : There were 1543 patients in the control arm that were included in our analysis.

PRIMARY AND SECONDARY OUTCOME MEASURES : Analyses from a study published on the real dataset were replicated on synthetic data to investigate the relationship between bowel obstruction and event-free survival. Information theoretic metrics were used to compare the univariate distributions between real and synthetic data. Percentage CI overlap was used to assess the similarity in the size of the bivariate relationships, and similarly for the multivariate Cox models derived from the two datasets.

RESULTS : Analysis results were similar between the real and synthetic datasets. The univariate distributions were within 1% of difference on an information theoretic metric. All of the bivariate relationships had CI overlap on the tau statistic above 50%. The main conclusion from the published study, that lack of bowel obstruction has a strong impact on survival, was replicated directionally and the HR CI overlap between the real and synthetic data was 61% for overall survival (real data: HR 1.56, 95% CI 1.11 to 2.2; synthetic data: HR 2.03, 95% CI 1.44 to 2.87) and 86% for disease-free survival (real data: HR 1.51, 95% CI 1.18 to 1.95; synthetic data: HR 1.63, 95% CI 1.26 to 2.1).

CONCLUSIONS : The high concordance between the analytical results and conclusions from synthetic and real data suggests that synthetic data can be used as a reasonable proxy for real clinical trial datasets.

TRIAL REGISTRATION NUMBER : NCT00079274.

Azizi Zahra, Zheng Chaoyi, Mosquera Lucy, Pilote Louise, El Emam Khaled

2021-Apr-16

epidemiology, health informatics, information management, information technology, statistics & research methods

In Applied and environmental microbiology

Levent Gizem, Schlochtermeier Ashlynn, Ives Samuel E, Norman Keri N, Lawhon Sara D, Loneragan Guy H, Anderson Robin C, Vinasco Javier, Bakker Henk C den, Scott H Morgan

2021-Apr-16

In Journal of pediatric surgery ; h5-index 38.0

PURPOSE : We aimed to introduce an explainable machine learning technology to help clinicians understand the risk factors for neonatal postoperative mortality at different levels.

METHODS : A total of 1481 neonatal surgeries performed between May 2016 and December 2019 at a children's hospital were included in this study. Perioperative variables, including vital signs during surgery, were collected and used to predict postoperative mortality. Several widely used machine learning methods were trained and evaluated on split datasets. The model with the best performance was explained by SHAP (SHapley Additive exPlanations) at different levels.

RESULTS : The random forest model achieved the best performance with an area under the receiver operating characteristic curve of 0.72 in the validation set. TreeExplainer of SHAP was used to identify the risk factors for neonatal postoperative mortality. The explainable machine learning model not only explains the risk factors identified by traditional statistical analysis but also identifies additional risk factors. The visualization of feature contributions at different levels by SHAP makes the "black-box" machine learning model easily understood by clinicians and families. Based on this explanation, vital signs during surgery play an important role in eventual survival.

CONCLUSIONS : The explainable machine learning model not only exhibited good performance in predicting neonatal surgical mortality but also helped clinicians understand each risk factor and each individual case.

Hu Yaoqin, Gong Xiaojue, Shu Liqi, Zeng Xian, Duan Huilong, Luo Qinyu, Zhang Baihui, Ji Yaru, Wang Xiaofeng, Shu Qiang, Li Haomin

2021-Apr-05

Machine learning, Neonatal surgery, Postoperative mortality

In Journal of clinical neuroscience : official journal of the Neurosurgical Society of Australasia

Hagiwara Akifumi, Otsuka Yujiro, Andica Christina, Kato Shimpei, Yokoyama Kazumasa, Hori Masaaki, Fujita Shohei, Kamagata Koji, Hattori Nobutaka, Aoki Shigeki

2021-May

Deep learning, Magnetic resonance imaging, Multiparametric quantitative imaging, Multiple sclerosis, Neuromyelitis optica spectrum disorder

In Analytica chimica acta

Lee Taeha, Kim Insu, Cheong Da Yeon, Roh Seokbeom, Jung Hyo Gi, Lee Sang Won, Kim Hyun Soo, Yoon Dae Sung, Hong Yoochan, Lee Gyudo

2021-May-08

Colorimetric paper sensor, High selectivity, Noninvasive, Polyaniline nanoparticles, Red blood cell membrane, Urine glucose

In BMC genomics ; h5-index 78.0

BACKGROUND : Sepsis is the major cause of death in Intensive Care Unit (ICU) globally. Molecular detection enables rapid diagnosis that allows early intervention to minimize the death rate. Recent studies showed that long non-coding RNAs (lncRNAs) regulate proinflammatory genes and are related to the dysfunction of organs in sepsis. Identifying lncRNA signature with absolute abundance is challenging because of the technical variation and the systematic experimental bias.

RESULTS : Cohorts (n = 768) containing whole blood lncRNA profiling of sepsis patients in the Gene Expression Omnibus (GEO) database were included. We proposed a novel diagnostic strategy that made use of the relative expressions of lncRNA pairs, which are reversed between sepsis patients and normal controls (eg. lncRNA_i > lncRNA_j in sepsis patients and lncRNA_i < lncRNA_j in normal controls), to identify 14 lncRNA pairs as a sepsis diagnostic signature. The signature was then applied to independent cohorts (n = 644) to evaluate its predictive performance across different ages and normalization methods. Comparing to common machine learning models and existing signatures, SepSigLnc consistently attains better performance on the validation cohorts from the same age group (AUC = 0.990 & 0.995 in two cohorts) and across different groups (AUC = 0.878 on average), as well as cohorts processed by an alternative normalization method (AUC = 0.953 on average). Functional analysis demonstrates that the lncRNA pairs in SepsigLnc are functionally similar and tend to implicate in the same biological processes including cell fate commitment and cellular response to steroid hormone stimulus.

CONCLUSION : Our study identified 14 lncRNA pairs as signature that can facilitate the diagnosis of septic patients at an intervenable point when clinical manifestations are not dramatic. Also, the computational procedure can be generalized to a standard procedure for discovering diagnostic molecule signatures.

Zheng Xubin, Leung Kwong-Sak, Wong Man-Hon, Cheng Lixin

2021-Apr-16

Diagnostics, Long non-coding RNA, Relative expression, Sepsis, Signature

In Biochemical and biophysical research communications

Lu Shiting, Kong Wei, Wang Shuaiqun

2021-Apr-14

Alzheimer’s disease, Blood brain barrier, Mutual information, PANDA

In Health informatics journal ; h5-index 25.0

Duong Duc Long, Kabir Malitha Humayun, Kuo Rong Fu

artificial intelligence, caries detection, computer modeling, digital imaging, image analysis, machine learning

In The Science of the total environment

Wu Chengcheng, Zhang Xiaoqin, Wang Wanjie, Lu Chengpeng, Zhang Yong, Qin Wei, Tick Geoffrey R, Liu Bo, Shu Longcang

2021-Apr-08

Groundwater level simulation, Long short-term memory (LSTM) model, Surface water, Wavelet transform (WT)

In European journal of radiology ; h5-index 47.0

OBJECTIVES : New markers are required to predict chemoradiation response in oropharyngeal squamous cell carcinoma (OPSCC) patients. This study evaluated the ability of magnetic resonance (MR) radiomics to predict locoregional control (LRC) and overall survival (OS) after chemoradiation and aimed to determine whether this has added value to traditional clinical outcome predictors.

METHODS : 177 OPSCC patients were eligible for this study. Radiomic features were extracted from the primary tumor region in T1-weighted postcontrast MRI acquired before chemoradiation. Logistic regression models were created using either clinical variables (clinical model), radiomic features (radiomic model) or clinical and radiomic features combined (combined model) to predict LRC and OS 2-years posttreatment. Model performance was evaluated using area under the curve (AUC), 95 % confidence intervals were calculated using 500 iterations of bootstrap. All analyses were performed for the total population and the Human papillomavirus (HPV) negative tumor subgroup.

RESULTS : A combined model predicted treatment outcome with a higher AUC (LRC: 0.745 [0.734-0.757], OS: 0.744 [0.735-0.753]) than the clinical model (LRC: 0.607 [0.594-0.620], OS: 0.708 [0.697-0.719]). Performance of the radiomic model was comparable to the combined model for LRC (AUC: 0.740 [0.729-0.750]), but not for OS prediction (AUC: 0.654 [0.646-0.662]). In HPV negative patients, the performance of all models was not sufficient with AUCs ranging from 0.587 to 0.660 for LRC and 0.559 to 0.600 for OS prediction.

CONCLUSION : Predictive models that include clinical variables and radiomic tumor features derived from MR images of OPSCC better predict LRC after chemoradiation than models based on only clinical variables. Predictive models that include clinical variables perform better than models based on only radiomic features for the prediction of OS.

Bos Paula, van den Brekel Michiel W M, Gouw Zeno A R, Al-Mamgani Abrahim, Taghavi Marjaneh, Waktola Selam, Aerts Hugo J W L, Castelijns Jonas A, Beets-Tan Regina G H, Jasperse Bas

2021-Apr-08

Head and neck neoplasms, Machine learning, Magnetic Resonance Imaging, Oropharyngeal neoplasms, Radiomics, Treatment outcome

In Computers in biology and medicine

Ji Zhiwei, Liu Changan, Zhao Weiling, Soto Claudio, Zhou Xiaobo

2021-Apr-05

AD, Agent-based model, Microglia, Plaque, TREM2

In Computers in biology and medicine

BACKGROUND : Recent advances in robotics and deep learning can be used in endoscopic surgeries and can provide numerous advantages by freeing one of the surgeon's hands. This study aims to automatically detect the tip of the instrument, localize a point, and evaluate the detection accuracy in biportal endoscopic spine surgery (BESS). The tip detection could serve as a preliminary study for the development of vision intelligence in robotic endoscopy.

METHODS : The dataset contains 2310 frames from 9 BESS videos with x and y coordinates of the tip annotated by an expert. We trained two state-of-the-art detectors, RetinaNet and YOLOv2, with bounding boxes centered around the tip annotations with specific margin sizes to determine the optimal margin size for detecting the tip of the instrument and localizing the point. We calculated the recall, precision, and F1-score with a fixed box size for both ground truth tip coordinates and predicted midpoints to compare the performance of the models trained with different margin size bounding boxes.

RESULTS : For RetinaNet, a margin size of 150 pixels was optimal with a recall of 1.000, precision of 0.733, and F1-score of 0.846. For YOLOv2, a margin size of 150 pixels was optimal with a recall of 0.864, precision of 0.808, F1-score of 0.835. Also, the optimal margin size of 150 pixels of RetinaNet was used to cross-validate its overall robustness. The resulting mean recall, precision, and F1-score were 1.000 ± 0.000, 0.767 ± 0.033, and 0.868 ± 0.022, respectively.

CONCLUSIONS : In this study, we evaluated an automatic tip detection method for surgical instruments in endoscopic surgery, compared two state-of-the-art detection algorithms, RetinaNet and YOLOv2, and validated the robustness with cross-validation. This method can be applied in different types of endoscopy tip detection.

Cho Sue Min, Kim Young-Gon, Jeong Jinhoon, Kim Inhwan, Lee Ho-Jin, Kim Namkug

2021-Apr-14

Convolutional neural network, Deep learning, Detection, Endoscopic surgery, Localization

In Computers in biology and medicine

Vafaeezadeh Majid, Behnam Hamid, Hosseinsabet Ali, Gifani Parisa

2021-Apr-14

DCNN, Echocardiographic, EfficientNet, Prosthetic mitral valve

In Medical image analysis

Grammatikopoulou Maria, Flouty Evangello, Kadkhodamohammadi Abdolrahim, Quellec Gwenolé, Chow Andre, Nehme Jean, Luengo Imanol, Stoyanov Danail

2021-Mar-31

Cataract surgery, Dataset, Semantic segmentation

In Journal of neuroscience methods

BACKGROUND : . In this work, we explore the possibility of decoding Imagined Speech (IS) brain waves using machine learning techniques.

APPROACH : . We design two finite state machines to create an interface for controlling a computer system using an IS-based brain-computer interface. To decode IS signals, we propose a covariance matrix of Electroencephalogram channels as input features, covariance matrices projection to tangent space for obtaining vectors from matrices, principal component analysis for dimension reduction of vectors, an artificial neural network (ANN) as a classification model, and bootstrap aggregation for creating an ensemble of ANN models.

RESULT : . Based on these findings, we are first to use an IS-based system to operate a computer and obtain an information transfer rate of 21-bits-per-minute. The proposed approach can decode the IS signal with a mean classification accuracy of 85% on classifying one long vs. short word. Our proposed approach can also differentiate between IS and rest state brain signals with a mean classification accuracy of 94%. COMPARISON: . After comparison, we show that our approach performs equivalent to the state-of-the-art approach (SOTA) on decoding long vs. short word classification task. We also show that the proposed method outperforms SOTA significantly on decoding three short words and vowels with an average margin of 11% and 9%, respectively.

CONCLUSION : . These results show that the proposed approach can decode a wide variety of IS signals and is practically applicable in a real-time environment.

Singh Abhiram, Gumaste Ashwin

2021-Apr-14

Artificial neural network, Brain-computer interface, Electroencephalogram, Finite state MACHINE, Imagined speech

In Journal of neuroscience methods

BACKGROUND : Neonatal seizures are a common occurrence in clinical settings, requiring immediate attention and detection. Previous studies have proposed using manual feature extraction coupled with machine learning, or deep learning to classify between seizure and non-seizure states.

NEW METHOD : In this paper a deep learning based approach is used for neonatal seizure classification using electroencephalogram (EEG) signals. The architecture detects seizure activity in raw EEG signals as opposed to common state-of-art, where manual feature extraction with machine learning algorithms is used. The architecture is a two-dimensional (2D) convolutional neural network (CNN) to classify between seizure/non-seizure states.

RESULTS : The dataset used for this study is annotated by three experts and as such three separate models are trained on individual annotations, resulting in average accuracies (ACC) of 95.6%, 94.8% and 90.1% respectively, and average area under the receiver operating characteristic curve (AUC) of 99.2%, 98.4% and 96.7% respectively. The testing was done using 10-cross fold validation, so that the performance can be an accurate representation of the architectures classification capability in a clinical setting. After training/testing of the three individual models, a final ensemble model is made consisting of the three models. The ensemble model gives an average ACC and AUC of 96.3% and 99.3% respectively.

COMPARISON WITH EXISTING METHODS : This study outperforms previous studies, with increased ACC and AUC results coupled with use of small time windows (1 second) used for evaluation.

CONCLUSION : The proposed approach is promising for detecting seizure activity in unseen neonate data in a clinical setting.

Tanveer M Asjid, Khan Muhammad Jawad, Sajid Hasan, Naseer Noman

2021-Apr-14

EEG, convolutional neural network, detection, ensemble model, neonatal, seizures

In Biometrics

Basak Piyali, Linero Antonio, Sinha Debajyoti, Lipsitz Stuart

2021-Apr-17

Bayesian Additive Regression Trees, machine learning, nonproportional hazards, semiparametric, survival analysis

In European radiology ; h5-index 62.0

OBJECTIVES : To investigate the feasibility and prognostic implications of coronary CT angiography (CCTA) derived fractional flow reserve (FFR_CT) in patients who have undergone stents implantation.

METHODS : Firstly, the feasibility of FFR_CT in stented vessels was validated. The diagnostic performance of FFR_CT in identifying hemodynamically in-stent restenosis (ISR) in 33 patients with invasive FFR ≤ 0.88 as reference standard, intra-group correlation coefficient (ICC) between FFR_CT and FFR was calculated. Secondly, prognostic value was assessed with 115 patients with serial CCTA scans after PCI. Stent characteristics (location, diameter, length, etc.), CCTA measurements (minimum lumen diameter [MLD], minimum lumen area [MLA], ISR), and FFR_CT measurements (FFR_CT, ΔFFR_CT, ΔFFR_CT/stent length) both at baseline and follow-up were recorded. Longitudinal analysis included changes of MLD, MLA, ISR, and FFR_CT. The primary endpoint was major adverse cardiovascular events (MACE).

RESULTS : Per-patient accuracy of FFR_CT was 0.85 in identifying hemodynamically ISR. FFR_CT had a good correlation with FFR (ICC = 0.84). 15.7% (18/115) developed MACE during 25 months since follow-up CCTA. Lasso regression identified age and follow-up ΔFFR_CT/length as candidate variables. In the Cox proportional hazards model, age (hazard ratio [HR], 1.102 [95% CI, 1.032-1.177]; p = 0.004) and follow-up ΔFFR_CT/length (HR, 1.014 [95% CI, 1.006-1.023]; p = 0.001) were independently associated with MACE (c-index = 0.856). Time-dependent ROC analysis showed AUC was 0.787 (95% CI, 0.594-0.980) at 25 months to predict adverse outcome. After bootstrap validation with 1000 resamplings, the bias-corrected c-index was 0.846.

CONCLUSIONS : Noninvasive ML-based FFR_CT is feasible in patients following stents implantation and shows prognostic value in predicting adverse events after stents implantation in low-moderate risk patients.

KEY POINTS : • Machine-learning-based FFR_CT is feasible to evaluate the functional significance of in-stent restenosis in patients with stent implantation. • Follow-up △FFR_CT along with the stent length might have prognostic implication in patients with stent implantation and low-to-moderate risk after 2 years follow-up. The prognostic role of FFR_CT in patients with moderate-to-high or high risk needs to be further studied. • FFR_CT might refine the clinical pathway of patients with stent implantation to invasive catheterization.

Tang Chun Xiang, Guo Bang Jun, Schoepf Joseph U, Bayer Richard R, Liu Chun Yu, Qiao Hong Yan, Zhou Fan, Lu Guang Ming, Zhou Chang Sheng, Zhang Long Jiang

2021-Apr-17

Computed tomography angiography, Coronary artery disease, Coronary restenosis, Myocardial fractional flow reserve, Stents

In Physical review. E

Çakmak Burak, Opper Manfred

2021-Mar

In Database : the journal of biological databases and curation

Fadlelmola Faisal M, Ghedira Kais, Hamdi Yosr, Hanachi Mariem, Radouani Fouzia, Allali Imane, Kiran Anmol, Zass Lyndon, Alsayed Nihad, Fassatoui Meriem, Samtal Chaimae, Ahmed Samah, Da Rocha Jorge, Chaqsare Souad, Sallam Reem M, Chaouch Melek, Farahat Mohammed, Ssekagiri Alfred, Parker Ziyaad, Adil Mai, Turkson Michael, Benchaalia Aymen, Benkahla Alia, Panji Sumir, Kassim Samar, Souiai Oussema, Mulder Nicola

2021-Apr-17

In Clinical and experimental dermatology

Agarwala S, Mata D A, Hafeez F

2021-Apr-17

In Global change biology

Duarte de Paula Costa Micheli, Lovelock Catherine E, Waltham Nathan J, Young Mary, Adame Maria Fernanda, Bryant Catherine V, Butler Don, Green David, Rasheed Michael A, Salinas Cristian, Serrano Oscar, York Paul H, Whitt Ashley A, Macreadie Peter I

2021-Apr-16

blue carbon, climate change, coastal wetlands, mangroves, seagrass meadows, soil carbon stocks, tidal marshes

In Cognitive, affective & behavioral neuroscience

Wen Xiaotong, Li Wen, Liu Yuan, Liu Zhenghao, Zhao Ping, Zhu Zhiyuan, Wu Xia

2021-Apr-17

Cognitive control, Dynamic functional connectivity, Effective connectivity, Middle thalamus

In International journal of computer assisted radiology and surgery

PURPOSE : Pelvic bone segmentation in CT has always been an essential step in clinical diagnosis and surgery planning of pelvic bone diseases. Existing methods for pelvic bone segmentation are either hand-crafted or semi-automatic and achieve limited accuracy when dealing with image appearance variations due to the multi-site domain shift, the presence of contrasted vessels, coprolith and chyme, bone fractures, low dose, metal artifacts, etc. Due to the lack of a large-scale pelvic CT dataset with annotations, deep learning methods are not fully explored.

METHODS : In this paper, we aim to bridge the data gap by curating a large pelvic CT dataset pooled from multiple sources, including 1184 CT volumes with a variety of appearance variations. Then, we propose for the first time, to the best of our knowledge, to learn a deep multi-class network for segmenting lumbar spine, sacrum, left hip, and right hip, from multiple-domain images simultaneously to obtain more effective and robust feature representations. Finally, we introduce a post-processor based on the signed distance function (SDF).

RESULTS : Extensive experiments on our dataset demonstrate the effectiveness of our automatic method, achieving an average Dice of 0.987 for a metal-free volume. SDF post-processor yields a decrease of 15.1% in Hausdorff distance compared with traditional post-processor.

CONCLUSION : We believe this large-scale dataset will promote the development of the whole community and open source the images, annotations, codes, and trained baseline models at https://github.com/ICT-MIRACLE-lab/CTPelvic1K .

Liu Pengbo, Han Hu, Du Yuanqi, Zhu Heqin, Li Yinhao, Gu Feng, Xiao Honghu, Li Jun, Zhao Chunpeng, Xiao Li, Wu Xinbao, Zhou S Kevin

2021-Apr-16

CT dataset, Deep learning, Pelvic segmentation, SDF post-processing

In Scientific reports ; h5-index 158.0

Kim Taesung, Kim Jinhee, Choi Hyuk Soon, Kim Eun Sun, Keum Bora, Jeen Yoon Tae, Lee Hong Sik, Chun Hoon Jai, Han Sung Yong, Kim Dong Uk, Kwon Soonwook, Choo Jaegul, Lee Jae Min

2021-Apr-16

In Scientific reports ; h5-index 158.0

Tang Siyi, Ghorbani Amirata, Yamashita Rikiya, Rehman Sameer, Dunnmon Jared A, Zou James, Rubin Daniel L

2021-Apr-16

In Scientific data

Leuschner Johannes, Schmidt Maximilian, Baguer Daniel Otero, Maass Peter

2021-Apr-16

In Nature communications ; h5-index 260.0

Wang Xiao, Alnabati Eman, Aderinwale Tunde W, Maddhuri Venkata Subramaniya Sai Raghavendra, Terashi Genki, Kihara Daisuke

2021-Apr-16

In The European respiratory journal

Berra Gregory, Farkona Sofia, Mohammed-Ali Zahraa, Kotlyar Max, Levy Liran, Clotet-Freixas Sergi, Ly Phillip, Renaud-Picard Benjamin, Zehong Guan, Daigneault Tina, Duong Allen, Batruch Ihor, Jurisica Igor, Konvalinka Ana, Martinu Tereza

2021-Apr-16

In European urology focus

BACKGROUND : It has been shown that metrics recorded for instrument kinematics during robotic surgery can predict urinary continence outcomes.

OBJECTIVE : To evaluate the contributions of patient and treatment factors, surgeon efficiency metrics, and surgeon technical skill scores, especially for vesicourethral anastomosis (VUA), to models predicting urinary continence recovery following robot-assisted radical prostatectomy (RARP).

DESIGN, SETTING, AND PARTICIPANTS : Automated performance metrics (APMs; instrument kinematics and system events) and patient data were collected for RARPs performed from July 2016 to December 2017. Robotic Anastomosis Competency Evaluation (RACE) scores during VUA were manually evaluated. Training datasets included: (1) patient factors; (2) summarized APMs (reported over RARP steps); (3) detailed APMs (reported over suturing phases of VUA); and (4) technical skills (RACE). Feature selection was used to compress the dimensionality of the inputs.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS : The study outcome was urinary continence recovery, defined as use of 0 or 1 safety pads per day. Two predictive models (Cox proportional hazards [CoxPH] and deep learning survival analysis [DeepSurv]) were used.

RESULTS AND LIMITATIONS : Of 115 patients undergoing RARP, 89 (77.4%) recovered their urinary continence and the median recovery time was 166 d (interquartile range [IQR] 82-337). VUAs were performed by 23 surgeons. The median RACE score was 28/30 (IQR 27-29). Among the individual datasets, technical skills (RACE) produced the best models (C index: CoxPH 0.695, DeepSurv: 0.708). Among summary APMs, posterior/anterior VUA yielded superior model performance over other RARP steps (C index 0.543-0.592). Among detailed APMs, metrics for needle driving yielded top-performing models (C index 0.614-0.655) over other suturing phases. DeepSurv models consistently outperformed CoxPH; both approaches performed best when provided with all the datasets. Limitations include feature selection, which may have excluded relevant information but prevented overfitting.

CONCLUSIONS : Technical skills and "needle driving" APMs during VUA were most contributory. The best-performing model used synergistic data from all datasets.

PATIENT SUMMARY : One of the steps in robot-assisted surgical removal of the prostate involves joining the bladder to the urethra. Detailed information on surgeon performance for this step improved the accuracy of predicting recovery of urinary continence among men undergoing this operation for prostate cancer.

Trinh Loc, Mingo Samuel, Vanstrum Erik B, Sanford Daniel I, Aastha Ma, Runzhuo Nguyen, Jessica H Liu, Yan Hung

2021-Apr-12

Artificial intelligence, Machine learning, Prostatectomy, Robotics, Survival analysis, Urinary incontinence

In BJPsych open

Cobo Aurora, Porras-Segovia Alejandro, Pérez-Rodríguez María Mercedes, Artés-Rodríguez Antonio, Barrigón Maria Luisa, Courtet Philippe, Baca-García Enrique

2021-Apr-16

COVID-19, Suicide, ecological momentary assessment, machine learning, suicide attempt

In Journal of cheminformatics

Halder Amit Kumar, Dias Soeiro Cordeiro M Natália

2021-Apr-15

Feature selection, Machine learning, Multitarget models, QSAR, Software tools

In Journal of cheminformatics

Hu Fan, Jiang Jiaxin, Wang Dongqi, Zhu Muchun, Yin Peng

2021-Apr-15

Deep learning, Drug discovery, Interpretable, Multi‐task

In Bioinspiration & biomimetics

Bhardwaj Ananya, Khyam Mohammad Omar, Mueller Rolf

2021-Apr-16

bats, biosonar, dynamics, machine learning, sensing, soft robotics

In Computer methods and programs in biomedicine

BACKGROUND AND OBJECTIVE : The purpose of the present study was to investigate low-shot deep learning models applied to conjunctival melanoma detection using a small dataset with ocular surface images.

METHODS : A dataset was composed of anonymized images of four classes; conjunctival melanoma (136), nevus or melanosis (93), pterygium (75), and normal conjunctiva (94). Before training involving conventional deep learning models, two generative adversarial networks (GANs) were constructed to augment the training dataset for low-shot learning. The collected data were randomly divided into training (70%), validation (10%), and test (20%) datasets. Moreover, 3D melanoma phantoms were designed to build an external validation set using a smartphone. The GoogleNet, InceptionV3, NASNet, ResNet50, and MobileNetV2 architectures were trained through transfer learning and validated using the test and external validation datasets.

RESULTS : The deep learning model demonstrated a significant improvement in the classification accuracy of conjunctival lesions using synthetic images generated by the GAN models. MobileNetV2 with GAN-based augmentation displayed the highest accuracy of 87.5% in the four-class classification and 97.2% in the binary classification for the detection of conjunctival melanoma. It showed an accuracy of 94.0% using 3D melanoma phantom images captured using a smartphone camera.

CONCLUSIONS : The present study described a low-shot deep learning model that can detect conjunctival melanomas using ocular surface images. To the best of our knowledge, this study is the first to develop a deep learning model to detect conjunctival melanoma using a digital imaging device such as smartphone camera.

Yoo Tae Keun, Choi Joon Yul, Kim Hong Kyu, Ryu Ik Hee, Kim Jin Kuk

2021-Apr-03

Conjunctival melanoma, Conjunctival nevus, Deep learning, Low-shot learning, Melanosis

In American journal of otolaryngology ; h5-index 23.0

OBJECTIVES : Cervical lymph nodes with internal cystic changes are seen with several pathologies, including papillary thyroid carcinoma (PTC), tuberculosis (TB), and HPV-positive oropharyngeal squamous cell carcinoma (HPV+OPSCC). Differentiating these lymph nodes is difficult in the absence of a known primary tumor or reliable medical history. In this study, we assessed the utility of deep learning in differentiating the pathologic lymph nodes of PTC, TB, and HPV+OPSCC on CT.

METHODS : A total of 173 lymph nodes (55 PTC, 58 TB, and 60 HPV+OPSCC) were selected based on pathology records and suspicious morphological features. These lymph nodes were divided into the training set (n = 131) and the test set (n = 42). In deep learning analysis, JPEG lymph node images were extracted from the CT slice that included the largest area of each node and fed into a deep learning training session to create a diagnostic model. Transfer learning was used with the deep learning model architecture of ResNet-101. Using the test set, the diagnostic performance of the deep learning model was compared against the histopathological diagnosis and to the diagnostic performances of two board-certified neuroradiologists.

RESULTS : Diagnostic accuracy of the deep learning model was 0.76 (=32/42), whereas those of Radiologist 1 and Radiologist 2 were 0.48 (=20/42) and 0.41 (=17/42), respectively. Deep learning derived diagnostic accuracy was significantly higher than both of the two neuroradiologists (P < 0.01, respectively).

CONCLUSION : Deep learning algorithm holds promise to become a useful diagnostic support tool in interpreting cervical lymphadenopathy.

Onoue Keita, Fujima Noriyuki, Andreu-Arasa V Carlota, Setty Bindu N, Sakai Osamu

2021-Apr-09

Cervical lymphadenopathy, Computed tomography, Deep learning, Human papillomavirus, Machine learning, Papillary thyroid carcinoma, Tuberculosis

In International journal of medical informatics ; h5-index 49.0

INTRODUCTION : Understanding and managing clinician workload is important for clinician (nurses, physicians and advanced practice providers) occupational health as well as patient safety. Efforts have been made to develop strategies for managing clinician workload by improving patient assignment. The goal of the current study is to use electronic health record (EHR) data to predict the amount of work that individual patients contribute to clinician workload (patient-related workload).

METHODS : One month of EHR data was retrieved from an emergency department (ED). A list of workload indicators and five potential workload proxies were extracted from the data. Linear regression and four machine learning classification algorithms were utilized to model the relationship between the indicators and the proxies.

RESULTS : Linear regression proved that the indicators explained a substantial amount of variance of the proxies (four out of five proxies were modeled with R² > 0.80). Classification algorithms also showed success in classifying a patient as having high or low task demand based on data from early in the ED visit (e.g. 80 % accurate binary classification with data from the first hour).

CONCLUSION : The main contribution of this study is demonstrating the potential of using EHR data to predict patient-related workload automatically in the ED. The predicted workload can potentially help in managing clinician workload by supporting decisions around the assignment of new patients to providers. Future work should focus on identifying the relationship between workload proxies and actual workload, as well as improving prediction performance of regression and multi-class classification.

Wang Xiaomei, Blumenthal H Joseph, Hoffman Daniel, Benda Natalie, Kim Tracy, Perry Shawna, Franklin Ella S, Roth Emilie M, Hettinger A Zachary, Bisantz Ann M

2021-Apr-09

Electronic health record, Emergency department, Machine learning, Workload

In Biosensors & bioelectronics

Lim Ji Won, Kim Tai-Yong, Woo Min-Ah

2021-Apr-07

Colorimetry, Electrochemistry, Field detection, Fluorescence, Mercury, Sensor platform

In Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy

Wang Yuan, Ma Hongbing, Wang Jingzhe, Liu Li, Pietikäinen Matti, Zhang Zipeng, Chen Xiangyue

2021-Mar-26

Data enhancement (DA), Deep neural network (DNN), Soil heavy metal pollution, Soil hyperspectrum, Support vector machine (SVM), k-nearest neighbour (KNN)

In Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society

Mofrad Samaneh Abolpour, Lundervold Arvid, Lundervold Alexander Selvikvåg

2021-Apr-02

“Alzheimers disease”, Longitudinal data analysis, MRI, Machine learning, Mild cognitive impairment, Mixed effects models

In Medical image analysis

Signoroni Alberto, Savardi Mattia, Benini Sergio, Adami Nicola, Leonardi Riccardo, Gibellini Paolo, Vaccher Filippo, Ravanelli Marco, Borghesi Andrea, Maroldi Roberto, Farina Davide

2021-Mar-31

COVID-19 severity assessment, Chest X-rays, Convolutional neural networks, End-to-end learning, Semi-quantitative rating

In BioData mining

BACKGROUND : As per the 2017 WHO fact sheet, Coronary Artery Disease (CAD) is the primary cause of death in the world, and accounts for 31% of total fatalities. The unprecedented 17.6 million deaths caused by CAD in 2016 underscores the urgent need to facilitate proactive and accelerated pre-emptive diagnosis. The innovative and emerging Machine Learning (ML) techniques can be leveraged to facilitate early detection of CAD which is a crucial factor in saving lives. The standard techniques like angiography, that provide reliable evidence are invasive and typically expensive and risky. In contrast, ML model generated diagnosis is non-invasive, fast, accurate and affordable. Therefore, ML algorithms can be used as a supplement or precursor to the conventional methods. This research demonstrates the implementation and comparative analysis of K Nearest Neighbor (k-NN) and Random Forest ML algorithms to achieve a targeted "At Risk" CAD classification using an emerging set of 35 cytokine biomarkers that are strongly indicative predictive variables that can be potential targets for therapy. To ensure better generalizability, mechanisms such as data balancing, repeated k-fold cross validation for hyperparameter tuning, were integrated within the models. To determine the separability efficacy of "At Risk" CAD versus Control achieved by the models, Area under Receiver Operating Characteristic (AUROC) metric is used which discriminates the classes by exhibiting tradeoff between the false positive and true positive rates.

RESULTS : A total of 2 classifiers were developed, both built using 35 cytokine predictive features. The best AUROC score of .99 with a 95% Confidence Interval (CI) (.982,.999) was achieved by the Random Forest classifier using 35 cytokine biomarkers. The second-best AUROC score of .954 with a 95% Confidence Interval (.929,.979) was achieved by the k-NN model using 35 cytokines. A p-value of less than 7.481e-10 obtained by an independent t-test validated that Random Forest classifier was significantly better than the k-NN classifier with regards to the AUROC score. Presently, as large-scale efforts are gaining momentum to enable early, fast, reliable, affordable, and accessible detection of individuals at risk for CAD, the application of powerful ML algorithms can be leveraged as a supplement to conventional methods such as angiography. Early detection can be further improved by incorporating 65 novel and sensitive cytokine biomarkers. Investigation of the emerging role of cytokines in CAD can materially enhance the detection of risk and the discovery of mechanisms of disease that can lead to new therapeutic modalities.

Saharan Seema Singh, Nagar Pankaj, Creasy Kate Townsend, Stock Eveline O, Feng James, Malloy Mary J, Kane John P

2021-Apr-15

AUROC, CAD, Classification, Confidence interval, Distance metrics, ML, Plasma cytokines, ROSE, Random Forest, k-NN, k-fold cross validation

In Developmental cognitive neuroscience ; h5-index 46.0

Adise Shana, Allgaier Nicholas, Laurent Jennifer, Hahn Sage, Chaarani Bader, Owens Max, Yuan DeKang, Nyugen Philip, Mackey Scott, Potter Alexandra, Garavan Hugh P

2021-Mar-30

Childhood obesity, Inhibitory control, Machine-learning, Reward, Weight gain, Weight stability, fMRI

In Biological psychiatry. Cognitive neuroscience and neuroimaging

BACKGROUND : Individuals with autism spectrum disorder (ASD) exhibit frequent behavioral deficits in facial emotion recognition (FER). It remains unknown whether these deficits arise because facial emotion information is not encoded in their neural signal, or because it is encoded, but fails to translate to FER behavior (deployment). This distinction has functional implications, including constraining when differences in social information processing occur in ASD, and guiding interventions (i.e., developing prosthetic FER-vs.-reinforcing existing skills).

METHODS : We utilized a discriminative and contemporary machine learning approach - Deep Convolutional Neural Networks (CNN) - to classify facial emotions viewed by individuals with and without ASD(N= 88) from concurrently-recorded electroencephalography signals.

RESULTS : The CNN classified facial emotions with high accuracy for both ASD and non-ASD groups, even though individuals with ASD performed more poorly on the concurrent FER task. In fact, CNN accuracy was greater in the ASD group, and was not related to behavioral performance. This pattern of results replicated across three independent participant samples. Moreover, feature-importance analyses suggest that a late temporal window of neural activity (1000-1500ms) may be uniquely important in facial emotion classification for individuals for ASD.

CONCLUSIONS : Our results reveal for the first time that facial emotion information is encoded in the neural signal of individuals with (and without) ASD. Thus, observed difficulties in behavioral FER associated with ASD likely arise from difficulties in decoding or deployment of facial emotion information within the neural signal. Interventions should focus on capitalizing on this intact encoding rather than promoting compensation or FER prosthetics.

Mayor Torres Juan Manuel, Clarkson Tessa, Hauschild Kathryn M, Luhmann Christian C, Lerner Matthew D, Riccardi Giuseppe

2021-Apr-13

Autism Spectrum Disorder, Deep Convolutional Neural Networks, Electroencephalography, Facial Emotion Recognition

In Journal of biomedical informatics ; h5-index 55.0

INTRODUCTION : Clustering analyses in clinical contexts hold promise to improve the understanding of patient phenotype and disease course in chronic and acute clinical medicine. However, work remains to ensure that solutions are rigorous, valid, and reproducible. In this paper, we evaluate best practices for dissimilarity matrix calculation and clustering on mixed-type, clinical data.

METHODS : We simulate clinical data to represent problems in clinical trials, cohort studies, and EHR data, including single-type datasets (binary, continuous, categorical) and 4 data mixtures. We test 5 single distance metrics (Jaccard, Hamming, Gower, Manhattan, Euclidean) and 3 mixed distance metrics (DAISY, Supersom, and Mercator) with 3 clustering algorithms (hierarchical (HC), k-medoids, self-organizing maps (SOM)). We quantitatively and visually validate by Adjusted Rand Index (ARI) and silhouette width (SW).

RESULTS : HC outperformed k-medoids and SOM by ARI across data types. DAISY produced the highest mean ARI for mixed data types for all mixtures except unbalanced mixtures dominated by continuous data. Compared to the Hamming distance, a real data application of DAISY with HC uncovered superior, separable clusters.

DISCUSSION : Selecting an appropriate mixed-type metric allows the investigator to obtain optimal separation of patient clusters and get maximum use of their data. Superior metrics for mixed-type data handle multiple data types using multiple, type-focused distances. Better subclassification of disease opens avenues for targeted treatments, precision medicine, clinical decision support, and improved patient outcomes.

Coombes Caitlin E, Abrams Zachary B, Coombes Kevin R, Brock Guy

2021-Apr-13

Clustering, clinical informatics, clinical trial, electronic health record, unsupervised machine learning

In Cell ; h5-index 250.0

Belbin Gillian M, Cullina Sinead, Wenric Stephane, Soper Emily R, Glicksberg Benjamin S, Torre Denis, Moscati Arden, Wojcik Genevieve L, Shemirani Ruhollah, Beckmann Noam D, Cohain Ariella, Sorokin Elena P, Park Danny S, Ambite Jose-Luis, Ellis Steve, Auton Adam, Bottinger Erwin P, Cho Judy H, Loos Ruth J F, Abul-Husn Noura S, Zaitlen Noah A, Gignoux Christopher R, Kenny Eimear E

2021-Apr-15

biobanks, computational genomics, electronic health records, genetic ancestry, genomic medicine, health disparities, machine learning, population health

In PloS one ; h5-index 176.0

BACKGROUND : Accurately predicting the survival rate of breast cancer patients is a major issue for cancer researchers. Machine learning (ML) has attracted much attention with the hope that it could provide accurate results, but its modeling methods and prediction performance remain controversial. The aim of this systematic review is to identify and critically appraise current studies regarding the application of ML in predicting the 5-year survival rate of breast cancer.

METHODS : In accordance with the PRISMA guidelines, two researchers independently searched the PubMed (including MEDLINE), Embase, and Web of Science Core databases from inception to November 30, 2020. The search terms included breast neoplasms, survival, machine learning, and specific algorithm names. The included studies related to the use of ML to build a breast cancer survival prediction model and model performance that can be measured with the value of said verification results. The excluded studies in which the modeling process were not explained clearly and had incomplete information. The extracted information included literature information, database information, data preparation and modeling process information, model construction and performance evaluation information, and candidate predictor information.

RESULTS : Thirty-one studies that met the inclusion criteria were included, most of which were published after 2013. The most frequently used ML methods were decision trees (19 studies, 61.3%), artificial neural networks (18 studies, 58.1%), support vector machines (16 studies, 51.6%), and ensemble learning (10 studies, 32.3%). The median sample size was 37256 (range 200 to 659820) patients, and the median predictor was 16 (range 3 to 625). The accuracy of 29 studies ranged from 0.510 to 0.971. The sensitivity of 25 studies ranged from 0.037 to 1. The specificity of 24 studies ranged from 0.008 to 0.993. The AUC of 20 studies ranged from 0.500 to 0.972. The precision of 6 studies ranged from 0.549 to 1. All of the models were internally validated, and only one was externally validated.

CONCLUSIONS : Overall, compared with traditional statistical methods, the performance of ML models does not necessarily show any improvement, and this area of research still faces limitations related to a lack of data preprocessing steps, the excessive differences of sample feature selection, and issues related to validation. Further optimization of the performance of the proposed model is also needed in the future, which requires more standardization and subsequent validation.

Li Jiaxin, Zhou Zijun, Dong Jianyu, Fu Ying, Li Yuan, Luan Ze, Peng Xin

2021

In PloS one ; h5-index 176.0

Bernas Antoine, Breuer Lisanne E M, Aldenkamp Albert P, Zinger Svitlana

2021

In PloS one ; h5-index 176.0

Englbrecht Fabian, Ruider Iris E, Bausch Andreas R

2021

In PloS one ; h5-index 176.0

OBJECTIVE : To assess whether machine learning algorithms (MLA) can predict eyes that will undergo rapid glaucoma progression based on an initial visual field (VF) test.

DESIGN : Retrospective analysis of longitudinal data.

SUBJECTS : 175,786 VFs (22,925 initial VFs) from 14,217 patients who completed ≥5 reliable VFs at academic glaucoma centers were included.

METHODS : Summary measures and reliability metrics from the initial VF and age were used to train MLA designed to predict the likelihood of rapid progression. Additionally, the neural network model was trained with point-wise threshold data in addition to summary measures, reliability metrics and age. 80% of eyes were used for a training set and 20% were used as a test set. MLA test set performance was assessed using the area under the receiver operating curve (AUC). Performance of models trained on initial VF data alone was compared to performance of models trained on data from the first two VFs.

MAIN OUTCOME MEASURES : Accuracy in predicting future rapid progression defined as MD worsening more than 1 dB/year.

RESULTS : 1,968 eyes (8.6%) underwent rapid progression. The support vector machine model (AUC 0.72 [95% CI 0.70-0.75]) most accurately predicted rapid progression when trained on initial VF data. Artificial neural network, random forest, logistic regression and naïve Bayes classifiers produced AUC of 0.72, 0.70, 0.69, 0.68 respectively. Models trained on data from the first two VFs performed no better than top models trained on the initial VF alone. Based on the odds ratio (OR) from logistic regression and variable importance plots from the random forest model, older age (OR: 1.41 per 10 year increment [95% CI: 1.34 to 1.08]) and higher pattern standard deviation (OR: 1.31 per 5-dB increment [95% CI: 1.18 to 1.46]) were the variables in the initial VF most strongly associated with rapid progression.

CONCLUSIONS : MLA can be used to predict eyes at risk for rapid progression with modest accuracy based on an initial VF test. Incorporating additional clinical data to the current model may offer opportunities to predict patients most likely to rapidly progress with even greater accuracy.

Shuldiner Scott R, Boland Michael V, Ramulu Pradeep Y, De Moraes C Gustavo, Elze Tobias, Myers Jonathan, Pasquale Louis, Wellik Sarah, Yohannan Jithin

2021

In PloS one ; h5-index 176.0

Al-Rajab Murad, Lu Joan, Xu Qiang

2021

In PLoS computational biology

Shi Kaiwen, Liu Tao, Fu Hanjiang, Li Wuju, Zheng Xiaofei

2021-Apr-16

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

BACKGROUND : Semantic categorization analysis of clinical trials eligibility criteria based on natural language processing technology is crucial for the task of optimizing clinical trials design and building automated patient recruitment system. However, most of related researches focused on English eligibility criteria, and to the best of our knowledge, there are no researches studied the Chinese eligibility criteria. Thus in this study, we aimed to explore the semantic categories of Chinese eligibility criteria.

METHODS : We downloaded the clinical trials registration files from the website of Chinese Clinical Trial Registry (ChiCTR) and extracted both the Chinese eligibility criteria and corresponding English eligibility criteria. We represented the criteria sentences based on the Unified Medical Language System semantic types and conducted the hierarchical clustering algorithm for the induction of semantic categories. Furthermore, in order to explore the classification performance of Chinese eligibility criteria with our developed semantic categories, we implemented multiple classification algorithms, include four baseline machine learning algorithms (LR, NB, kNN, SVM), three deep learning algorithms (CNN, RNN, FastText) and two pre-trained language models (BERT, ERNIE).

RESULTS : We totally developed 44 types of semantic categories, summarized 8 topic groups, and investigated the average incidence and prevalence in 272 hepatocellular carcinoma related Chinese clinical trials. Compared with the previous proposed categories in English eligibility criteria, 13 novel categories are identified in Chinese eligibility criteria. The classification result shows that most of semantic categories performed quite well, the pre-trained language model ERNIE achieved best performance with macro-average F1 score of 0.7980 and micro-average F1 score of 0.8484.

CONCLUSION : As a pilot study of Chinese eligibility criteria analysis, we developed the 44 semantic categories by hierarchical clustering algorithms for the first times, and validated the classification capacity with multiple classification algorithms.

Zong Hui, Yang Jinxuan, Zhang Zeyu, Li Zuofeng, Zhang Xiaoyan

2021-Apr-15

Classification, Clinical trials, Clustering, Eligibility criteria, Semantic category

In IEEE transactions on neural networks and learning systems

Lou Yang, He Yaodong, Wang Lin, Tsang Kim Fung, Chen Guanrong

2021-Apr-16

In IEEE transactions on neural systems and rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society

Lv Bo, Chai Guohong, Sheng Xinjun, Ding Han, Zhu Xiangyang

2021-Apr-16

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

Wu Gaochang, Liu Yebin, Fang Lu, Chai Tianyou

2021-Apr-16

In La Radiologia medica

Gurgitano Martina, Angileri Salvatore Alessio, Rodà Giovanni Maria, Liguori Alessandro, Pandolfi Marco, Ierardi Anna Maria, Wood Bradford J, Carrafiello Gianpaolo

2021-Apr-16

Artificial intelligence (AI), Augmented reality (AR), Deep learning (DL), Interventional radiology (IR), Machine learning (ML), Virtual reality (VR)

In Archives of pathology & laboratory medicine ; h5-index 49.0

CONTEXT.— : Pathology studies using convolutional neural networks (CNNs) have focused on neoplasms, while studies in inflammatory pathology are rare. We previously demonstrated a CNN differentiates reactive gastropathy, Helicobacter pylori gastritis (HPG), and normal gastric mucosa.

OBJECTIVE.— : To determine whether a CNN can differentiate the following 2 gastric inflammatory patterns: autoimmune gastritis (AG) and HPG.

DESIGN.— : Gold standard diagnoses were blindly established by 2 gastrointestinal (GI) pathologists. One hundred eighty-seven cases were scanned for analysis by HALO-AI. All levels and tissue fragments per slide were included for analysis. The cases were randomized, 112 (60%; 60 HPG, 52 AG) in the training set and 75 (40%; 40 HPG, 35 AG) in the test set. A HALO-AI correct area distribution (AD) cutoff of 50% or more was required to credit the CNN with the correct diagnosis. The test set was blindly reviewed by pathologists with different levels of GI pathology expertise as follows: 2 GI pathologists, 2 general surgical pathologists, and 2 residents. Each pathologist rendered their preferred diagnosis, HPG or AG.

RESULTS.— : At the HALO-AI AD percentage cutoff of 50% or more, the CNN results were 100% concordant with the gold standard diagnoses. On average, autoimmune gastritis cases had 84.7% HALO-AI autoimmune gastritis AD and HP cases had 87.3% HALO-AI HP AD. The GI pathologists, general anatomic pathologists, and residents were on average, 100%, 86%, and 57% concordant with the gold standard diagnoses, respectively.

CONCLUSIONS.— : A CNN can distinguish between cases of HPG and autoimmune gastritis with accuracy equal to GI pathologists.

Franklin Michael M, Schultz Fred A, Tafoya Marissa A, Kerwin Audra A, Broehm Cory J, Fischer Edgar G, Gullapalli Rama R, Clark Douglas P, Hanson Joshua A, Martin David R

2021-Apr-15

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : Asthma affects a large proportion of the population and leads to many hospital encounters involving both hospitalizations and emergency department visits every year. To lower the number of such encounters, many health care systems and health plans deploy predictive models to prospectively identify patients at high risk and offer them care management services for preventive care. However, the previous models do not have sufficient accuracy for serving this purpose well. Embracing the modeling strategy of examining many candidate features, we built a new machine learning model to forecast future asthma hospital encounters of patients with asthma at Intermountain Healthcare, a nonacademic health care system. This model is more accurate than the previously published models. However, it is unclear how well our modeling strategy generalizes to academic health care systems, whose patient composition differs from that of Intermountain Healthcare.

OBJECTIVE : This study aims to evaluate the generalizability of our modeling strategy to the University of Washington Medicine (UWM), an academic health care system.

METHODS : All adult patients with asthma who visited UWM facilities between 2011 and 2018 served as the patient cohort. We considered 234 candidate features. Through a secondary analysis of 82,888 UWM data instances from 2011 to 2018, we built a machine learning model to forecast asthma hospital encounters of patients with asthma in the subsequent 12 months.

RESULTS : Our UWM model yielded an area under the receiver operating characteristic curve (AUC) of 0.902. When placing the cutoff point for making binary classification at the top 10% (1464/14,644) of patients with asthma with the largest forecasted risk, our UWM model yielded an accuracy of 90.6% (13,268/14,644), a sensitivity of 70.2% (153/218), and a specificity of 90.91% (13,115/14,426).

CONCLUSIONS : Our modeling strategy showed excellent generalizability to the UWM, leading to a model with an AUC that is higher than all of the AUCs previously reported in the literature for forecasting asthma hospital encounters. After further optimization, our model could be used to facilitate the efficient and effective allocation of asthma care management resources to improve outcomes.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID) : RR2-10.2196/resprot.5039.

Tong Yao, Messinger Amanda I, Wilcox Adam B, Mooney Sean D, Davidson Giana H, Suri Pradeep, Luo Gang

2021-Apr-16

asthma, forecasting, machine learning, patient care management, risk factors

In JMIR medical informatics ; h5-index 23.0

BACKGROUND : A 12-lead electrocardiogram (ECG) is the most commonly used method to diagnose patients with cardiovascular diseases. However, there are a number of possible misinterpretations of the ECG that can be caused by several different factors, such as the misplacement of chest electrodes.

OBJECTIVE : The aim of this study is to build advanced algorithms to detect precordial (chest) electrode misplacement.

METHODS : In this study, we used traditional machine learning (ML) and deep learning (DL) to autodetect the misplacement of electrodes V1 and V2 using features from the resultant ECG. The algorithms were trained using data extracted from high-resolution body surface potential maps of patients who were diagnosed with myocardial infarction, diagnosed with left ventricular hypertrophy, or a normal ECG.

RESULTS : DL achieved the highest accuracy in this study for detecting V1 and V2 electrode misplacement, with an accuracy of 93.0% (95% CI 91.46-94.53) for misplacement in the second intercostal space. The performance of DL in the second intercostal space was benchmarked with physicians (n=11 and age 47.3 years, SD 15.5) who were experienced in reading ECGs (mean number of ECGs read in the past year 436.54, SD 397.9). Physicians were poor at recognizing chest electrode misplacement on the ECG and achieved a mean accuracy of 60% (95% CI 56.09-63.90), which was significantly poorer than that of DL (P<.001).

CONCLUSIONS : DL provides the best performance for detecting chest electrode misplacement when compared with the ability of experienced physicians. DL and ML could be used to help flag ECGs that have been incorrectly recorded and flag that the data may be flawed, which could reduce the number of erroneous diagnoses.

Rjoob Khaled, Bond Raymond, Finlay Dewar, McGilligan Victoria, J Leslie Stephen, Rababah Ali, Iftikhar Aleeha, Guldenring Daniel, Knoery Charles, McShane Anne, Peace Aaron

2021-Apr-16

ECG, ECG interpretation, cardiovascular disease, deep learning, electrode misplacement, engineering, feature engineering, machine learning, myocardial, myocardial infarction, physicians

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : Acute kidney injury (AKI) is commonly encountered in clinical practice and is associated with poor patient outcomes and increased health care costs. Despite it posing significant challenges for clinicians, effective measures for AKI prediction and prevention are lacking. Previously published AKI prediction models mostly have a simple design without external validation. Furthermore, little is known about the process of linking model output and clinical decisions due to the black-box nature of neural network models.

OBJECTIVE : We aimed to present an externally validated recurrent neural network (RNN)-based continuous prediction model for in-hospital AKI and show applicable model interpretations in relation to clinical decision support.

METHODS : Study populations were all patients aged 18 years or older who were hospitalized for more than 48 hours between 2013 and 2017 in 2 tertiary hospitals in Korea (Seoul National University Bundang Hospital and Seoul National University Hospital). All demographic data, laboratory values, vital signs, and clinical conditions of patients were obtained from electronic health records of each hospital. We developed 2-stage hierarchical prediction models (model 1 and model 2) using RNN algorithms. The outcome variable for model 1 was the occurrence of AKI within 7 days from the present. Model 2 predicted the future trajectory of creatinine values up to 72 hours. The performance of each developed model was evaluated using the internal and external validation data sets. For the explainability of our models, different model-agnostic interpretation methods were used, including Shapley Additive Explanations, partial dependence plots, individual conditional expectation, and accumulated local effects plots.

RESULTS : We included 69,081 patients in the training, 7675 in the internal validation, and 72,352 in the external validation cohorts for model development after excluding cases with missing data and those with an estimated glomerular filtration rate less than 15 mL/min/1.73 m2 or end-stage kidney disease. Model 1 predicted any AKI development with an area under the receiver operating characteristic curve (AUC) of 0.88 (internal validation) and 0.84 (external validation), and stage 2 or higher AKI development with an AUC of 0.93 (internal validation) and 0.90 (external validation). Model 2 predicted the future creatinine values within 3 days with mean-squared errors of 0.04-0.09 for patients with higher risks of AKI and 0.03-0.08 for those with lower risks. Based on the developed models, we showed AKI probability according to feature values in total patients and each individual with partial dependence, accumulated local effects, and individual conditional expectation plots. We also estimated the effects of feature modifications such as nephrotoxic drug discontinuation on future creatinine levels.

CONCLUSIONS : We developed and externally validated a continuous AKI prediction model using RNN algorithms. Our model could provide real-time assessment of future AKI occurrences and individualized risk factors for AKI in general inpatient cohorts; thus, we suggest approaches to support clinical decisions based on prediction models for in-hospital AKI.

Kim Kipyo, Yang Hyeonsik, Yi Jinyeong, Son Hyung-Eun, Ryu Ji-Young, Kim Yong Chul, Jeong Jong Cheol, Chin Ho Jun, Na Ki Young, Chae Dong-Wan, Han Seung Seok, Kim Sejoong

2021-Apr-16

acute kidney injury, external validation, internal validation, kidney, neural networks, prediction model, recurrent neural network

In Journal of veterinary medical education

Boller Elise, Courtman Natalie, Chiavaroli Neville, Beck Catherine

2021-Apr

clinical integrative puzzle, clinical reasoning, collaborative learning, extended matching question

In ACS nano ; h5-index 203.0

Kalinin Sergei V, Zhang Shuai, Valleti Mani, Pyles Harley, Baker David, De Yoreo James J, Ziatdinov Maxim

2021-Apr-16

deep learning, latent space models, representation learning, self-assembly, variational autoencoder

In BMC bioinformatics

BACKGROUND : The Cox proportional hazards model is commonly used to predict hazard ratio, which is the risk or probability of occurrence of an event of interest. However, the Cox proportional hazard model cannot directly generate an individual survival time. To do this, the survival analysis in the Cox model converts the hazard ratio to survival times through distributions such as the exponential, Weibull, Gompertz or log-normal distributions. In other words, to generate the survival time, the Cox model has to select a specific distribution over time.

RESULTS : This study presents a method to predict the survival time by integrating hazard network and a distribution function network. The Cox proportional hazards network is adapted in DeepSurv for the prediction of the hazard ratio and a distribution function network applied to generate the survival time. To evaluate the performance of the proposed method, a new evaluation metric that calculates the intersection over union between the predicted curve and ground truth was proposed. To further understand significant prognostic factors, we use the 1D gradient-weighted class activation mapping method to highlight the network activations as a heat map visualization over an input data. The performance of the proposed method was experimentally verified and the results compared to other existing methods.

CONCLUSIONS : Our results confirmed that the combination of the two networks, Cox proportional hazards network and distribution function network, can effectively generate accurate survival time.

Baek Eu-Tteum, Yang Hyung Jeong, Kim Soo Hyung, Lee Guee Sang, Oh In-Jae, Kang Sae-Ryung, Min Jung-Joon

2021-Apr-15

Cox proportional hazards network, Deep learning, Distribution function network, Prognosis, Survival time prediction

In Medical image analysis

Signoroni Alberto, Savardi Mattia, Benini Sergio, Adami Nicola, Leonardi Riccardo, Gibellini Paolo, Vaccher Filippo, Ravanelli Marco, Borghesi Andrea, Maroldi Roberto, Farina Davide

2021-Mar-31

COVID-19 severity assessment, Chest X-rays, Convolutional neural networks, End-to-end learning, Semi-quantitative rating

In Infectious diseases and therapy

INTRODUCTION : We aimed to assess risk factors for multidrug-resistant Gram-negative bacilli (MDR-GNB) from a large amount of data retrieved from electronic health records (EHRs) and determine whether machine learning (ML) may be useful in assessing the risk of MDR-GNB infection at febrile neutropenia (FN) onset.

METHODS : Retrospective study of almost 7 million pieces of structured data from all consecutive episodes of FN in hematological patients in a tertiary hospital in Barcelona (January 2008-December 2017). Conventional multivariate analysis and ML algorithms (random forest, gradient boosting machine, XGBoost, and GLM) were done.

RESULTS : A total of 3235 episodes of FN in 349 patients were documented; MDR-GNB caused 180 (5.6%) infections in 132 patients. The most frequent MDR-GNBs were MDR-Pseudomonas aeruginosa (53%) and extended-spectrum beta-lactamase-producing Enterobacterales (46%). According to conventional logistic regression analysis, independent factors associated with MDR-GNB infection were age older than 45 years (OR 2.07; 95% CI 1.31-3.24), prior antibiotics (2.62; 1.39-4.92), first-ever FN in this hospitalization (2.94; 1.33-6.52), prior hospitalizations for FN (1.72; 1.02-2.89); at least 15 prior hospital visits (2.65; 1.31-5.33), high-risk hematological diseases (3.62; 1.12-11.67), and hospitalization in a room formerly occupied by patients with MDR-GNB isolation (1.69; 1.20-2.38). ML algorithms achieved the following AUC and F1 score for MDR-GNB prediction: random forest, 0.79-0.9711; GMB, 0.79-0.9705; XGBoost, 0.79-0.9670; and GLM, 0.78-0.9716.

CONCLUSION : Data generated in EHRs proved useful in assessing risk factors for MDR-GNB infections in patients with FN. The great number of analyzed variables allowed us to identify new factors related to MDR infection, as well as to train ML algorithms for infection predictions. This information may be used by clinicians to make better clinical decisions.

Garcia-Vidal Carolina, Puerta-Alcalde Pedro, Cardozo Celia, Orellana Miquel A, Besanson Gaston, Lagunas Jaime, Marco Francesc, Del Rio Ana, Martínez Jose A, Chumbita Mariana, Garcia-Pouton Nicole, Mensa Josep, Rovira Montserrat, Esteve Jordi, Soriano Alex

2021-Apr-16

Electronic health records, Machine learning, Multiresistance, Neutropenia

In European radiology ; h5-index 62.0

OBJECTIVES : To perform a systematic review of design and reporting of imaging studies applying convolutional neural network models for radiological cancer diagnosis.

METHODS : A comprehensive search of PUBMED, EMBASE, MEDLINE and SCOPUS was performed for published studies applying convolutional neural network models to radiological cancer diagnosis from January 1, 2016, to August 1, 2020. Two independent reviewers measured compliance with the Checklist for Artificial Intelligence in Medical Imaging (CLAIM). Compliance was defined as the proportion of applicable CLAIM items satisfied.

RESULTS : One hundred eighty-six of 655 screened studies were included. Many studies did not meet the criteria for current design and reporting guidelines. Twenty-seven percent of studies documented eligibility criteria for their data (50/186, 95% CI 21-34%), 31% reported demographics for their study population (58/186, 95% CI 25-39%) and 49% of studies assessed model performance on test data partitions (91/186, 95% CI 42-57%). Median CLAIM compliance was 0.40 (IQR 0.33-0.49). Compliance correlated positively with publication year (ρ = 0.15, p = .04) and journal H-index (ρ = 0.27, p < .001). Clinical journals demonstrated higher mean compliance than technical journals (0.44 vs. 0.37, p < .001).

CONCLUSIONS : Our findings highlight opportunities for improved design and reporting of convolutional neural network research for radiological cancer diagnosis.

KEY POINTS : • Imaging studies applying convolutional neural networks (CNNs) for cancer diagnosis frequently omit key clinical information including eligibility criteria and population demographics. • Fewer than half of imaging studies assessed model performance on explicitly unobserved test data partitions. • Design and reporting standards have improved in CNN research for radiological cancer diagnosis, though many opportunities remain for further progress.

O’Shea Robert J, Sharkey Amy Rose, Cook Gary J R, Goh Vicky

2021-Apr-16

Artificial intelligence, Deep learning, Diagnosis, computer-assisted, Neoplasms, Research design

In European radiology ; h5-index 62.0

OBJECTIVES : To examine the various roles of radiologists in different steps of developing artificial intelligence (AI) applications.

MATERIALS AND METHODS : Through the case study of eight companies active in developing AI applications for radiology, in different regions (Europe, Asia, and North America), we conducted 17 semi-structured interviews and collected data from documents. Based on systematic thematic analysis, we identified various roles of radiologists. We describe how each role happens across the companies and what factors impact how and when these roles emerge.

RESULTS : We identified 9 roles that radiologists play in different steps of developing AI applications: (1) problem finder (in 4 companies); (2) problem shaper (in 3 companies); (3) problem dominator (in 1 company); (4) data researcher (in 2 companies); (5) data labeler (in 3 companies); (6) data quality controller (in 2 companies); (7) algorithm shaper (in 3 companies); (8) algorithm tester (in 6 companies); and (9) AI researcher (in 1 company).

CONCLUSIONS : Radiologists can play a wide range of roles in the development of AI applications. How actively they are engaged and the way they are interacting with the development teams significantly vary across the cases. Radiologists need to become proactive in engaging in the development process and embrace new roles.

KEY POINTS : • Radiologists can play a wide range of roles during the development of AI applications. • Both radiologists and developers need to be open to new roles and ways of interacting during the development process. • The availability of resources, time, expertise, and trust are key factors that impact how actively radiologists play roles in the development process.

Scheek Damian, Rezazade Mehrizi Mohammad H, Ranschaert Erik

2021-Apr-16

AI, Artificial intelligence, Development, Radiologists, Roles

In eLife

Schmidt Calvin M, Smolke Christina D

2021-Apr-16

S. cerevisiae, computational biology, systems biology

In Future cardiology

Sotirakos Sara, Fouda Basem, Mohamed Razif Noor Adeebah, Cribben Niall, Mulhall Cormac, O’Byrne Aisling, Moran Bridget, Connolly Ruairi

2021-Apr-16

arrhythmia, artificial intelligence, cardiac disease, cardiac rehabilitation, heart failure, smartwatch

In American journal of physical anthropology

OBJECTIVES : Convolutional neural network (CNN) is a state-of-art deep learning (DL) method with superior performance in image classification. Here, a CNN-based workflow is proposed to discriminate hominid teeth. Our hope is that this method could help confirm otherwise questionable records of Homo from Pleistocene deposits where there is a standing risk of mis-attributing molars of Pongo to Homo.

METHODS AND MATERIALS : A two-step workflow was designed. The first step is converting the enamel-dentine junction (EDJ) into EDJ card, that is, a two-dimensional image conversion of the three-dimensional EDJ surface. In this step, researchers must carefully orient the teeth according to the cervical plane. The second step is training the CNN learner with labeled EDJ cards. A sample consisting of 53 fossil Pongo and 53 Homo (modern human and Neanderthal) was adopted to generate EDJ cards, which were then separated into training set (n = 84) and validation set (n = 22). To assess the feasibility of this workflow, a Pongo-Homo classifier was trained from the aforementioned EDJ card set, and then the classifier was used to predict the taxonomic affinities of six samples (test set) from von Koenigswald's Chinese Apothecary collection.

RESULTS : Results show that EDJ cards in validation set are classified accurately by the CNN learner. More importantly, taxonomic predictions for six specimens in test set match well with the diagnosis results deduced from multiple lines of evidence, implying the great potential of CNN method.

DISCUSSION : This workflow paves a way for future studies using CNN to address taxonomic complexity (e.g., distinguishing Pongo and Homo teeth from the Pleistocene of Asia). Further improvements include visual interpretation and extending the applicability to moderately worn teeth.

Yi Zhixing, Zanolli Clément, Liao Wei, Wang Wei

2021-Apr-16

convolutional neural network, deep learning, hominid teeth, taxonomy

In Journal of clinical monitoring and computing

Nistal-Nuño Beatriz

2021-Apr-15

Cardiac Surgery Recovery Unit, Coronary Care Unit, Decision-support systems, Machine learning, Mortality

In Neurological sciences : official journal of the Italian Neurological Society and of the Italian Society of Clinical Neurophysiology

BACKGROUND : Determining the rupture risk of unruptured intracranial aneurysm is crucial for treatment strategy. The purpose of this study was to predict the rupture risk of middle cerebral artery (MCA) aneurysms using a machine learning technique.

METHODS : We retrospectively reviewed 403 MCA aneurysms and randomly partitioned them into the training and testing datasets with a ratio of 8:2. A generalized linear model with logit link was developed using training dataset to predict the aneurysm rupture risk based on the clinical variables and morphological features manually measured from computed tomography angiography. To facilitate the clinical application, we further constructed an easy-to-use nomogram based on the developed model.

RESULTS : Ruptured MCA aneurysm had larger aneurysm size, aneurysm height, perpendicular height, aspect ratio, size ratio, bottleneck factor, and height-width ratio. Presence of a daughter-sac was more common in ruptured than in unruptured MCA aneurysms. Six features, including aneurysm multiplicity, lobulations, size ratio, bottleneck factor, height-width ratio, and aneurysm angle, were adopted in the model after feature selection. The model achieved a relatively good performance with areas under the receiver operating characteristic curves of 0.77 in the training dataset and 0.76 in the testing dataset. The nomogram provided a visual interpretation of our model, and the rupture risk probability of MCA aneurysms can be directly read from it.

CONCLUSION : Our model can be used to predict the rupture risk of MCA aneurysm.

Liu Jinjin, Chen Yongchun, Zhu Dongqin, Li Qiong, Chen Zhonggang, Zhou Jiafeng, Lin Boli, Yang Yunjun, Jia Xiufen

2021-Apr-15

Aneurysm, Computed tomography angiography, Morphology, Prediction, Rupture

In Chemical senses

Gerkin Richard C

2021-Apr-16

In Wellcome open research

Belesova Kristine, Callaghan Max, Minx Jan C, Creutzig Felix, Turcu Catalina, Hutchinson Emma, Milner James, Crane Melanie, Haines Andy, Davies Michael, Wilkinson Paul

2021

actions, adaptation, case studies, cities, climate action, climate change, evaluation, implementation, intervention, mitigation, planetary health, public health, solutions, urban health, wellbeing

In Journal of hazardous materials

Nam Ki Jeon, Li Qian, Heo Sung Ku, Tariq Shahzeb, Loy-Benitez Jorge, Woo Tae Yong, Yoo Chang Kyoo

2021-Jun-05

Climate change, Deep learning, Dynamic fate analysis, Polycyclic aromatic hydrocarbons, Potential risk

In Scientific reports ; h5-index 158.0

Tran Nam K, Howard Taylor, Walsh Ryan, Pepper John, Loegering Julia, Phinney Brett, Salemi Michelle R, Rashidi Hooman H

2021-Apr-15

In BioMed research international ; h5-index 102.0

Objective : To establish a machine learning model for identifying patients coinfected with hepatitis B virus (HBV) and human immunodeficiency virus (HIV) through two sexual transmission routes in Jiangsu, China.

Methods : A total of 14197 HIV cases transmitted by homosexual and heterosexual routes were recruited. After data processing, 12469 cases (HIV and HBV, 1033; HIV, 11436) were left for further analysis, including 7849 cases with homosexual transmission and 4620 cases with heterosexual transmission. Univariate logistic regression was used to select variables with significant P value and odds ratio for multivariable analysis. In homosexual transmission and heterosexual transmission groups, 10 and 6 variables were selected, respectively. For identifying HIV individuals coinfected with HBV, a machine learning model was constructed with four algorithms, including Decision Tree, Random Forest, AdaBoost with decision tree (AdaBoost), and extreme gradient boosting decision tree (XGBoost). The detective value of each variable was calculated using the optimal machine learning algorithm.

Results : AdaBoost algorithm showed the highest efficiency in both transmission groups (homosexual transmission group: accuracy = 0.928, precision = 0.915, recall = 0.944, F - 1 = 0.930, and AUC = 0.96; heterosexual transmission group: accuracy = 0.892, precision = 0.881, recall = 0.905, F - 1 = 0.893, and AUC = 0.98). Calculated by AdaBoost algorithm, the detective value of PLA was the highest in homosexual transmission group, followed by CR, AST, HB, ALT, TBIL, leucocyte, age, marital status, and treatment condition; in the heterosexual transmission group, the detective value of PLA was the highest (consistent with the condition in the homosexual group), followed by ALT, AST, TBIL, leucocyte, and symptom severity.

Conclusions : The univariate logistics regression combined with the AdaBoost algorithm could accurately screen the risk factors of HBV in HIV coinfection without invasive testing. Further studies are needed to evaluate the utility and feasibility of this model in various settings.

Yin Yi, Xue Mingyue, Shi Lingen, Qiu Tao, Xia Derun, Fu Gengfeng, Peng Zhihang

2021

In Journal of healthcare engineering

Khan Taha Muthar, Xu Shengjun, Khan Zullatun Gull, Uzair Chishti Muhammad

2021

In Computational and mathematical methods in medicine

Andrade Evandro, Portela Samuel, Pinheiro Plácido Rogério, Nunes Luciano Comin, Filho Marum Simão, Costa Wagner Silva, Pinheiro Mirian Caliope Dantas

2021

In Evidence-based complementary and alternative medicine : eCAM

Xu Tiancheng, Xia Youbing

2021

In Computational intelligence and neuroscience

Zhang Junming, Tang Zhen, Gao Jinfeng, Lin Li, Liu Zhiliang, Wu Haitao, Liu Fang, Yao Ruxian

2021

In Frontiers in plant science

Schirrmann Michael, Landwehr Niels, Giebel Antje, Garz Andreas, Dammer Karl-Heinz

2021

ResNet, camera sensor, deep learning, image recognition, monitoring, smart farming, wheat crops, yellow rust

In Signal processing. Image communication

Liaqat Sidrah, Wu Chongruo, Duggirala Prashanth Reddy, Cheung Sen-Ching Samson, Chuah Chen-Nee, Ozonoff Sally, Young Gregory

2021-May

Autism Spectrum Disorders, Deep Learning, Eye Gaze Data

In Annals of statistics

Abbe Emmanuel, Fan Jianqing, Wang Kaizheng, Zhong Yiqiao

2020-Jun

62H12, Primary 62H25, community detection, eigenvector perturbation, low-rank structures, matrix completion, random matrices, secondary 60B20, spectral analysis, synchronization

In Nature medicine ; h5-index 170.0

Gehrung Marcel, Crispin-Ortuzar Mireia, Berman Adam G, O’Donovan Maria, Fitzgerald Rebecca C, Markowetz Florian

2021-Apr-15

In Nature biomedical engineering

Mason Derek M, Friedensohn Simon, Weber Cédric R, Jordi Christian, Wagner Bastian, Meng Simon M, Ehling Roy A, Bonati Lucia, Dahinden Jan, Gainza Pablo, Correia Bruno E, Reddy Sai T

2021-Apr-15

In Journal of hazardous materials

Roostaei Javad, Colley Sarah, Mulhern Riley, May Andrew A, Gibson Jacqueline MacDonald

2021-Jun-05

Bayesian network, Drinking water, GenX, Machine-learning, PFAS, Well water

In BJPsych open

Cobo Aurora, Porras-Segovia Alejandro, Pérez-Rodríguez María Mercedes, Artés-Rodríguez Antonio, Barrigón Maria Luisa, Courtet Philippe, Baca-García Enrique

2021-Apr-16

COVID-19, Suicide, ecological momentary assessment, machine learning, suicide attempt

In NPJ digital medicine

Perslev Mathias, Darkner Sune, Kempfner Lykke, Nikolic Miki, Jennum Poul Jørgen, Igel Christian

2021-Apr-15

In Scientific reports ; h5-index 158.0

Stuit S M, Kootstra T M, Terburg D, van den Boomen C, van der Smagt M J, Kenemans J L, Van der Stigchel S

2021-Apr-15

In Scientific reports ; h5-index 158.0

Hadaeghi Fatemeh, Diercks Björn-Philipp, Schetelig Daniel, Damicelli Fabrizio, Wolf Insa M A, Werner René

2021-Apr-15

In Scientific reports ; h5-index 158.0

Chapman Ross, Cock James, Samson Marianne, Janetski Noel, Janetski Kate, Gusyana Dadang, Dutta Sudarshan, Oberthür Thomas

2021-Apr-15

In Scientific reports ; h5-index 158.0

Bhagirath Divya, Liston Michael, Akoto Theresa, Lui Byron, Bensing Barbara A, Sharma Ashok, Saini Sharanjot

2021-Apr-15

In Scientific reports ; h5-index 158.0

Tran Nam K, Howard Taylor, Walsh Ryan, Pepper John, Loegering Julia, Phinney Brett, Salemi Michelle R, Rashidi Hooman H

2021-Apr-15

In Thorax ; h5-index 75.0

BACKGROUND : The optimal interface for the delivery of home non-invasive ventilation (NIV) to treat chronic respiratory failure has not yet been determined. The aim of this individual participant data (IPD) meta-analysis was to compare the effect of nasal and oronasal masks on treatment efficacy and adherence in patients with COPD and obesity hypoventilation syndrome (OHS).

METHODS : We searched Medline and Cochrane Central Register of Controlled Trials for prospective randomised controlled trials (RCTs) of at least 1 month's duration, published between January 1994 and April 2019, that assessed NIV efficacy in patients with OHS and COPD. The main outcomes were diurnal PaCO₂, PaO₂ and NIV adherence (PROSPERO CRD42019132398).

FINDINGS : Of 1576 articles identified, 34 RCTs met the inclusion criteria and IPD were obtained for 18. Ten RCTs were excluded because only one type of mask was used, or mask data were missing. Data from 8 RCTs, including 290 IPD, underwent meta-analysis. Oronasal masks were used in 86% of cases. There were no differences between oronasal and nasal masks for PaCO₂ (0.61 mm Hg (95% CI -2.15 to 3.38); p=0.68), PaO₂ (-0.00 mm Hg (95% CI -4.59 to 4.58); p=1) or NIV adherence (0·29 hour/day (95% CI -0.74 to 1.32); p=0.58). There was no interaction between the underlying pathology and the effect of mask type on any outcome.

INTERPRETATION : Oronasal masks are the most used interface for the delivery of home NIV in patients with OHS and COPD; however, there is no difference in the efficacy or tolerance of oronasal or nasal masks.

Lebret Marius, Léotard Antoine, Pépin Jean Louis, Windisch Wolfram, Ekkernkamp Emelie, Pallero Mercedes, Sánchez-Quiroga M-Ángeles, Hart Nicholas, Kelly Julia L, Patout Maxime, Funk Georg Chistian, Duiverman Marieke L, Masa Juan F, Simonds Anita, Murphy Patrick Brian, Wijkstra Peter J, Dreher Michael, Storre Jan, Khouri Charles, Borel Jean-Christian

2021-Apr-15

COPD pathology, non invasive ventilation, sleep apnoea

In Journal for immunotherapy of cancer

BACKGROUND : The human leucocyte antigen (HLA) complex controls adaptive immunity by presenting defined fractions of the intracellular and extracellular protein content to immune cells. Understanding the benign HLA ligand repertoire is a prerequisite to define safe T-cell-based immunotherapies against cancer. Due to the poor availability of benign tissues, if available, normal tissue adjacent to the tumor has been used as a benign surrogate when defining tumor-associated antigens. However, this comparison has proven to be insufficient and even resulted in lethal outcomes. In order to match the tumor immunopeptidome with an equivalent counterpart, we created the HLA Ligand Atlas, the first extensive collection of paired HLA-I and HLA-II immunopeptidomes from 227 benign human tissue samples. This dataset facilitates a balanced comparison between tumor and benign tissues on HLA ligand level.

METHODS : Human tissue samples were obtained from 16 subjects at autopsy, five thymus samples and two ovary samples originating from living donors. HLA ligands were isolated via immunoaffinity purification and analyzed in over 1200 liquid chromatography mass spectrometry runs. Experimentally and computationally reproducible protocols were employed for data acquisition and processing.

RESULTS : The initial release covers 51 HLA-I and 86 HLA-II allotypes presenting 90,428 HLA-I- and 142,625 HLA-II ligands. The HLA allotypes are representative for the world population. We observe that immunopeptidomes differ considerably between tissues and individuals on source protein and HLA-ligand level. Moreover, we discover 1407 HLA-I ligands from non-canonical genomic regions. Such peptides were previously described in tumors, peripheral blood mononuclear cells (PBMCs), healthy lung tissues and cell lines. In a case study in glioblastoma, we show that potential on-target off-tumor adverse events in immunotherapy can be avoided by comparing tumor immunopeptidomes to the provided multi-tissue reference.

CONCLUSION : Given that T-cell-based immunotherapies, such as CAR-T cells, affinity-enhanced T cell transfer, cancer vaccines and immune checkpoint inhibition, have significant side effects, the HLA Ligand Atlas is the first step toward defining tumor-associated targets with an improved safety profile. The resource provides insights into basic and applied immune-associated questions in the context of cancer immunotherapy, infection, transplantation, allergy and autoimmunity. It is publicly available and can be browsed in an easy-to-use web interface at https://hla-ligand-atlas.org .

Marcu Ana, Bichmann Leon, Kuchenbecker Leon, Kowalewski Daniel Johannes, Freudenmann Lena Katharina, Backert Linus, Mühlenbruch Lena, Szolek András, Lübke Maren, Wagner Philipp, Engler Tobias, Matovina Sabine, Wang Jian, Hauri-Hohl Mathias, Martin Roland, Kapolou Konstantina, Walz Juliane Sarah, Velz Julia, Moch Holger, Regli Luca, Silginer Manuela, Weller Michael, Löffler Markus W, Erhard Florian, Schlosser Andreas, Kohlbacher Oliver, Stevanović Stefan, Rammensee Hans-Georg, Neidert Marian Christoph

2021-Apr

adaptive immunity, antigen presentation, antigens, carbohydrate, immunotherapy, translational medical research, tumor-associated

In JMIR medical informatics ; h5-index 23.0

BACKGROUND : Federated learning (FL) is a decentralized approach to machine learning which is attracting attention as a training strategy that overcomes medical data privacy regulations and generalization of deep learning algorithms. FL mitigates many systemic privacy risks by sharing only the model and parameters for training, without the need to export existing medical datasets. In this study, we performed ultrasound (US) image analysis using FL to predict the benignity or malignancy of thyroid nodules.

OBJECTIVE : The goal of this study was to evaluate whether the performance of FL is comparable with that of conventional deep learning.

METHODS : A total of 8,457 (5,375 malignant, 3,082 benign) US images were collected from six institutions, and used for FL and conventional deep learning. Five deep learning networks (VGG19, ResNet50, ResNext50, SE-ResNet50, and SE-ResNext50) were used. Using stratified random sampling, we selected 20% (1,075 malignant, 616 benign) of the total images for internal validation. For external validation, we used 100 US images (50 malignant, 50 benign) from another institution.

RESULTS : For internal validation, the area under the receiver operating characteristic (AUROC) curve of the FL was between 78.88 and 87.56, and the AUC of the conventional deep learning was between 82.61 and 91.57. For external validation, the AUROC of the FL was between 75.20 and 86.72, and the AUROC curve of the conventional deep learning was between 73.04 and 91.04.

CONCLUSIONS : We demonstrated the performance of FL using decentralized data was comparable to that of conventional deep learning using pooled data. FL might be potentially useful for analyzing medical images while protecting patients' personal information.

Lee Haeyun, Chai Young Jun, Joo Hyunjin, Lee Kyungsu, Hwang Jae Youn, Kim Seok-Mo, Kim Kwangsoon, Nam Inn-Chul, Choi June Young, Yu Hyeong Won, Lee Myung-Chul, Masuoka Hiroo, Miyauchi Akira, Lee Kyu Eun, Kim Sungwan, Kong Hyoun-Joong

2021-Apr-03

In The Lancet. Digital health

BACKGROUND : Survival of liver transplant recipients beyond 1 year since transplantation is compromised by an increased risk of cancer, cardiovascular events, infection, and graft failure. Few clinical tools are available to identify patients at risk of these complications, which would flag them for screening tests and potentially life-saving interventions. In this retrospective analysis, we aimed to assess the ability of deep learning algorithms of longitudinal data from two prospective cohorts to predict complications resulting in death after liver transplantation over multiple timeframes, compared with logistic regression models.

METHODS : In this machine learning analysis, model development was done on a set of 42 146 liver transplant recipients (mean age 48·6 years [SD 17·3]; 17 196 [40·8%] women) from the Scientific Registry of Transplant Recipients (SRTR) in the USA. Transferability of the model was further evaluated by fine-tuning on a dataset from the University Health Network (UHN) in Canada (n=3269; mean age 52·5 years [11·1]; 1079 [33·0%] women). The primary outcome was cause of death, as recorded in the databases, due to cardiovascular causes, infection, graft failure, or cancer, within 1 year and 5 years of each follow-up examination after transplantation. We compared the performance of four deep learning models against logistic regression, assessing performance using the area under the receiver operating characteristic curve (AUROC).

FINDINGS : In both datasets, deep learning models outperformed logistic regression, with the Transformer model achieving the highest AUROCs in both datasets (p<0·0001). The AUROC for the Transformer model across all outcomes in the SRTR dataset was 0·804 (99% CI 0·795-0·854) for 1-year predictions and 0·733 (0·729-0·769) for 5-year predictions. In the UHN dataset, the AUROC for the top-performing deep learning model was 0·807 (0·795-0·842) for 1-year predictions and 0·722 (0·705-0·764) for 5-year predictions. AUROCs ranged from 0·695 (0·680-0·713) for prediction of death from infection within 5 years to 0·859 (0·847-0·871) for prediction of death by graft failure within 1 year.

INTERPRETATION : Deep learning algorithms can incorporate longitudinal information to continuously predict long-term outcomes after liver transplantation, outperforming logistic regression models. Physicians could use these algorithms at routine follow-up visits to identify liver transplant recipients at risk for adverse outcomes and prevent these complications by modifying management based on ranked features.

FUNDING : Canadian Donation and Transplant Research Program, CIFAR AI Chairs Program.

Nitski Osvald, Azhie Amirhossein, Qazi-Arisar Fakhar Ali, Wang Xueqi, Ma Shihao, Lilly Leslie, Watt Kymberly D, Levitsky Josh, Asrani Sumeet K, Lee Douglas S, Rubin Barry B, Bhat Mamatha, Wang Bo

2021-Apr-09

In Scientific reports ; h5-index 158.0

Kanavati Fahdi, Toyokawa Gouji, Momosaki Seiya, Takeoka Hiroaki, Okamoto Masaki, Yamazaki Koji, Takeo Sadanori, Iizuka Osamu, Tsuneki Masayuki

2021-Apr-14

In Scientific reports ; h5-index 158.0

Kaufmann Kevin, Lane Hobson, Liu Xiao, Vecchio Kenneth S

2021-Apr-14

In Scientific reports ; h5-index 158.0

Borovsky Arielle, Thal Donna, Leonard Laurence B

2021-Apr-14

In BMJ health & care informatics

OBJECTIVE : To examine how and to what extent medical devices using machine learning (ML) support clinician decision making.

METHODS : We searched for medical devices that were (1) approved by the US Food and Drug Administration (FDA) up till February 2020; (2) intended for use by clinicians; (3) in clinical tasks or decisions and (4) used ML. Descriptive information about the clinical task, device task, device input and output, and ML method were extracted. The stage of human information processing automated by ML-based devices and level of autonomy were assessed.

RESULTS : Of 137 candidates, 59 FDA approvals for 49 unique devices were included. Most approvals (n=51) were since 2018. Devices commonly assisted with diagnostic (n=35) and triage (n=10) tasks. Twenty-three devices were assistive, providing decision support but left clinicians to make important decisions including diagnosis. Twelve automated the provision of information (autonomous information), such as quantification of heart ejection fraction, while 14 automatically provided task decisions like triaging the reading of scans according to suspected findings of stroke (autonomous decisions). Stages of human information processing most automated by devices were information analysis, (n=14) providing information as an input into clinician decision making, and decision selection (n=29), where devices provide a decision.

CONCLUSION : Leveraging the benefits of ML algorithms to support clinicians while mitigating risks, requires a solid relationship between clinician and ML-based devices. Such relationships must be carefully designed, considering how algorithms are embedded in devices, the tasks supported, information provided and clinicians' interactions with them.

Lyell David, Coiera Enrico, Chen Jessica, Shah Parina, Magrabi Farah

2021-Apr

medical informatics

In BMJ health & care informatics

OBJECTIVES : To conduct systematic review and meta-analysis of interventional studies to investigate the impact of diabetes self-management education and support (DSMES) apps on adherence in patients with type 2 diabetes mellitus (T2D).

METHODS : PubMed, Embase, CENTRAL, Web of Science, Scopus and ProQuest were searched, in addition to references of identified articles and similar reviews. Experimental studies, reported in English, assessing DSMES app intervention's impact on adherence and clinical outcomes of patients with T2D compared with usual care were included. Study bias was assessed using Cochrane Risk of Bias V.2.0 tool. Analysis plan involved narrative synthesis, moderator and meta-analysis.

RESULTS : Six randomised controlled trials were included, involving 696 participants (average age 57.6 years, SD 10.59). Mobile apps were mostly used for imputing clinical data, dietary intake or physical activity, and transmitting information to the provider. At 3 months, DSMES apps proved effective in improving medication adherence (standardized mean difference (SMD)=0.393, 95% CI 0.17 to 0.61), glycated haemoglobin (HbA1c) (mean difference (MD)=-0.314, 95% CI -0.477 to -0.151) and Body Mass Index (BMI) (MD=-0.28, 95% CI -0.545 to -0.015). All pooled estimates had low heterogeneity (I² 0%). Four studies had moderate risk of bias while one each was judged to be low and high risks, respectively.

CONCLUSION : DSMES apps had significant small to moderate effects on medication adherence, HbA1c and BMI of patients with T2D compared with usual care. Apps were described as reliable, easy to use and convenient, though participants were required to be phone literate. Evidence comes from feasibility trials with generally moderate risk of bias. Larger trials with longer follow-up periods using theory-based interventions are required to improve current evidence.

Enricho Nkhoma Dumisani, Jenya Soko Charles, Joseph Banda Kondwani, Greenfield David, Li Yu-Chuan Jack, Iqbal Usman

2021-Apr

patient care, primary health care, public health

ArXiv Preprint

Zach Wood-Doughty, Isabel Cachola, Mark Dredze

2021-04-16

In Chinese medical sciences journal = Chung-kuo i hsueh k'o hsueh tsa chih

Li Da Sheng, Wang Da Wei, Wang Na Na, Xu Hai Wang, Huang He, Dong Jian Ping, Xia Chen

2021-Mar-31

ArXiv Preprint

Abanoub M. Girgis, Hyowoon Seo, Jihong Park, Mehdi Bennis, Jinho Choi

2021-04-16

In BMC anesthesiology ; h5-index 31.0

BACKGROUND : The intrathecal hyperbaric bupivacaine dosage for cesarean section is difficult to predetermine. This study aimed to develop a decision-support model using a machine-learning algorithm for assessing intrathecal hyperbaric bupivacaine dose based on physical variables during cesarean section.

METHODS : Term parturients presenting for elective cesarean section under spinal anaesthesia were enrolled. Spinal anesthesia was performed at the L3/4 interspace with 0.5% hyperbaric bupivacaine at dosages determined by the anesthesiologist. A spinal spread level between T4-T6 was considered the appropriate block level. We used a machine-learning algorithm to identify relevant parameters. The dataset was split into derivation (80%) and validation (20%) cohorts. A decision-support model was developed for obtaining the regression equation between optimized intrathecal 0.5% hyperbaric bupivacaine volume and physical variables.

RESULTS : A total of 684 parturients were included, of whom 516 (75.44%) and 168 (24.56%) had block levels between T4 and T6, and less than T6 or higher than T4, respectively. The appropriate block level rate was 75.44%, with the mean bupivacaine volume [1.965, 95%CI (1.945,1.984)]ml. In lasso regression, based on the principle of predicting a reasonable dose of intrathecal bupivacaine with fewer physical variables, the model is "Y=0.5922+ 0.055117* X₁-0.017599*X₂" (Y: bupivacaine volume; X₁: vertebral column length; X₂: abdominal girth), with λ 0.055, MSE 0.0087, and R² 0.807.

CONCLUSIONS : After applying a machine-learning algorithm, we developed a decision model with R² 0.8070 and MSE due to error 0.0087 using abdominal girth and vertebral column length for predicting the optimized intrathecal 0.5% hyperbaric bupivacaine dosage during term cesarean sections.

Wei Chang-Na, Wang Li-Ying, Chang Xiang-Yang, Zhou Qing-He

2021-Apr-14

Bupivacaine dosage, Cesarean section, Machine learning algorithm, Physical variables, Spinal anesthesia

ArXiv Preprint

Danilo Numeroso, Davide Bacciu

2021-04-16

In Critical reviews in toxicology

Cox Louis A, Ketelslegers Hans B, Lewis R Jeffrey

2021-Apr-15

Bayesian networks, Benzene metabolism, Chinese workers, PBPK model, machine learning, supralinearity

ArXiv Preprint

Helena Russello, Rik van der Tol, Gert Kootstra

2021-04-16

In Interventional neuroradiology : journal of peritherapeutic neuroradiology, surgical procedures and related neurosciences

INTRODUCTION : There is increased interest in the use of artificial intelligence-based (AI) software packages in the evaluation of neuroimaging studies for acute ischemic stroke. We studied whether, compared to standard image interpretation without AI, Brainomix e-ASPECTS software improved interobserver agreement and accuracy in detecting ASPECTS regions affected in anterior circulation LVO.

METHODS : We included 60 consecutive patients with anterior circulation LVO who had TICI 3 revascularization within 60 minutes of their baseline CT. A total of 16 readers, including senior neuroradiologists, junior neuroradiologists and vascular neurologists participated. Readers interpreted CT scans on independent workstations and assessed final ASPECTS and evaluated whether each individual ASPECTS region was affected. Two months later, readers again evaluated the CT scans, but with assistance of e-ASPECTS software. We assessed interclass correlation coefficient for total ASPECTS and interobserver agreement with Fleiss' Kappa for each ASPECTS region with and without assistance of the e-ASPECTS. We also assessed accuracy for the readers with and without e-ASPECTS assistance. In our assessment of accuracy, ground truth was the 24 hour CT in this cohort of patients who had prompt and complete revascularization.

RESULTS : Interclass correlation coefficient for total ASPECTS without e-ASPECTS assistance was 0.395, indicating fair agreement compared, to 0.574 with e-ASPECTS assistance, indicating good agreement (P < 0.01). There was significant improvement in inter-rater agreement with e-ASPECTS assistance for each individual region with the exception of M6 and caudate. The e-ASPECTS software had higher accuracy than the overall cohort of readers (with and without e-ASPECTS assistance) for every region except the caudate.

CONCLUSIONS : Use of Brainomix e-ASPECTS software resulted in significant improvements in inter-rater agreement and accuracy of ASPECTS score evaluation in a large group of neuroradiologists and neurologists. e-ASPECTS software was more predictive of final infarct/ASPECTS than the overall group interpreting the CT scans with and without e-ASPECTS assistance.

Brinjikji Waleed, Abbasi Mehdi, Arnold Catherine, Benson John C, Braksick Sherry A, Campeau Norbert, Carr Carrie M, Cogswell Petrice M, Klaas James P, Liebo Greta B, Little Jason T, Luetmer Patrick H, Messina Steven A, Nagelschneider Alex A, Schwartz Kara M, Wood Christopher P, Nasr Deena M, Kallmes David F

2021-Apr-14

ASPECTS, CT, Stroke, artificial intelligence, large vessel occlusion

ArXiv Preprint

Shiva Azimi, Rohan Wadhawan, Tapan K. Gandhi

2021-04-16

ArXiv Preprint

Shiva Azimi, Rohan Wadhawan, Tapan K. Gandhi

2021-04-16

In Advanced science (Weinheim, Baden-Wurttemberg, Germany)

Kim Kyungsoo, Sim Minkyung, Lim Sung-Ho, Kim Dongsu, Lee Doyoung, Shin Kwonsik, Moon Cheil, Choi Ji-Woong, Jang Jae Eun

2021-Apr

P(VDF‐TrFE), machine learning, piezoelectric effect, tactile avatars

In Oncoimmunology

Kim Yeongjoo, Kang Ji Wan, Kang Junho, Kwon Eun Jung, Ha Mihyang, Kim Yoon Kyeong, Lee Hansong, Rhee Je-Keun, Kim Yun Hak

2021-Mar-29

Head and neck cancer, cibersort, deep learning, international cancer genome consortium, oral cancer, the cancer genome atlas, tumor microenvironment, tumor-infiltrating lymphocytes

In Journal of healthcare engineering

Du Hongwei, Feng Linxing, Xu Yan, Zhan Enbo, Xu Wei

2021

In Journal of Cancer

Gong Ting-Ting, He Xin-Hui, Gao Song, Wu Qi-Jun

2021

chemoresistant, gut microbiota, machine learning, ovarian cancer, random forest

In Computational intelligence and neuroscience

Noh Sun-Kuk

2021

In Clinical Medicine Insights. Oncology

Objective : Over 1 million new cases of hepatocellular carcinoma (HCC) are diagnosed worldwide every year. Its prognosis remains poor, and the 5-year survival rate in all disease stages is estimated to be between 10% and 20%. Radiofrequency ablation (RFA) has become an important local treatment for liver cancer, and machine learning (ML) can provide many shortcuts for liver cancer medical research. Therefore, we explore the role of ML in predicting the total mortality of liver cancer patients undergoing RFA.

Methods : This study is a secondary analysis of public database data from 578 liver cancer patients. We used Python for ML to establish the prognosis model.

Results : The results showed that the 5 most important factors were platelet count (PLT), Alpha-fetoprotein (AFP), age, tumor size, and total bilirubin, respectively. Results of the total death model for liver cancer patients in test group: among the 5 algorithm models, the highest accuracy rate was that of gbm (0.681), followed by the Logistic algorithm (0.672); among the 5 algorithms, area under the curve (AUC) values, from high to low, were Logistic (0.738), DecisionTree (0.723), gbm (0.717), GradientBoosting (0.714), and Forest (0.693); Among the 5 algorithms, gbm had the highest precision rate (0.721), followed by the Logistic algorithm (0.714). Among the 5 algorithms, DecisionTree had the highest recall rate (0.642), followed by the GradientBoosting algorithm (0.571).

Conclusion : Machine learning can predict total death after RFA in liver cancer patients. Therefore, ML research has great potential for both personalized treatment and prognosis of liver cancer.

Tong Jianhua, Liu Panmiao, Ji Muhuo, Wang Ying, Xue Qiong, Yang Jian-Jun, Zhou Cheng-Mao

2021

Treatment algorithms, hepatocellular carcinoma, individual outcomes, patient stratification, radiofrequency ablation

In Clinical ophthalmology (Auckland, N.Z.)

Background : In recent years, increasing interest has arisen in the application of data from corneal biomechanics in many areas of ophthalmology, particularly to assist in the detection of early corneal ectasia or ectasia susceptibility, to predict corneal response to surgical or therapeutic interventions and in glaucoma management. Technology has evolved and, recently, the Scheimpflug principle was associated with a non-contact air-puff tonometer, allowing a thorough analysis of corneal biomechanics and a biomechanically corrected intraocular pressure assessment, opening up new perspectives both in ophthalmology and in other medical areas. Data from corneal biomechanics assessment are being integrated in artificial intelligence models in order to increase its value in clinical practice.

Objective : To review the state of the art in the field of corneal biomechanics assessment with special emphasis to the technology based on ultra-high-speed Scheimpflug imaging during non-contact tonometry.

Summary : A meticulous literature review was performed until the present day. We used 136 published manuscripts as our references. Both information from healthy individuals and descriptions of possible associations with systemic diseases are described. Additionally, it exposed information regarding several fields of ocular pathology, from cornea and ocular surface through areas of refractive surgery and glaucoma until vascular and structural diseases of the chorioretinal unit.

Baptista Pedro Manuel, Ambrosio Renato, Oliveira Luis, Meneres Pedro, Beirao Joao Melo

2021

Corvis, cornea, corneal biomechanics, ultra-high speed Szcheimpflug camera

In Physiology and molecular biology of plants : an international journal of functional plant biology

Unnikrishnan Remya, Sumod M, Jayaraj R, Sujanapal P, Dev Suma Arun

2021-Mar

Artificial intelligence, DNA barcoding, Machine learning algorithm, Medicinal plants, Threatened species

In Scientific reports ; h5-index 158.0

Chemura Abel, Mudereri Bester Tawona, Yalew Amsalu Woldie, Gornott Christoph

2021-Apr-14

In Scientific reports ; h5-index 158.0

Wagner Dennis, Heider Dominik, Hattab Georges

2021-Apr-14

In Nature communications ; h5-index 260.0

Park Bo-Yong, Hong Seok-Jun, Valk Sofie L, Paquola Casey, Benkarim Oualid, Bethlehem Richard A I, Di Martino Adriana, Milham Michael P, Gozzi Alessandro, Yeo B T Thomas, Smallwood Jonathan, Bernhardt Boris C

2021-04-13

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : In this age of global COVID-19 pandemic, the urgency of addressing an epidemic of obesity and associated inflammatory illnesses has come to the fore. Studies have demonstrated that interactions between single nucleotide polymorphisms (SNPs) and lifestyle interventions like food and exercise may vary metabolic outcomes, contributing to obesity and therapeutic response. However, there is a paucity of research relating outcomes from digital therapeutics to inclusion of genetic data in care interventions.

OBJECTIVE : This study aims to describe and model weight loss of subjects enrolled in a precision digital weight loss program informed by machine learning analysis of subject data, including genomic. It was hypothesized that weight loss models would exhibit better fit when incorporating genomic data than utilizing demographic and engagement variables alone.

METHODS : A cohort of 393 participants enrolled in Digbi's personalized digital care program for 120 days was analyzed retrospectively. Care protocol included the use of subject data informing precision coaching by mobile app and personal coach. Two linear regression models of weight loss in this cohort (pounds lost, percentage lost) as a function of demographic and behavioral engagement variables were fit. Genomic-enhanced models were built by adding 197 SNPs from subject genomic data as predictors, then refitting, employing Lasso regression on SNPs for variable selection. Success/failure logistic regression models were also fit, with and without genomic data.

RESULTS : 72% of subjects in this cohort lost weight, while 17% maintained stable weight. 142 subjects lost 5% within 120 days. Models describe the impact of demographic and clinical factors, behavioral engagement, and genomic risk on weight loss. The addition of genomic predictors improved the mean squared error of weight loss models (pounds lost and percent) from 70 to 60 and 16 to 13 respectively. The logistic model improved pseudo R2 from 0.193 to 0.285. Gender, engagement and specific SNPs were significantly associated with weight loss. SNPs within genes involved in metabolic pathways that process food and regulate storage of fat were associated with weight loss in this cohort. This included rs17300539_G (insulin resistance, monounsaturated fat metabolism), rs2016520_C (BMI, waist circumference, cholesterol metabolism), and rs4074995_A (calcium-potassium transport, serum calcium levels). Models described greater average weight loss for subjects having more of these risk alleles. Notably, coaching for dietary modification was personalized to these genetic risks.

CONCLUSIONS : Adding genomic information in modeling outcomes of a digital precision weight loss program greatly enhanced model accuracy. Interpretable weight loss models pointed to efficacy of coaching informed by subjects' genomic risk, accompanied by active engagement of subjects in their own success. While large-scale validation is needed, our study preliminarily supports precision dietary interventions for weight loss utilizing genetic risk, with digitally delivered recommendations alongside health-coaching to improve intervention efficacy.

CLINICALTRIAL :

Sinha Ranjan, Kachru Dashyanng, Ricchetti Roshni Ray, Singh-Rambiritch Simitha, Muthukumar Karthik Marimuthu, Singaravel Vidhya, Irudayanathan Carmel, Reddy-Sinha Chandana, Junaid Imran, Sharma Garima, Airey Catherine, Francis-Lyon Patricia Alice

2021-Apr-11

In Clinical genitourinary cancer

INTRODUCTION : Immunotherapy is effective in a small percentage of patients with cancer and no reliable predictive biomarkers are currently available. Artificial Intelligence algorithms may automatically quantify radiologic characteristics associated with disease response to medical treatments.

METHODS : We investigated an innovative approach based on a 3-dimensional (3D) deep radiomics pipeline to classify visual features of chest-abdomen computed tomography (CT) scans with the aim of distinguishing disease control from progressive disease to immune checkpoint inhibitors (ICIs). Forty-two consecutive patients with metastatic urothelial cancer had progressed on first-line platinum-based chemotherapy and had baseline CT scans at immunotherapy initiation. The 3D-pipeline included self-learned visual features and a deep self-attention mechanism. According to the outcome to the ICIs, a 3D deep classifier semiautomatically categorized the most discriminative region of interest on the CT scans.

RESULTS : With a median follow-up of 13.3 months (95% CI, 11.1-15.6), the median overall survival was 8.5 months (95% CI, 3.1-13.8). According to disease response to immunotherapy, the median overall survival was 3.6 months (95% CI, 2.0-5.2) for patients with progressive disease; it was not yet reached for those with disease control. The predictive accuracy of the 3D-pipeline was 82.5% (sensitivity 96%; specificity, 60%). The addition of baseline clinical factors increased the accuracy to 92.5% by improving specificity to 87%; the accuracy of other architectures ranged from 72.5% to 90%.

CONCLUSION : Artificial Intelligence by 3D deep radiomics is a potential noninvasive biomarker for the prediction of disease control to ICIs in metastatic urothelial cancer and deserves validation in larger series.

Rundo Francesco, Bersanelli Melissa, Urzia Valeria, Friedlaender Alex, Cantale Ornella, Calcara Giacomo, Addeo Alfredo, Banna Giuseppe Luigi

2021-Mar-19

3D-CNN, Artificial intelligence, Deep learning, Immune-checkpoint inhibitors, Machine learning

In Journal of voice : official journal of the Voice Foundation

OBJECTIVE : Estimation of physiological control parameters of the vocal system from the produced voice outcome has important applications in clinical management of voice disorders . Previously we developed a simulation-based neural network for estimation of vocal fold geometry, mechanical properties, and subglottal pressure from voice outcome features that characterize the acoustics of the produced voice. The goals of this study are to (1) explore the possibility of improving the estimation accuracy of physiological control parameters by including voice outcome features characterizing vocal fold vibration; and (2) identify voice feature sets that optimize both estimation accuracy and robustness to measurement noise.

METHODS : Feedforward neural networks are trained to solve the inversion problem of estimating the physiological control parameters of a three-dimensional body-cover vocal fold model from different sets of voice outcome features that characterize the simulated voice acoustics, glottal flow, and vocal fold vibration. A sensitivity analysis is then performed to evaluate the contribution of individual voice features to the overall performance of the neural networks in estimating the physiologic control parameters.

RESULTS AND CONCLUSIONS : While including voice outcome features characterizing vocal fold vibration increases estimation accuracy, it also reduces the network's robustness to measurement noise, due to high sensitivity of network performance to voice outcome features measuring the absolute amplitudes of the glottal flow and area waveforms, which are also difficult to measure accurately in practical applications. By excluding such glottal flow-based features and replacing glottal area-based features by their normalized counterparts, we are able to significantly improve both estimation accuracy and robustness to noise. We further show that similar estimation accuracy and robustness can be achieved with an even smaller set of voice outcome features by excluding features of small sensitivity.

Zhang Zhaoyan

2021-Apr-10

Voice inversion—Vocal fold geometry—Vocal fold stiffness—Machine learning

In BMC medical imaging

BACKGROUND : In oncology, the correct determination of nodal metastatic disease is essential for patient management, as patient treatment and prognosis are closely linked to the stage of the disease. The aim of the study was to develop a tool for automatic 3D detection and segmentation of lymph nodes (LNs) in computed tomography (CT) scans of the thorax using a fully convolutional neural network based on 3D foveal patches.

METHODS : The training dataset was collected from the Computed Tomography Lymph Nodes Collection of the Cancer Imaging Archive, containing 89 contrast-enhanced CT scans of the thorax. A total number of 4275 LNs was segmented semi-automatically by a radiologist, assessing the entire 3D volume of the LNs. Using this data, a fully convolutional neuronal network based on 3D foveal patches was trained with fourfold cross-validation. Testing was performed on an unseen dataset containing 15 contrast-enhanced CT scans of patients who were referred upon suspicion or for staging of bronchial carcinoma.

RESULTS : The algorithm achieved a good overall performance with a total detection rate of 76.9% for enlarged LNs during fourfold cross-validation in the training dataset with 10.3 false-positives per volume and of 69.9% in the unseen testing dataset. In the training dataset a better detection rate was observed for enlarged LNs compared to smaller LNs, the detection rate for LNs with a short-axis diameter (SAD) ≥ 20 mm and SAD 5-10 mm being 91.6% and 62.2% (p < 0.001), respectively. Best detection rates were obtained for LNs located in Level 4R (83.6%) and Level 7 (80.4%).

CONCLUSIONS : The proposed 3D deep learning approach achieves an overall good performance in the automatic detection and segmentation of thoracic LNs and shows reasonable generalizability, yielding the potential to facilitate detection during routine clinical work and to enable radiomics research without observer-bias.

Iuga Andra-Iza, Carolus Heike, Höink Anna J, Brosch Tom, Klinder Tobias, Maintz David, Persigehl Thorsten, Baeßler Bettina, Püsken Michael

2021-Apr-13

Artificial intelligence, Computed tomography, Deep learning, Lymph nodes, Staging

ArXiv Preprint

Nasrin Taghizadeh, Ehsan Doostmohammadi, Elham Seifossadat, Hamid R. Rabiee, Maedeh S. Tahaei

2021-04-15

In BMC cancer

BACKGROUND : Bronchoscopy is a common procedure used for evaluation of suspicious lung nodules, but the low diagnostic sensitivity of bronchoscopy often results in inconclusive results and delays in treatment. Percepta Genomic Sequencing Classifier (GSC) was developed to assist with patient management in cases where bronchoscopy is inconclusive. Studies have shown that exposure to tobacco smoke alters gene expression in airway epithelial cells in a way that indicates an increased risk of developing lung cancer. Percepta GSC leverages this idea of a molecular "field of injury" from smoking and was developed using RNA sequencing data generated from lung bronchial brushings of the upper airway. A Percepta GSC score is calculated from an ensemble of machine learning algorithms utilizing clinical and genomic features and is used to refine a patient's risk stratification.

METHODS : The objective of the analysis described and reported here is to validate the analytical performance of Percepta GSC. Analytical performance studies characterized the sensitivity of Percepta GSC test results to input RNA quantity, the potentially interfering agents of blood and genomic DNA, and the reproducibility of test results within and between processing runs and between laboratories.

RESULTS : Varying the amount of input RNA into the assay across a nominal range had no significant impact on Percepta GSC classifier results. Bronchial brushing RNA contaminated with up to 10% genomic DNA by nucleic acid mass also showed no significant difference on classifier results. The addition of blood RNA, a potential contaminant in the bronchial brushing sample, caused no change to classifier results at up to 11% contamination by RNA proportion. Percepta GSC scores were reproducible between runs, within runs, and between laboratories, varying within less than 4% of the total score range (standard deviation of 0.169 for scores on 4.57 scale).

CONCLUSIONS : The analytical sensitivity, analytical specificity, and reproducibility of Percepta GSC laboratory results were successfully demonstrated under conditions of expected day to day variation in testing. Percepta GSC test results are analytically robust and suitable for routine clinical use.

Johnson Marla K, Wu Shuyang, Pankratz Daniel G, Fedorowicz Grazyna, Anderson Jessica, Ding Jie, Wong Mei, Cao Manqiu, Babiarz Joshua, Lofaro Lori, Walsh P Sean, Kennedy Giulia C, Huang Jing

2021-Apr-13

Analytical validation, Bronchial brushing specimen, Genomic sequencing classifier, Lung lesion, Molecular diagnostic, Percepta

In PloS one ; h5-index 176.0

Eastman Brydon, Meaney Cameron, Przedborski Michelle, Kohandel Mohammad

2021

ArXiv Preprint

Sola Shirai, Oshani Seneviratne, Deborah L. McGuinness

2021-04-15

In Physiological measurement ; h5-index 36.0

OBJECTIVE : Fetal heart rate monitoring is routinely used during pregnancy and labor to assess fetal well-being. The non-invasive fetal electrocardiogram (ECG), obtained by electrodes on the maternal abdomen, is a promising alternative to standard fetal monitoring. Subtraction of the maternal ECG from the abdominal measurements results in fetal ECG signals, in which the fetal heart rate can be determined typically through R-peak detection. However, the low signal-to-noise ratio and the nonstationary nature of the fetal ECG make R-peak detection a challenging task.

APPROACH : We propose an alternative approach that instead of performing R-peak detection employs deep learning to directly determine the fetal heart rate from the extracted fetal ECG signals. We introduce a combination of dilated inception convolutional neural networks (CNN) with long short-term memory networks to capture both short-term and long-term temporal dynamics of the fetal heart rate. The robustness of the method is reinforced by a separate CNN-based classifier that estimates the reliability of the outcome.

MAIN RESULTS : Our method achieved a positive percent agreement (within 10% of the actual fetal heart rate value) of 97.3% on a dataset recorded during labor and 99.6% on set-A of the 2013 Physionet/Computing in Cardiology Challenge exceeding top-performing state-of-the-art algorithms from the literature.

SIGNIFICANCE : The proposed method can potentially improve the accuracy and robustness of fetal heart rate extraction in clinical practice.

Fotiadou Eleni, van Sloun Ruud J G, van Laar Judith O E H, Vullings Rik

2021-Apr-14

convolutional neural networks, dilated convolution, fetal electrocardiogram, fetal heart rate, long short-term memory networks, noninvasive fetal ECG

In Journal of chromatography. A

Kensert Alexander, Collaerts Gilles, Efthymiadis Kyriakos, Van Broeck Peter, Desmet Gert, Cabooter Deirdre

2021-Mar-23

Autoencoder, Baseline drift, Baseline noise, Deep learning, Denoising, Machine learning, Noise reduction

In Gastrointestinal endoscopy ; h5-index 72.0

BACKGROUND AND AIMS : Endoscopic submucosal dissection (ESD) and endoscopic mucosal resection (EMR) are applied in treating superficial colorectal neoplasms but contraindicated by deeply invasive colorectal cancer (CRC). The invasion depth of neoplasms can be examined by an automated artificial intelligence (AI) system to determine the applicability of ESD and EMR.

METHODS : A deep convolutional neural network (CNN) with tumor localization branch to guide invasion depth classification was constructed on the GoogLeNet architecture. The model was trained using 7,734 nonmagnified white-light colonoscopy (WLC) images supplemented by image augmentation, from 657 lesions labeled with histopathological analysis of invasion depth. An independent testing dataset consisting of 1,631 WLC images from 156 lesions was used to validate the model.

RESULTS : For predicting noninvasive and superficially invasive neoplasms, the model achieved an overall accuracy of 91.1% (95% confidence interval [CI], 89.6% - 92.4%), with 91.2% sensitivity (95% CI, 88.8% - 93.3%) and 91.0% specificity (95% CI, 89.0% - 92.7%) at an optimal cutoff of 0.41 and the area under the receiver operating characteristic (AUROC) of 0.970 (95% CI, 0.962 - 0.978). Inclusion of the advanced CRC data significantly increased the sensitivity in differentiating superficial neoplasms from deeply invasive early CRC to 65.3% (95% CI, 61.9% - 68.8%) with AUROC of 0.729 (95% CI, 0.699 - 0.759) similar to experienced endoscopists (0.691; 95% CI, 0.624 - 0.758).

CONCLUSION : We have developed an AI-enhanced attention-guided WLC system that differentiates noninvasive or superficially submucosal invasive neoplasms from deeply invasive CRC with high accuracy, sensitivity, and specificity.

Xiaobei Luo, Jiahao Wang, Zelong Han, Yang Yu, Zhenyu Chen, Feiyang Huang, Yumeng Xu, Jianqun Cai, Qiang Zhang, Weiguang Qiao, Chuan Ng Inn, Robby Tan, Side Liu, Hanry Yu

2021-Apr-11

artificial intelligence, colorectal cancer, invasion depth, white light colonoscopy

In Cell reports ; h5-index 119.0

Camiolo Matthew J, Zhou Xiaoying, Oriss Timothy B, Yan Qi, Gorry Michael, Horne William, Trudeau John B, Scholl Kathryn, Chen Wei, Kolls Jay K, Ray Prabir, Weisel Florian J, Weisel Nadine M, Aghaeepour Nima, Nadeau Kari, Wenzel Sally E, Ray Anuradha

2021-Apr-13

BAL, CyTOF, FceRI+, ICLite, IFN-g, RNA-seq, clusters, immune, multi-omics, severe asthma

In Neuropharmacology ; h5-index 72.0

Chen Xinling, Xu Yongbo, Li Bingjie, Wu Xiaoyan, Li Ting, Wang Li, Zhang Yijie, Lin Wanghuan, Qu Chen, Feng Chunliang

2021-Apr-11

amplitude of low-frequency fluctuation, arginine vasopressin, functional decoding, large-scale network, machine learning, resting-state fMRI

In PloS one ; h5-index 176.0

Eastman Brydon, Meaney Cameron, Przedborski Michelle, Kohandel Mohammad

2021

In PloS one ; h5-index 176.0

Hu Jingfei, Wang Hua, Wang Jie, Wang Yunqi, He Fang, Zhang Jicong

2021

In IEEE transactions on neural networks and learning systems

Li Xiang, Zhang Wei, Ma Hui, Luo Zhong, Li Xu

2021-Apr-14

In IEEE transactions on neural networks and learning systems

Pan Pingping, Zhang Yunjian, Deng Zhenmiao, Qi Wei

2021-Apr-14

In Journal of the Royal Society, Interface

Alexiadis A, Simmons M J H, Stamatopoulos K, Batchelor H K, Moulitsas I

2021-Apr

coupling multiphysics with artificial intelligence, mathematical modelling of the intestine, multiphysics, reinforcement learning, virtual human

In medRxiv : the preprint server for health sciences

Rationale : Viral infection of the respiratory tract can be associated with propagating effects on the airway microbiome, and microbiome dysbiosis may influence viral disease.

Objective : To define the respiratory tract microbiome in COVID-19 and relationship disease severity, systemic immunologic features, and outcomes.

Methods and Measurements : We examined 507 oropharyngeal, nasopharyngeal and endotracheal samples from 83 hospitalized COVID-19 patients, along with non-COVID patients and healthy controls. Bacterial communities were interrogated using 16S rRNA gene sequencing, commensal DNA viruses Anelloviridae and Redondoviridae were quantified by qPCR, and immune features were characterized by lymphocyte/neutrophil (L/N) ratios and deep immune profiling of peripheral blood mononuclear cells (PBMC).

Main Results : COVID-19 patients had upper respiratory microbiome dysbiosis, and greater change over time than critically ill patients without COVID-19. Diversity at the first time point correlated inversely with disease severity during hospitalization, and microbiome composition was associated with L/N ratios and PBMC profiles in blood. Intubated patients showed patient-specific and dynamic lung microbiome communities, with prominence of Staphylococcus . Anelloviridae and Redondoviridae showed more frequent colonization and higher titers in severe disease. Machine learning analysis demonstrated that integrated features of the microbiome at early sampling points had high power to discriminate ultimate level of COVID-19 severity.

Conclusions : The respiratory tract microbiome and commensal virome are disturbed in COVID-19, correlate with systemic immune parameters, and early microbiome features discriminate disease severity. Future studies should address clinical consequences of airway dysbiosis in COVID-19, possible use as biomarkers, and role of bacterial and viral taxa identified here in COVID-19 pathogenesis.

Merenstein Carter, Liang Guanxiang, Whiteside Samantha A, Cobián-Güemes Ana G, Merlino Madeline S, Taylor Louis J, Glascock Abigail, Bittinger Kyle, Tanes Ceylan, Graham-Wooten Jevon, Khatib Layla A, Fitzgerald Ayannah S, Reddy Shantan, Baxter Amy E, Giles Josephine R, Oldridge Derek A, Meyer Nuala J, Wherry E John, McGinniss John E, Bushman Frederic D, Collman Ronald G

2021-Apr-05

ArXiv Preprint

Samuel Greenbank, David A. Howey

2021-04-15

In IEEE transactions on neural networks and learning systems

Quan Yuhui, Lin Peikang, Xu Yong, Nan Yuesong, Ji Hui

2021-Apr-14

In IEEE transactions on neural networks and learning systems

Wu Hao, Qin Shaowei, Nie Rencan, Cao Jinde, Gorbachev Sergey

2021-Apr-14

In IEEE transactions on bio-medical engineering

OBJECTIVE : Many recent studies have suggested that brain deformation resulting from a head impact is linked to the corresponding clinical outcome, such as mild traumatic brain injury (mTBI). Even though several finite element (FE) head models have been developed and validated to calculate brain deformation based on impact kinematics, the clinical application of these FE head models is limited due to the time-consuming nature of FE simulations. This work aims to accelerate the process of brain deformation calculation and thus improve the potential for clinical applications.

METHODS : We propose a deep learning head model with a five-layer deep neural network and feature engineering, and trained and tested the model on 2511 total head impacts from a combination of head model simulations and on-field college football and mixed martial arts impacts.

RESULTS : The proposed deep learning head model can calculate the maximum principal strain (Green Lagrange) for every element in the entire brain in less than 0.001s with an average root mean squared error of 0.022, and with a standard deviation of 0.001 over twenty repeats with random data partition and model initialization.

CONCLUSION : Trained and tested using the dataset of 2511 head impacts, this model can be applied to various sports in the calculation of brain strain with accuracy, and its applicability can even further be extended by incorporating data from other types of head impacts.

SIGNIFICANCE : In addition to the potential clinical application in real-time brain deformation monitoring, this model will help researchers estimate the brain strain from a large number of head impacts more efficiently than using FE models.

Zhan Xianghao, Liu Yuzhe, Raymond Samuel J, Vahid Alizadeh Hossein, Domel August, Gevaert Olivier, Zeineh Michael, Grant Gerald, Camarillo David Benjamin

2021-Apr-14

In Annual review of chemical and biomolecular engineering

Pertseva Margarita, Gao Beichen, Neumeier Daniel, Yermanos Alexander, Reddy Sai T

2021-Apr-14

In JCO clinical cancer informatics

PURPOSE : Recent advances in structured reporting are providing an opportunity to enhance cancer imaging assessment to drive value-based care and improve patient safety.

METHODS : The computer-assisted reporting and decision support (CAR/DS) framework has been developed to enable systematic ingestion of guidelines as clinical decision structured reporting tools embedded within the radiologist's workflow.

RESULTS : CAR/DS tools can reduce the radiology reporting variability and increase compliance with clinical guidelines. The lung cancer use-case is used to describe various scenarios of a cancer imaging structured reporting pathway, including incidental findings, screening, staging, and restaging or continued care. Various aspects of these tools are also described using cancer-related examples for different imaging modalities and applications such as calculators. Such systems can leverage artificial intelligence (AI) algorithms to assist with the generation of structured reports and there are opportunities for new AI applications to be created using the structured data associated with CAR/DS tools.

CONCLUSION : These AI-enabled systems are starting to allow information from multiple sources to be integrated and inserted into structured reports to drive improvements in clinical decision support and patient care.

Bizzo Bernardo C, Almeida Renata R, Alkasab Tarik K

2021-Apr

In Physical and engineering sciences in medicine

Yücelbaş Cüneyt

2021-Apr-14

Artificial intelligence systems, Information gain approach, KNN, Parkinson’s disease, Speech signals

In European radiology ; h5-index 62.0

OBJECTIVES : To evaluate a deep learning model for predicting gestational age from fetal brain MRI acquired after the first trimester in comparison to biparietal diameter (BPD).

MATERIALS AND METHODS : Our Institutional Review Board approved this retrospective study, and a total of 184 T2-weighted MRI acquisitions from 184 fetuses (mean gestational age: 29.4 weeks) who underwent MRI between January 2014 and June 2019 were included. The reference standard gestational age was based on the last menstruation and ultrasonography measurements in the first trimester. The deep learning model was trained with T2-weighted images from 126 training cases and 29 validation cases. The remaining 29 cases were used as test data, with fetal age estimated by both the model and BPD measurement. The relationship between the estimated gestational age and the reference standard was evaluated with Lin's concordance correlation coefficient (ρc) and a Bland-Altman plot. The ρc was assessed with McBride's definition.

RESULTS : The ρc of the model prediction was substantial (ρc = 0.964), but the ρc of the BPD prediction was moderate (ρc = 0.920). Both the model and BPD predictions had greater differences from the reference standard at increasing gestational age. However, the upper limit of the model's prediction (2.45 weeks) was significantly shorter than that of BPD (5.62 weeks).

CONCLUSIONS : Deep learning can accurately predict gestational age from fetal brain MR acquired after the first trimester.

KEY POINTS : • The prediction of gestational age using ultrasound is accurate in the first trimester but becomes inaccurate as gestational age increases. • Deep learning can accurately predict gestational age from fetal brain MRI acquired in the second and third trimester. • Prediction of gestational age by deep learning may have benefits for prenatal care in pregnancies that are underserved during the first trimester.

Kojita Yasuyuki, Matsuo Hidetoshi, Kanda Tomonori, Nishio Mizuho, Sofue Keitaro, Nogami Munenobu, Kono Atsushi K, Hori Masatoshi, Murakami Takamichi

2021-Apr-14

Brain, Deep learning, Fetus, Gestational age, Pregnancy

In European radiology ; h5-index 62.0

OBJECTIVES : Quantum noise is a random process in X-ray-based imaging systems. We addressed and measured the uncertainty of radiomics features against this quantum noise in computed tomography (CT) images.

METHODS : A clinical multi-detector CT scanner, two homogeneous phantom sets, and four heterogeneous samples were used. A solid tumor tissue removed from a male BALB/c mouse was included. We the placed phantom sets on the CT scanning table and repeated 20 acquisitions with identical imaging settings. Regions of interest were delineated for feature extraction. Statistical quantities-average, standard deviation, and percentage uncertainty-were calculated from these 20 repeated scans. Percentage uncertainty was used to measure and quantify feature stability against quantum noise. Twelve radiomics features were measured. Random noise was added to study the robustness of machine learning classifiers against feature uncertainty.

RESULTS : We found the ranges of percentage uncertainties from homogeneous soft tissue phantoms, homogeneous bone phantoms, and solid tumor tissue to be 0.01-2138%, 0.02-15%, and 0.18-16%, respectively. Overall, it was found that the CT features ShortRunHighGrayLevelEmpha (SRHGE) (0.01-0.18%), ShortRunLowGrayLevelEmpha (SRLGE) (0.01-0.41%), LowGrayLevelRunEmpha (LGRE) (0.01-0.39%), and LongRunLowGrayLevelEmpha (LRLGE) (0.02-0.66%) were the most stable features against the inherent quantum noise. The most unstable features were cluster shade (1-2138%) and max probability (1-16%). The impact of random noise to the prediction accuracy by different machine learning classifiers was found to be between 0 and 12%.

CONCLUSIONS : Twelve features were used for uncertainty measurements. The upper and lower bounds of percentage uncertainties were determined. The quantum noise effect on machine learning classifiers is model dependent.

KEY POINTS : • Quantum noise is a random process and is intrinsic to X-ray-based imaging systems. This inherent quantum noise creates unpredictable fluctuations in the gray-level intensities of image pixels. Extra cautions and further validations are strongly recommended when unstable radiomics features are selected by a predictive model for disease classification or treatment outcome prognosis. • We addressed and used the statistical quantity of percentage uncertainty to measure the uncertainty of radiomics features against the inherent quantum noise in computed tomography (CT) images. • A clinical multi-detector CT scanner, two homogeneous phantom sets, and four heterogeneous samples were used in the stability measurement. A solid tumor tissue removed from a male BALB/c mouse was included in the heterogeneous sample.

Tu Shu-Ju, Chen Wei-Yuan, Wu Chen-Te

2021-Apr-14

Health care quality assurance, Medical informatics computing, Radiomics, Uncertainty, X-ray computed tomography

In Informatics for health & social care

Khan Jahidur Rahman, Tomal Jabed H, Raheem Enayetur

2021-Apr-14

Bangladesh, Machine learning, prediction, stunting, variable importance

In Biological chemistry

Kamp Christel, Becker Björn, Matheis Walter, Öppling Volker, Bekeredjian-Ding Isabelle

2021-Apr-13

Raman spectroscopy, machine learning, pre-processing, quality control, standardisation, vaccine

In Traffic injury prevention

OBJECTIVE : The aim of this study was to explore a suitable method for analyzing road transport accidents that involve hazardous materials and to explore the main factors that influence the occurrence of accidents of varying severity.

METHODS : The 2015-2019 reported crash data from the Ministry of Transport of the People's Republic of China were obtained, and road transport crashes involving hazardous materials were extracted as the analysis data. The dataset was classified into three injury severity categories: property damage only (PDO), injured (INJ), and fatal (FAT). A statistical model and three machine learning-based models were developed: a random parameters logit model (RPLM), multilayer perceptron (MLP), decision tree C5.0 (C5.0) and support vector machine (SVM). The four models were trained/estimated using the training/estimation dataset, and the best model was used to model accidents of the three different severity levels. The main factors that influence the occurrence of accidents at each crash severity level were obtained.

RESULTS : C5.0 had the best modeling performance. The direct accident form (DAF), indirect accident form (IAF) and road segment (RS) were determined to be the critical determinants of PDO accidents. The DAF, IAF, road type, RS and time had a substantial effect on INJ accidents. The DAF, IAF, hazardous material type (HMT) and road surface condition were important factors in the occurrence of FAT accidents.

CONCLUSIONS : Different data have unique characteristics, and the best modeling and analysis method should be chosen accordingly. The safety of road transport of hazardous materials in China is poor, and the losses caused by accidents are substantial. Strengthening the monitoring of travel speed and travel time; improving driver safety awareness, driving skills and the ability to mitigate emergencies; improving the configuration of vehicle safety equipment and the linkage with the control center and rescue center; improving the environmental differences between inside a tunnel and outside a tunnel; reducing the design of long downhill and steep slope sections; reducing the transport plan in unsafe environments; and improving the ability of road management to mitigate bad environments can be effective measures to reduce the severity of road transport accidents involving hazardous materials.

Shen Xiaoyan, Wei Shanshan

2021-Apr-13

Hazardous materials, accident severities, influencing factors, machine learning, road transportation

In Global health journal (Amsterdam, Netherlands)

Zhang Yanjie, Tambo Ernest, Djuikoue Ingrid C, Tazemda Gildas K, Fotsing Michael F, Zhou Xiao-Nong

2021-Mar

Coronavirus disease 2019 (COVID-19), Governance, Pandemic, Response, Risk communication and community engagement (RCCE), Trust, Vaccination

ArXiv Preprint

Pooja Prajod, Dominik Schiller, Tobias Huber, Elisabeth André

2021-04-15

In JMIR medical informatics ; h5-index 23.0

BACKGROUND : Statistical analysis, which has become an integral part of evidence-based medicine, relies heavily on data quality that is of critical importance in modern clinical research. Input data are not only at risk of being falsified or fabricated, but also of being mishandled by investigators.

OBJECTIVE : The urgent need to assure the highest data quality possible has led to the implementation of various auditing strategies designed to monitor clinical trials and detect errors of different origin that frequently occur in the field.

METHODS : A particular electronic data capture (EDC) system, which is used for data management in clinical registries, is presented including its architecture and data structure. This EDC system features an algorithm based on machine learning designed to detect anomalous patterns in quantitative data. The detection algorithm combines clustering with a series of seven distance metrics that serve to determine the strength of an anomaly. For the detection process, the thresholds and combinations of the metrics were used and the detection performance was evaluated and validated in the experiments involving simulated anomalous data and real-world data.

RESULTS : Five different clinical registries related to neuroscience were presented - all of them running in the given EDC system. Two of the registries were selected for the evaluation experiments and served also to validate the detection performance on an independent dataset. The best performing combination of the distance metrics was that of Canberra, Manhattan and Mahalanobis, whereas Cosine and Chebyshev metrics had been excluded from further analysis due to the lowest performance when used as single distance metric-based classifiers.

CONCLUSIONS : The experimental results demonstrate that the algorithm, which is universal in nature, and as such may be implemented in other EDC systems, is capable of anomalous data detection with a sensitivity exceeding 85 %.

CLINICALTRIAL :

Churová Vendula, Vyškovský Roman, Maršálová Kateřina, Kudláček David, Schwarz Daniel

2021-Apr-12

In Substance use & misuse ; h5-index 30.0

BACKGROUND : Researchers have not studied or used novel methods for identifying potential disparities for sexual minorities, those with criminal pasts, and veterans in (DOT).

METHODS : We used Bayesian logistic regression to identify factors associated with DOT, tested interaction effects, and used machine learning to classify qualitative responses.

FINDINGS : With 2,772 clients from two inpatient clinics in the Southwest United States, we found sexual minorities and females had 52% and 61%, increases and African Americans had 54% decreases in the odds of DOT. Additionally, those with a criminal past and 34.5 and older were less likely to DOT by 5% relative to clients with no prior involvement in the criminal justice system.

CONCLUSIONS : This study illustrated the disparities for women and sexual minorities in DOT as well as demonstrated novel methodological approaches to addressing previously unanswered questions.

Hanauer Matthew, Sielbeck-Mathes Kathryn, Banks Bre, Mitori Jessica, Reuveny Adi

2021-Apr-14

Bayesian analysis, against medical advice, machine learning, substance misuse

In International journal of cancer ; h5-index 82.0

Lee Sung Hak, Song In Hye, Jang Hyun-Jong

2021-Apr-13

Computational pathology, computer-aided diagnosis, convolutional neural network, digital pathology

In Pediatric radiology

Moore Michael M, Iyer Ramesh S, Sarwani Nabeel I, Sze Raymond W

2021-Apr-13

Artificial intelligence, Children, Convolutional neural network, Magnetic resonance imaging, Optimization, Segmentation

In European journal of nuclear medicine and molecular imaging ; h5-index 66.0

PURPOSE : The objective of this review was to explore the potential clinical application of unconventional non-amino acid PET radiopharmaceuticals in patients with gliomas.

METHODS : A comprehensive search strategy was used based on SCOPUS and PubMed databases using the following string: ("perfusion" OR "angiogenesis" OR "hypoxia" OR "neuroinflammation" OR proliferation OR invasiveness) AND ("brain tumor" OR "glioma") AND ("Positron Emission Tomography" OR PET). From all studies published in English, the most relevant articles were selected for this review, evaluating the mostly used PET radiopharmaceuticals in research centers, beyond amino acid radiotracers and 2-[¹⁸F]fluoro-2-deoxy-D-glucose ([¹⁸F]FDG), for the assessment of different biological features, such as perfusion, angiogenesis, hypoxia, neuroinflammation, cell proliferation, tumor invasiveness, and other biological characteristics in patients with glioma.

RESULTS : At present, the use of non-amino acid PET radiopharmaceuticals specifically designed to assess perfusion, angiogenesis, hypoxia, neuroinflammation, cell proliferation, tumor invasiveness, and other biological features in glioma is still limited.

CONCLUSION : The use of investigational PET radiopharmaceuticals should be further explored considering their promising potential and studies specifically designed to validate these preliminary findings are needed. In the clinical scenario, advancements in the development of new PET radiopharmaceuticals and new imaging technologies (e.g., PET/MR and the application of the artificial intelligence to medical images) might contribute to improve the clinical translation of these novel radiotracers in the assessment of gliomas.

Laudicella R, Quartuccio N, Argiroffi G, Alongi P, Baratto L, Califaretti E, Frantellizzi V, De Vincentis G, Del Sole A, Evangelista L, Baldari S, Bisdas S, Ceci Francesco, Iagaru Andrei

2021-Apr-13

Angiogenesis, Glioma, Neuroinflammation, PET, Perfusion, Proliferation

In Briefings in bioinformatics

Yuan Danyang, He Xiaoyu, Han Xinyin, Yang Chunyan, Liu Fei, Zhang Shuying, Luan Haijing, Li Ruilin, He Jiayin, Duan Xiaohong, Wang Dongliang, Zhou Qiming, Gao Sujun, Niu Beifang

2021-Apr-13

\n FLT3-ITD, acute myeloid leukaemia, bioinformatics, next-generation sequencing

In medRxiv : the preprint server for health sciences

Rationale : Viral infection of the respiratory tract can be associated with propagating effects on the airway microbiome, and microbiome dysbiosis may influence viral disease.

Objective : To define the respiratory tract microbiome in COVID-19 and relationship disease severity, systemic immunologic features, and outcomes.

Methods and Measurements : We examined 507 oropharyngeal, nasopharyngeal and endotracheal samples from 83 hospitalized COVID-19 patients, along with non-COVID patients and healthy controls. Bacterial communities were interrogated using 16S rRNA gene sequencing, commensal DNA viruses Anelloviridae and Redondoviridae were quantified by qPCR, and immune features were characterized by lymphocyte/neutrophil (L/N) ratios and deep immune profiling of peripheral blood mononuclear cells (PBMC).

Main Results : COVID-19 patients had upper respiratory microbiome dysbiosis, and greater change over time than critically ill patients without COVID-19. Diversity at the first time point correlated inversely with disease severity during hospitalization, and microbiome composition was associated with L/N ratios and PBMC profiles in blood. Intubated patients showed patient-specific and dynamic lung microbiome communities, with prominence of Staphylococcus . Anelloviridae and Redondoviridae showed more frequent colonization and higher titers in severe disease. Machine learning analysis demonstrated that integrated features of the microbiome at early sampling points had high power to discriminate ultimate level of COVID-19 severity.

Conclusions : The respiratory tract microbiome and commensal virome are disturbed in COVID-19, correlate with systemic immune parameters, and early microbiome features discriminate disease severity. Future studies should address clinical consequences of airway dysbiosis in COVID-19, possible use as biomarkers, and role of bacterial and viral taxa identified here in COVID-19 pathogenesis.

Merenstein Carter, Liang Guanxiang, Whiteside Samantha A, Cobián-Güemes Ana G, Merlino Madeline S, Taylor Louis J, Glascock Abigail, Bittinger Kyle, Tanes Ceylan, Graham-Wooten Jevon, Khatib Layla A, Fitzgerald Ayannah S, Reddy Shantan, Baxter Amy E, Giles Josephine R, Oldridge Derek A, Meyer Nuala J, Wherry E John, McGinniss John E, Bushman Frederic D, Collman Ronald G

2021-Apr-05

In Plant phenomics (Washington, D.C.)

Wang Jian, Wu Bizhi, Kohnen Markus V, Lin Daqi, Yang Changcai, Wang Xiaowei, Qiang Ailing, Liu Wei, Kang Jianbin, Li Hua, Shen Jing, Yao Tianhao, Su Jun, Li Bangyu, Gu Lianfeng

2021

In Data in brief

Delesma Cornelio, Amador-Bedolla Carlos, Robles Miguel, Muñiz Jesús

2021-Apr

Density functional theory, Excited state, Organic photovoltaic, Photo-isomerization

In Annals of translational medicine

Background : Rolandic epilepsy (RE) is a common pediatric idiopathic partial epilepsy syndrome. Children with RE display varying degrees of cognitive impairment. In epilepsy, age-related neuroanatomic and cognitive changes differ greatly from those observed in the healthy brain, and may be defined as accelerated brain aging. Connectome-based predictive modeling (CPM) is a recently developed machine learning approach that uses whole-brain connectivity measured with neuroimaging data ("neural fingerprints") to predict brain-behavior relationships. The aim of the study will be to develop and validate a CPM for predicting brain age in patients with RE.

Methods : A multicenter, cross-sectional study will be conducted in 5 Chinese hospitals. A total of 100 RE patients (including 50 patients receiving anti-epileptic drugs and 50 drug-naïve patients) and 100 healthy children will be recruited to undergo a neuropsychological test using the Wechsler Intelligence Scale. Magnetic resonance images will also be collected. CPM will be applied to predict the brain age of children with RE based on brain functional connectivity.

Discussion : The findings of the study will facilitate our understanding of developmental changes in the brain in children with RE and could also be an important milestone in the journey toward developing effective early interventions for this disorder.

Trial registration : The study has been registered with Chinese Clinical Trial Registry (ChiCTR2000032984).

Wang Fuqin, Yin Yu, Yang Yang, Liang Ting, Huang Tingting, He Cheng, Hu Jie, Zhang Jingjing, Yang Yanli, Xing Qianlu, Zhang Tijiang, Liu Heng

2021-Mar

MRI, Rolandic epilepsy (RE), brain age, machine learning

In Journal of public health research

BACKGROUND : Body weight has been implicated as a risk factor for latent tuberculosis infection (LTBI) and the active disease.

DESIGN AND METHODS : This study aimed to develop artificial neural network (ANN) models for predicting LTBI from body weight and other host-related disease risk factors. We used datasets from participants of the US-National Health and Nutrition Examination Survey (NHANES; 2012; n=5,156; 514 with LTBI and 4,642 controls) to develop three ANNs employing body mass index (BMI, Network I), BMI and HbA1C (as a proxy for diabetes; Network II) and BMI, HbA1C and education (as a proxy for socioeconomic status; Network III). The models were trained on n=1018 age- and sex-matched subjects equally distributed between the control and LTBI groups. The endpoint was the prediction of LTBI.

RESULTS : When data was adjusted for age, sex, diabetes and level of education, odds ratio (OR) and 95% confidence intervals (CI) for risk of LTBI with increased BMI was 0.85 (95%CI: 0.77 - 0.96, p=0.01). The three ANNs had a predictive accuracy varied from 75 to 80% with sensitivities ranged from 85% to 94% and specificities of approximately 70%. Areas under the receiver operating characteristic curve (AUC) were between 0.82 and 0.87. Optimal ANN performance was noted using BMI as a risk indicator.

CONCLUSION : Body weight can be employed in developing artificial intelligence-based tool to predict LTBI. This can be useful in precise decision making in clinical and public health practices aiming to curb the burden of tuberculosis, e.g., in the management and monitoring of the tuberculosis prevention programs and to evaluate the impact of healthy weight on tuberculosis risk and burden.

Badawi Alaa, Liu Christina J, Rehim Anas A, Gupta Alind

2021-Mar-15

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : In this age of global COVID-19 pandemic, the urgency of addressing an epidemic of obesity and associated inflammatory illnesses has come to the fore. Studies have demonstrated that interactions between single nucleotide polymorphisms (SNPs) and lifestyle interventions like food and exercise may vary metabolic outcomes, contributing to obesity and therapeutic response. However, there is a paucity of research relating outcomes from digital therapeutics to inclusion of genetic data in care interventions.

OBJECTIVE : This study aims to describe and model weight loss of subjects enrolled in a precision digital weight loss program informed by machine learning analysis of subject data, including genomic. It was hypothesized that weight loss models would exhibit better fit when incorporating genomic data than utilizing demographic and engagement variables alone.

METHODS : A cohort of 393 participants enrolled in Digbi's personalized digital care program for 120 days was analyzed retrospectively. Care protocol included the use of subject data informing precision coaching by mobile app and personal coach. Two linear regression models of weight loss in this cohort (pounds lost, percentage lost) as a function of demographic and behavioral engagement variables were fit. Genomic-enhanced models were built by adding 197 SNPs from subject genomic data as predictors, then refitting, employing Lasso regression on SNPs for variable selection. Success/failure logistic regression models were also fit, with and without genomic data.

RESULTS : 72% of subjects in this cohort lost weight, while 17% maintained stable weight. 142 subjects lost 5% within 120 days. Models describe the impact of demographic and clinical factors, behavioral engagement, and genomic risk on weight loss. The addition of genomic predictors improved the mean squared error of weight loss models (pounds lost and percent) from 70 to 60 and 16 to 13 respectively. The logistic model improved pseudo R2 from 0.193 to 0.285. Gender, engagement and specific SNPs were significantly associated with weight loss. SNPs within genes involved in metabolic pathways that process food and regulate storage of fat were associated with weight loss in this cohort. This included rs17300539_G (insulin resistance, monounsaturated fat metabolism), rs2016520_C (BMI, waist circumference, cholesterol metabolism), and rs4074995_A (calcium-potassium transport, serum calcium levels). Models described greater average weight loss for subjects having more of these risk alleles. Notably, coaching for dietary modification was personalized to these genetic risks.

CONCLUSIONS : Adding genomic information in modeling outcomes of a digital precision weight loss program greatly enhanced model accuracy. Interpretable weight loss models pointed to efficacy of coaching informed by subjects' genomic risk, accompanied by active engagement of subjects in their own success. While large-scale validation is needed, our study preliminarily supports precision dietary interventions for weight loss utilizing genetic risk, with digitally delivered recommendations alongside health-coaching to improve intervention efficacy.

CLINICALTRIAL :

Sinha Ranjan, Kachru Dashyanng, Ricchetti Roshni Ray, Singh-Rambiritch Simitha, Muthukumar Karthik Marimuthu, Singaravel Vidhya, Irudayanathan Carmel, Reddy-Sinha Chandana, Junaid Imran, Sharma Garima, Airey Catherine, Francis-Lyon Patricia Alice

2021-Apr-11

ArXiv Preprint

Sidharth R, Abhiraj Tiwari, Parthivi Choubey, Saisha Kashyap, Sahil Khose, Kumud Lakara, Nishesh Singh, Ujjwal Verma

2021-04-15

ArXiv Preprint

Sidharth R, Abhiraj Tiwari, Parthivi Choubey, Saisha Kashyap, Sahil Khose, Kumud Lakara, Nishesh Singh, Ujjwal Verma

2021-04-15

In Annals of translational medicine

Background : Artificial intelligence technology is widely used in the medical industry. Our retrospective study evaluated the effectiveness of an AI-CDSS in improving the incidence of hospital-related VTE and the impact of anticoagulant drug use.

Methods : This study collected relevant data on adult patients over 18 years of age who are not discharged 24 hours, from January to July 2019 and from January to July 2020, the VTE high-risk department of Ruijin Hospital. Before and after using AI-CDSS, the incidence of hospital-related VTE and using anticoagulants were analyzed.

Results : Between January to July 2019 and January to July 2020, 3,565 and 4,423 adult patients over 18 years old were hospitalized in our hospital and were designed as a control group and intervention group, respectively (7,988 in total). Both groups had similar baseline characteristics. There were 4,716 (59.03%) male patients, the mean age was 60.43±13.09 years, and the mean stay was 7.56±7.76 days. More than half of the patients (4,605, 57.58%) came from the respiratory. VTE events during hospitalization occurred in 41 patients; overall, 5.13/1,000 (41 episodes in 7,988 patients). Compared with the control group, before implementing AI-CDSS, the rate of VTE during hospitalization was reduced from 5.89/1,000 (21 episodes in 3,565 patients) to 4.75/1,000 patients (20 episodes in 4,423 patients) (relative reduction of 19.35%) in the intervention group. The use rate of anticoagulant drugs was increased from 19.97% (712/3,565) in the control group to 22.88% (1,012/4,423) in intervention group [P<0.01, odds ratio (OR): 1.19, 95 percent confidence interval (95% CI) (1.07-1.32)], (relative 14.57% increase). Poisson's regression results showed that department, age ≥75 years [OR: 3.09, 95% Cl (1.45-6.33)], duration of hospitalization [OR: 1.04, 95% CI (1.03-1.05)], heart failure [OR: 5.13, 95% CI (1.74-13.54)] and renal failure [OR: 3.60, 95% CI (0.90-11.34)] were high-risk factors for VTE events.

Conclusions : Implementing AI-CDSS can help clinicians identify hospitalized patients at increased VTE risk, take effective preventive measures, and improve clinicians' compliance with the American College of Chest Physicians (ACCP) guidelines.

Zhou Shuai, Ma Xudong, Jiang Songyi, Huang Xiaoyan, You Yi, Shang Hanbing, Lu Yong

2021-Mar

China, a retrospective study, artificial intelligence-based clinical decision support system (AI-CDSS), hospitalization rate, venous thromboembolism (VTE)

In Annals of translational medicine

Background : The shape, size, and surface information relating to the glenoid fossae and condyles in temporomandibular joints (TMJ) are essential for diagnosing and treating. Patients with TMJ disease often have surface abrasion which may cause fuzzy edges in computed tomography (CT) imaging, especially for low-dose CT, making TMJ segmentation more difficult.

Methods : In this paper, an automatic segmentation algorithm based on deep learning and post-processing was introduced. First, U-Net was used to divide images into 3 categories: glenoid fossae, condyles, and background. For structural fractures in these divided images, the internal force constraint of a snake model was used to replenish the integrity of the fracture boundary in a post-processing operation, and the initial boundary of the snake was obtained based on the basis of the tracking concept. A total of 206 cases of low-dose CT were used to verify the effectiveness of the algorithm, and such indicators as the Dice coefficient (DC) and mean surface distance (MSD) were used to evaluate the agreement between experimental results and the gold standard.

Results : The proposed method is tested on a self-collected dataset. The results demonstrate that proposed method achieves state-of-the-art performance in terms of DCs = 0.92±0.03 (condyles) and 0.90±0.04 (glenoid fossae), and MSDs =0.20±0.19 mm (condyles) and 0.19±0.08 mm (glenoid fossae).

Conclusions : This study is the first to focus on the simultaneous segmentation of TMJ glenoid fossae and condyles. The proposed U-Net + tracking-based algorithm showed a relatively high segmentation efficiency, enabling it to achieve sought-after segmentation accuracy.

Liu Yi, Lu Yao, Fan Yubo, Mao Longxia

2021-Mar

Biomedical imaging, computer-aided diagnosis, deep learning, image segmentation, low-dose computed tomography (low-dose CT), tracking

In Translational andrology and urology

Montironi Rodolfo, Cheng Liang, Cimadamore Alessia, Mazzucchelli Roberta, Scarpelli Marina, Santoni Matteo, Massari Francesco, Lopez-Beltran Antonio

2021-Mar

2005 ISUP Gleason modified system, 2014 ISUP Gleason modified system, Genitourinary Pathology Society (GUPS), Gleason grading, International Society of Urological Pathology (ISUP), Prostate cancer (PCa), prognostic grade grouping

In Global health journal (Amsterdam, Netherlands)

Zhang Yanjie, Tambo Ernest, Djuikoue Ingrid C, Tazemda Gildas K, Fotsing Michael F, Zhou Xiao-Nong

2021-Mar

Coronavirus disease 2019 (COVID-19), Governance, Pandemic, Response, Risk communication and community engagement (RCCE), Trust, Vaccination

In Healthcare technology letters

Decaro Cristoforo, Montanari Giovanni Battista, Bianconi Marco, Bellanca Gaetano

2021-Apr

In Proceedings. IEEE International Symposium on Biomedical Imaging

Yin Shi, Peng Qinmu, Li Hongming, Zhang Zhengqiang, You Xinge, Fischer Katherine, Furth Susan L, Tasian Gregory E, Fan Yong

2020-Apr

Multi-instance deep learning, congenital abnormalities of the kidney and urinary tract, sagittal view, transverse view, ultrasound images

In Scientific reports ; h5-index 158.0

Nishimori Makoto, Kiuchi Kunihiko, Nishimura Kunihiro, Kusano Kengo, Yoshida Akihiro, Adachi Kazumasa, Hirayama Yasutaka, Miyazaki Yuichiro, Fujiwara Ryudo, Sommer Philipp, El Hamriti Mustapha, Imada Hiroshi, Takemoto Makoto, Takami Mitsuru, Shinohara Masakazu, Toh Ryuji, Fukuzawa Koji, Hirata Ken-Ichi

2021-Apr-13

In NPJ digital medicine

Shang Ning, Khan Atlas, Polubriaginof Fernanda, Zanoni Francesca, Mehl Karla, Fasel David, Drawz Paul E, Carrol Robert J, Denny Joshua C, Hathcock Matthew A, Arruda-Olson Adelaide M, Peissig Peggy L, Dart Richard A, Brilliant Murray H, Larson Eric B, Carrell David S, Pendergrass Sarah, Verma Shefali Setia, Ritchie Marylyn D, Benoit Barbara, Gainer Vivian S, Karlson Elizabeth W, Gordon Adam S, Jarvik Gail P, Stanaway Ian B, Crosslin David R, Mohan Sumit, Ionita-Laza Iuliana, Tatonetti Nicholas P, Gharavi Ali G, Hripcsak George, Weng Chunhua, Kiryluk Krzysztof

2021-Apr-13

In Scientific reports ; h5-index 158.0

Yu Go-Eun, Shin Younhee, Subramaniyam Sathiyamoorthy, Kang Sang-Ho, Lee Si-Myung, Cho Chuloh, Lee Seung-Sik, Kim Chang-Kug

2021-Apr-13

In Scientific reports ; h5-index 158.0

Park Kawngwoo, Bae Gi Hwan, Kim Woo Kyung, Yoo Chan-Jong, Park Cheol Wan, Kim Soo-Ki, Cha Jihye, Kim Jin Wook, Jung Jaehun

2021-Apr-13

In ISA transactions

Al-Gabalawy Mostafa

2021-Apr-01

Artificial intelligence, Deep learning, Deep reinforcement learning, Hybrid controller, MPC, Machine learning, Planner robot, Reinforcement learning

In World journal of surgical oncology

BACKGROUND : Pancreatic adenocarcinoma (PAAD) is a nonimmunogenic tumor, and very little is known about the relationship between the host immune response and patient survival. We aimed to develop an immune prognostic model (IPM) and analyze its relevance to the tumor immune profiles of patients with PAAD.

METHODS : We investigated differentially expressed genes between tumor and normal tissues in the TCGA PAAD cohort. Immune-related genes were screened from highly variably expressed genes with weighted gene correlation network analysis (WGCNA) to construct an IPM. Then, the influence of IPM on the PAAD immune profile was comprehensively analyzed.

RESULTS : A total of 4902 genes highly variably expressed among primary tumors were used to construct a weighted gene coexpression network. One hundred seventy-five hub genes in the immune-related module were used for machine learning. Then, we established an IPM with four core genes (FCGR2B, IL10RA, and HLA-DRA) to evaluate the prognosis. The risk score predicted by IPM was an independent prognostic factor and had a high predictive value for the prognosis of patients with PAAD. Moreover, we found that the patients in the low-risk group had higher cytolytic activity and lower innate anti-PD-1 resistance (IPRES) signatures than patients in the high-risk group.

CONCLUSIONS : Unlike the traditional methods that use immune-related genes listed in public databases to screen prognostic genes, we constructed an IPM through WGCNA to predict the prognosis of PAAD patients. In addition, an IPM prediction of low risk indicated enhanced immune activity and a decreased anti-PD-1 therapeutic response.

Gu Xiaoqiang, Zhang Qiqi, Wu Xueying, Fan Yue, Qian Jianxin

2021-Apr-12

Immune profile, Immune prognostic model, Immunotherapy, Pancreatic cancer, WGCNA

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : Due to the COVID-19 pandemic and the imminent collapse of healthcare systems following the excessive consumption of financial, hospital, and medicinal resources, the WHO changed the alert level on the COVID-19 pandemic from high to very high. Meanwhile, the world began to favor less expensive and more precise COVID-19 detection methods.

OBJECTIVE : Machine vision-based COVID-19 detection methods especially Deep learning as a diagnostic technique in the early stages of the disease have found great importance during the pandemic. This study aimed to design a highly efficient CAD system for COVID-19 by using a NASNet-based algorithm.

METHODS : A state-of-the-art pre-trained CNN network for image feature extraction, called NASNet, was adopted to identify patients with COVID-19 in the first stages of the disease. A local dataset, comprising 10153 CT-scan images of 190 patients with COVID-19 and 59 with Non Covid-19, was used.

RESULTS : After fitting on the training dataset, hyper-parameter tuning and finally topological alterations of the classifier block, the proposed NASNet-based model was evaluated on the test dataset and yielded remarkable results. The proposed model's performance achieved a detection sensitivity, specificity, and accuracy of 0.999, 0.986, and 0.996, respectively.

CONCLUSIONS : The proposed model achieved acceptable results in the categorization of two data classes. Therefore, a CAD system was designed based on this model for COVID-19 detection using multiple lung CT scans. The system managed to differentiate all the COVID-19 cases from non-COVID-19 ones without any error in the application phase. Overall, the proposed deep learning-based CAD system can greatly aid radiologists in the detection of COVID-19 in its early stages. During the COVID-19 pandemic, the use of CAD system as a screening tool accelerates the process of disease detection and prevents the loss of healthcare resources.

CLINICALTRIAL :

Ghaderzadeh Mustafa, Asadi Farkhondeh, Jafari Ramezan, Bashash Davood, Abolghasemi Hassan, Aria Mehrad

2021-Apr-03

In BMC biology

BACKGROUND : Breast cancer is amongst the 10 first causes of death in women worldwide. Around 20% of patients are misdiagnosed leading to early metastasis, resistance to treatment and relapse. Many clinical and gene expression profiles have been successfully used to classify breast tumours into 5 major types with different prognosis and sensitivity to specific treatments. Unfortunately, these profiles have failed to subclassify breast tumours into more subtypes to improve diagnostics and survival rate. Alternative splicing is emerging as a new source of highly specific biomarkers to classify tumours in different grades. Taking advantage of extensive public transcriptomics datasets in breast cancer cell lines (CCLE) and breast cancer tumours (TCGA), we have addressed the capacity of alternative splice variants to subclassify highly aggressive breast cancers.

RESULTS : Transcriptomics analysis of alternative splicing events between luminal, basal A and basal B breast cancer cell lines identified a unique splicing signature for a subtype of tumours, the basal B, whose classification is not in use in the clinic yet. Basal B cell lines, in contrast with luminal and basal A, are highly metastatic and express epithelial-to-mesenchymal (EMT) markers, which are hallmarks of cell invasion and resistance to drugs. By developing a semi-supervised machine learning approach, we transferred the molecular knowledge gained from these cell lines into patients to subclassify basal-like triple negative tumours into basal A- and basal B-like categories. Changes in splicing of 25 alternative exons, intimately related to EMT and cell invasion such as ENAH, CD44 and CTNND1, were sufficient to identify the basal-like patients with the worst prognosis. Moreover, patients expressing this basal B-specific splicing signature also expressed newly identified biomarkers of metastasis-initiating cells, like CD36, supporting a more invasive phenotype for this basal B-like breast cancer subtype.

CONCLUSIONS : Using a novel machine learning approach, we have identified an EMT-related splicing signature capable of subclassifying the most aggressive type of breast cancer, which are basal-like triple negative tumours. This proof-of-concept demonstrates that the biological knowledge acquired from cell lines can be transferred to patients data for further clinical investigation. More studies, particularly in 3D culture and organoids, will increase the accuracy of this transfer of knowledge, which will open new perspectives into the development of novel therapeutic strategies and the further identification of specific biomarkers for drug resistance and cancer relapse.

Villemin Jean-Philippe, Lorenzi Claudio, Cabrillac Marie-Sarah, Oldfield Andrew, Ritchie William, Luco Reini F

2021-Apr-12

Alternative splicing, Basal-like, Breast Cancer, Epithelial-to-mesenchymal transition, Machine learning classification, Survival

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : Although the potential of big data analytics for health care is well recognized, evidence is lacking on its effects on public health.

OBJECTIVE : The aim of this study was to assess the impact of the use of big data analytics on people's health based on the health indicators and core priorities in the World Health Organization (WHO) General Programme of Work 2019/2023 and the European Programme of Work (EPW), approved and adopted by its Member States, in addition to SARS-CoV-2-related studies. Furthermore, we sought to identify the most relevant challenges and opportunities of these tools with respect to people's health.

METHODS : Six databases (MEDLINE, Embase, Cochrane Database of Systematic Reviews via Cochrane Library, Web of Science, Scopus, and Epistemonikos) were searched from the inception date to September 21, 2020. Systematic reviews assessing the effects of big data analytics on health indicators were included. Two authors independently performed screening, selection, data extraction, and quality assessment using the AMSTAR-2 (A Measurement Tool to Assess Systematic Reviews 2) checklist.

RESULTS : The literature search initially yielded 185 records, 35 of which met the inclusion criteria, involving more than 5,000,000 patients. Most of the included studies used patient data collected from electronic health records, hospital information systems, private patient databases, and imaging datasets, and involved the use of big data analytics for noncommunicable diseases. "Probability of dying from any of cardiovascular, cancer, diabetes or chronic renal disease" and "suicide mortality rate" were the most commonly assessed health indicators and core priorities within the WHO General Programme of Work 2019/2023 and the EPW 2020/2025. Big data analytics have shown moderate to high accuracy for the diagnosis and prediction of complications of diabetes mellitus as well as for the diagnosis and classification of mental disorders; prediction of suicide attempts and behaviors; and the diagnosis, treatment, and prediction of important clinical outcomes of several chronic diseases. Confidence in the results was rated as "critically low" for 25 reviews, as "low" for 7 reviews, and as "moderate" for 3 reviews. The most frequently identified challenges were establishment of a well-designed and structured data source, and a secure, transparent, and standardized database for patient data.

CONCLUSIONS : Although the overall quality of included studies was limited, big data analytics has shown moderate to high accuracy for the diagnosis of certain diseases, improvement in managing chronic diseases, and support for prompt and real-time analyses of large sets of varied input data to diagnose and predict disease outcomes.

TRIAL REGISTRATION : International Prospective Register of Systematic Reviews (PROSPERO) CRD42020214048; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=214048.

Borges do Nascimento Israel Júnior, Marcolino Milena Soriano, Abdulazeem Hebatullah Mohamed, Weerasekara Ishanka, Azzopardi-Muscat Natasha, Gonçalves Marcos André, Novillo-Ortiz David

2021-Apr-13

World Health Organization, big data, big data analytics, evidence-based medicine, health status, machine learning, overview, public health, secondary data analysis, systematic review

In BMC bioinformatics

BACKGROUND : Microsatellite instability (MSI) is a common genomic alteration in colorectal cancer, endometrial carcinoma, and other solid tumors. MSI is characterized by a high degree of polymorphism in microsatellite lengths owing to the deficiency in the mismatch repair system. Based on the degree, MSI can be classified as microsatellite instability-high (MSI-H) and microsatellite stable (MSS). MSI is a predictive biomarker for immunotherapy efficacy in advanced/metastatic solid tumors, especially in colorectal cancer patients. Several computational approaches based on target panel sequencing data have been used to detect MSI; however, they are considerably affected by the sequencing depth and panel size.

RESULTS : We developed MSIFinder, a python package for automatic MSI classification, using random forest classifier (RFC)-based genome sequencing, which is a machine learning technology. We included 19 MSI-H and 25 MSS samples as training sets. First, we selected 54 feature markers from the training sets, built an RFC model, and validated the classifier using a test set comprising 21 MSI-H and 379 MSS samples. With this test set, MSIFinder achieved a sensitivity (recall) of 1.0, a specificity of 0.997, an accuracy of 0.998, a positive predictive value of 0.954, an F1 score of 0.977, and an area under the curve of 0.999. To further verify the robustness and effectiveness of the model, we used a prospective cohort consisting of 18 MSI-H samples and 122 MSS samples. MSIFinder achieved a sensitivity (recall) of 1.0 and a specificity of 1.0. We discovered that MSIFinder is less affected by a low sequencing depth and can achieve a concordance of 0.993 while exhibiting a sequencing depth of 100×. Furthermore, we realized that MSIFinder is less affected by the panel size and can achieve a concordance of 0.99 when the panel size is 0.5 M (million bases).

CONCLUSION : These results indicate that MSIFinder is a robust and effective MSI classification tool that can provide reliable MSI detection for scientific and clinical purposes.

Zhou Tao, Chen Libin, Guo Jing, Zhang Mengmeng, Zhang Yanrui, Cao Shanbo, Lou Feng, Wang Haijun

2021-Apr-12

Genome sequencing, Immunotherapy, Machine learning technology, Microsatellite instability, Random forest classifier

In Chinese journal of academic radiology

Objective : This study aimed to compare quantifiable radiologic findings and their dynamic change throughout the clinical course of common and severe coronavirus disease 2019 (COVID-19), and to provide valuable evidence for radiologic classification of the two types of this disease.

Methods : 112 patients with laboratory-confirmed COVID-19 were retrospectively analyzed. Volumetric percentage of infection and density of the lung were measured by a computer-aided software. Clinical parameters were recorded to reflect disease progression. Baseline data and dynamic change were compared between two groups and a decision-tree algorithm was developed to determine the cut-off value for classification.

Results : 93 patients were finally included and were divided into common group (n = 76) and severe group (n = 17) based on current criteria. Compared with common patients, severe patients experienced shorter advanced stage, peak time and plateau, but longer absorption stage. The dynamic change of volume and density coincided with the clinical course. The interquartile range of volumetric percentage of the two groups were 1.0-7.2% and 11.4-31.2%, respectively. Baseline volumetric percentage of infection was significantly higher in severe group, and the cut-off value of it was 10.10%.

Conclusions : Volumetric percentage between severe and common patients was significantly different. Because serial CT scans are systemically performed in patients with COVID-19 pneumonia, this quantitative analysis can simultaneously provide valuable information for physicians to evaluate their clinical course and classify common and severe patients accurately.

Yang Chongtu, Cao Guijuan, Liu Fen, Liu Jiacheng, Huang Songjiang, Xiong Bin

2021-Apr-08

Artificial intelligence, Computer-assisted, Coronavirus disease 2019, Decision trees, Multidetector computed tomography, Numerical analysis

In BMC bioinformatics

BACKGROUND : Technological and research advances have produced large volumes of biomedical data. When represented as a network (graph), these data become useful for modeling entities and interactions in biological and similar complex systems. In the field of network biology and network medicine, there is a particular interest in predicting results from drug-drug, drug-disease, and protein-protein interactions to advance the speed of drug discovery. Existing data and modern computational methods allow to identify potentially beneficial and harmful interactions, and therefore, narrow drug trials ahead of actual clinical trials. Such automated data-driven investigation relies on machine learning techniques. However, traditional machine learning approaches require extensive preprocessing of the data that makes them impractical for large datasets. This study presents wide range of machine learning methods for predicting outcomes from biomedical interactions and evaluates the performance of the traditional methods with more recent network-based approaches.

RESULTS : We applied a wide range of 32 different network-based machine learning models to five commonly available biomedical datasets, and evaluated their performance based on three important evaluations metrics namely AUROC, AUPR, and F1-score. We achieved this by converting link prediction problem as binary classification problem. In order to achieve this we have considered the existing links as positive example and randomly sampled negative examples from non-existant set. After experimental evaluation we found that Prone, ACT and [Formula: see text] are the top 3 best performers on all five datasets.

CONCLUSIONS : This work presents a comparative evaluation of network-based machine learning algorithms for predicting network links, with applications in the prediction of drug-target and drug-drug interactions, and applied well known network-based machine learning methods. Our work is helpful in guiding researchers in the appropriate selection of machine learning methods for pharmaceutical tasks.

Abbas Khushnood, Abbasi Alireza, Dong Shi, Niu Ling, Yu Laihang, Chen Bolun, Cai Shi-Min, Hasan Qambar

2021-Apr-12

Data-driven drug discovery, Drug-target prediction, Network link prediction, Poly-pharmacy, Poly-pharmacy side effects prediction

In Biomedical signal processing and control

Sharifrazi Danial, Alizadehsani Roohallah, Roshanzamir Mohamad, Joloudari Javad Hassannataj, Shoeibi Afshin, Jafari Mahboobeh, Hussain Sadiq, Sani Zahra Alizadeh, Hasanzadeh Fereshteh, Khozeimeh Fahime, Khosravi Abbas, Nahavandi Saeid, Panahiazar Maryam, Zare Assef, Islam Sheikh Mohammed Shariful, Acharya U Rajendra

2021-Apr-08

CNN., Covid-19, Data Mining, Deep Learning, Feature Extraction, Image Processing, Machine Learning, SVM, Sobel operator

In Cerebrovascular diseases (Basel, Switzerland)

BACKGROUND AND PURPOSE : Randomized controlled trials have demonstrated the importance of time to endovascular therapy (EVT) in clinical outcomes in large vessel occlusion (LVO) acute ischemic stroke. Delays to treatment are particularly prevalent when patients require a transfer from hospitals without EVT capability onsite. A computer-aided triage system, Viz LVO, has the potential to streamline workflows. This platform includes an image viewer, a communication system, and an artificial intelligence (AI) algorithm that automatically identifies suspected LVO strokes on CTA imaging and rapidly triggers alerts. We hypothesize that the Viz application will decrease time-to-treatment, leading to improved clinical outcomes.

METHODS : A retrospective analysis of a prospectively maintained database was assessed for patients who presented to a stroke center currently utilizing Viz LVO and underwent EVT following transfer for LVO stroke between July 2018 and March 2020. Time intervals and clinical outcomes were compared for 55 patients divided into pre- and post-Viz cohorts.

RESULTS : The median initial door-to-neuroendovascular team (NT) notification time interval was significantly faster (25.0 min [IQR = 12.0] vs. 40.0 min [IQR = 61.0]; p = 0.01) with less variation (p < 0.05) following Viz LVO implementation. The median initial door-to-skin puncture time interval was 25 min shorter in the post-Viz cohort, although this was not statistically significant (p = 0.15).

CONCLUSIONS : Preliminary results have shown that Viz LVO implementation is associated with earlier, more consistent NT notification times. This application can serve as an early warning system and a failsafe to ensure that no LVO is left behind.

Morey Jacob R, Zhang Xiangnan, Yaeger Kurt A, Fiano Emily, Marayati Naoum Fares, Kellner Christopher P, De Leacy Reade A, Doshi Amish, Tuhrim Stanley, Fifi Johanna T

2021-Apr-13

Artificial intelligence, CT angiography, Stroke, Technology, Thrombectomy

In Annals of nutrition & metabolism

BACKGROUND : Fatty liver disease (FLD) has become a rampant condition. It is associated with a high rate of morbidity and mortality in a population. The condition is commonly referred as FLD. Early prediction of FLD would allow patients to take necessary preventive, diagnosis, and treatment. The main objective of this research is to develop a machine learning (ML) model to predict FLD that can help medics to classify individuals at high risk of FLD, make novel diagnosis, management, and prevention for FLD.

METHODS : Total of 3,419 subjects were recruited with 845 having been screened for FLD. Classification models were used in the detection of the disease. These models include logistic regression (LR), random forest (RF), artificial neural networks (ANNs), k-nearest neighbors (KNNs), extreme gradient boosting (XGBoost), and linear discriminant analysis (LDA). Predictive accuracy was assessed by area under curve (AUC), sensitivity, specificity, positive predictive value, and negative predictive value.

RESULTS : We demonstrated that ML models give more accurate predictions, the best accuracy reached to 0.9415 in the XGBoost model. Feature importance analysis not only confirmed some well-known FLD risk factors, but also demonstrated several novel features for predicting the risk of FLD, such as hemoglobin.

CONCLUSION : By implementing the XGBoost model, physicians can efficiently identify FLD in general patients; this would help in prevention, early treatment, and management of FLD.

Pei Xieyi, Deng Qingqing, Liu Zhuo, Yan Xiang, Sun Weiping

2021-Apr-13

Classification model, Extreme gradient boosting, Fatty liver disease, Machine learning

In Oncology

INTRODUCTION : Radiomics now has significant momentum in the era of precision medicine. Glioma is one of the pathologies that has been extensively evaluated by radiomics. However, this technique has not been incorporated into clinical practice. In this systematic review, we selected and reviewed the published studies about glioma grading by radiomics to evaluate this technique's feasibility and its challenges.

MATERIAL AND METHODS : Using seven different search strings, we considered all published English manuscripts from 2015 to September 2020 in PubMed, Embase, and Scopus databases. After implementing the exclusion and inclusion criteria, the final papers were selected for the methodological quality assessment based on our in-house Modified Radiomics Standard Scoring (RQS) containing 43 items (minimum score of 0, maximum score of 44). Finally, we offered our opinion about the challenges and weaknesses of the selected papers.

RESULTS : By our search, 1,177 manuscripts were found (485 in PubMed, 343 in Embase, and 349 in Scopus). After the implementation of inclusion and exclusion criteria, 18 papers remained for the final analysis by RQS. The total RQS score ranged from 26 (59% of maximum possible score) to 43 (97% of maximum possible score) with a mean of 33.5 (76% of maximum possible score).

CONCLUSION : The current studies are promising but very heterogeneous in design with high variation in the radiomics software, the number of extracted features, the number of selected features, and machine learning models. All of the studies were retrospective in design; many are based on small datasets and/or suffer from class imbalance and lack of external validation data-sets.

Tabatabaei Mohsen, Razaei Ali, Sarrami Amir Hossein, Saadatpour Zahra, Singhal Aparna, Sotoudeh Houman

2021-Apr-13

Artificial intelligence, Glioma, Neoplasm grading, Systematic review

In Journal of neural engineering ; h5-index 52.0

OBJECTIVE : To explore the viability of developing a computer-aided diagnostic system for Parkinsonian syndromes using dynamic [11C]raclopride PET and T1-weighted MRI data.

APPROACH : The biological heterogeneity of Parkinsonian syndromes renders their statistical classification a challenge. The unique combination of structural and molecular imaging data allowed different classifier designs to be tested. Datasets from dynamic [11C]raclopride PET and T1-weighted MRI scans were acquired from six groups of participants: healthy controls (CTRL n=15), patients with Parkinson's disease (PD n=27), multiple system atrophy (MSA n=8), corticobasal degeneration (CBD n=6), and dementia with Lewy bodies (DLB n=5). MSA, CBD, and DLB patients were classified into one category designated as atypical parkinsonism (AP). The distribution volume ratio (DVR) kinetic parameters obtained from PET data were used to quantify the reversible tracer binding to D2/D3 receptors in the subcortical regions of interest (ROI). Grey matter (GM) volumes obtained from the MRI data were used to quantify GM atrophy across cortical, subcortical, and cerebellar ROI.

RESULTS : The classifiers CTRL vs PD and CTRL vs AP achieved the highest balanced accuracy combining DVR and GM (DVR-GM) features (96.7%, 92.1%, respectively), followed by the classifiers designed with DVR features (93.3%, 88.8%, respectively), and GM features (69.6%, 86.1%, respectively). In contrast, the classifier PD vs AP showed the highest balanced accuracy (78.9%) using DVR features only. The integration of DVR-GM (77.9%) and GM features (72.7%) produced inferior performances. The classifier CTRL vs PD vs AP showed high weighted balanced accuracy when DVR (80.5%) or DVR-GM features (79.9%) were integrated. GM features revealed poorer performance (59.5%).

SIGNIFICANCE : This work was unique in its combination of structural and molecular imaging features in binary and triple category classifications. We were able to demonstrate improved binary classification of healthy/diseased status (concerning both PD and AP) and equate performance to DVR features in multiclass classifications.

Martins Ricardo, Oliveira Francisco Paulo Marques, Moreira Fradique, Moreira Ana Paula, Abrunhosa Antero, Januario Cristina, Castelo-Branco Miguel

2021-Apr-13

11C-Raclopride positron emission tomography, Computer-aided diagnosis, Parkinsonian syndromes, machine learning, magnetic resonance imaging, multimodality imaging

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : Due to the COVID-19 pandemic and the imminent collapse of healthcare systems following the excessive consumption of financial, hospital, and medicinal resources, the WHO changed the alert level on the COVID-19 pandemic from high to very high. Meanwhile, the world began to favor less expensive and more precise COVID-19 detection methods.

OBJECTIVE : Machine vision-based COVID-19 detection methods especially Deep learning as a diagnostic technique in the early stages of the disease have found great importance during the pandemic. This study aimed to design a highly efficient CAD system for COVID-19 by using a NASNet-based algorithm.

METHODS : A state-of-the-art pre-trained CNN network for image feature extraction, called NASNet, was adopted to identify patients with COVID-19 in the first stages of the disease. A local dataset, comprising 10153 CT-scan images of 190 patients with COVID-19 and 59 with Non Covid-19, was used.

RESULTS : After fitting on the training dataset, hyper-parameter tuning and finally topological alterations of the classifier block, the proposed NASNet-based model was evaluated on the test dataset and yielded remarkable results. The proposed model's performance achieved a detection sensitivity, specificity, and accuracy of 0.999, 0.986, and 0.996, respectively.

CONCLUSIONS : The proposed model achieved acceptable results in the categorization of two data classes. Therefore, a CAD system was designed based on this model for COVID-19 detection using multiple lung CT scans. The system managed to differentiate all the COVID-19 cases from non-COVID-19 ones without any error in the application phase. Overall, the proposed deep learning-based CAD system can greatly aid radiologists in the detection of COVID-19 in its early stages. During the COVID-19 pandemic, the use of CAD system as a screening tool accelerates the process of disease detection and prevents the loss of healthcare resources.

CLINICALTRIAL :

Ghaderzadeh Mustafa, Asadi Farkhondeh, Jafari Ramezan, Bashash Davood, Abolghasemi Hassan, Aria Mehrad

2021-Apr-03

In The Journal of surgical research

BACKGROUND : Machine learning (ML) has garnered increasing attention as a means to quantitatively analyze the growing and complex medical data to improve individualized patient care. We herein aim to critically examine the current state of ML in predicting surgical outcomes, evaluate the quality of currently available research, and propose areas of improvement for future uses of ML in surgery.

METHODS : A systematic review was conducted in accordance with the Preferred Reporting Items for a Systematic Review and Meta-Analysis (PRISMA) checklist. PubMed, MEDLINE, and Embase databases were reviewed under search syntax "machine learning" and "surgery" for papers published between 2015 and 2020.

RESULTS : Of the initial 2677 studies, 45 papers met inclusion and exclusion criteria. Fourteen different subspecialties were represented with neurosurgery being most common. The most frequently used ML algorithms were random forest (n = 19), artificial neural network (n = 17), and logistic regression (n = 17). Common outcomes included postoperative mortality, complications, patient reported quality of life and pain improvement. All studies which compared ML algorithms to conventional studies which used area under the curve (AUC) to measure accuracy found improved outcome prediction with ML models.

CONCLUSIONS : While still in its early stages, ML models offer surgeons an opportunity to capitalize on the myriad of clinical data available and improve individualized patient care. Limitations included heterogeneous outcome and imperfect quality of some of the papers. We therefore urge future research to agree upon methods of outcome reporting and require basic quality standards.

Elfanagely Omar, Toyoda Yoshiko, Othman Sammy, Mellia Joseph A, Basta Marten, Liu Tony, Kording Konrad, Ungar Lyle, Fischer John P

2021-Apr-10

Artificial intelligence, Machine learning, Natural language processing, Surgical outcomes

In Current medical imaging

Musa Aguiar Paula, Zarantonello Paola, Aparisi Gómez Maria Pilar

2021-04-12

CT, DXA, Imaging, MRI, Neoplastic, Osteoporosis, Radiographs, Vertebral Fracture

In Information sciences

Ouyang Liwei, Yuan Yong, Cao Yumeng, Wang Fei-Yue

2021-Apr-08

blockchain, collaborative early warning, federated learning, learning markets, smart contracts

In Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society

Hammouda K, Khalifa F, Soliman A, Ghazal M, El-Ghar M Abou, Badawy M A, Darwish H E, Khelifi A, El-Baz A

2021-Mar-31

CAD system, Classification bladder cancer staging, Functional features, Morphological features, Texture features

In Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society

Xiao Yi, Wang Xiang, Li Qingchu, Fan Rongrong, Chen Rutan, Shao Ying, Chen Yanbo, Gao Yaozong, Liu Aie, Chen Lei, Liu Shiyuan

2021-Mar-04

Computed tomography (CT), Deep learning, Lung nodule detection, Phantom

In Journal of stomatology, oral and maxillofacial surgery

Hennocq Quentin, Khonsari Roman Hossein, Benoît Vincent, Rio Marlène, Garcelon Nicolas

2021-Apr-10

deep learning, diagnosis, dysmorphology, literature review, machine learning, photograph

In Gene

Quoc Khanh Le Nguyen, Thi Do Duyen, Nguyen Trinh-Trung-Duong, Anh Le Quynh

2021-Apr-10

Kruppel-like factor, SMOTE imbalance, Zinc finger, eXtreme Gradient Boosting, feature selection, protein sequence

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To develop classification criteria for 25 of the most common uveitides.

DESIGN : Machine learning using 5766 cases of 25 uveitides.

METHODS : Cases were collected in an informatics-designed preliminary database. Using formal consensus techniques, a final database was constructed of 4046 cases achieving supermajority agreement on the diagnosis. Cases were analyzed within uveitic class and were split into a training set and a validation set. Machine learning used multinomial logistic regression with lasso regularization on the training set to determine a parsimonious set of criteria for each disease and to minimize misclassification rates. The resulting criteria were evaluated in the validation set. Accuracy of the rules developed to express the machine learning criteria was evaluated by a masked observer in a 10% random sample of cases.

RESULTS : Overall accuracy estimates by uveitic class in the validation set were: anterior uveitides 96.7% (95% confidence interval [CI] 92.4, 98.6); intermediate uveitides 99.3% (95% CI 96.1, 99.9); posterior uveitides 98.0% (95% CI 94.3, 99.3); panuveitides 94.0% (95% CI 89.0, 96.8); and infectious posterior/panuveitides 93.3% (95% CI 89.1, 96.3). Accuracies of the masked evaluation of the "rules" were: anterior uveitides 96.5% (95% CI 91.4, 98.6) intermediate uveitides 98.4% (91.5, 99.7), posterior uveitides 99.2% (95% CI 95.4, 99.9), panuveitides 98.9% (95% CI 94.3, 99.8), and infectious posterior/panuveitides 98.8% (95% CI 93.4, 99.9).

CONCLUSIONS : The classification criteria for these 25 uveitides had high overall accuracy (i.e. low misclassification rates) and appeared to perform well enough for use in clinical and translational research.

Jabs Douglas A, McCluskey Peter, Oden Neal, Palestine Alan G, Peterson Jan S, Saleem Sophia M, Thorne Jennifer E, Trusko Brett E

2021-Apr-10

In International journal of cosmetic science

OBJECTIVE : The first objective of this study was to apply computer vision and machine learning techniques to quantify the effects of haircare treatments on hair assembly and to identify correctly whether unknown tresses were treated or not. The second objective was to explore and compare the performance of human assessment with that obtained from artificial intelligence (AI) algorithms.

METHODS : Machine learning was applied to a dataset of hair tress images (virgin and bleached), both untreated and treated with a shampoo and conditioner set, aimed at increasing hair volume whilst improving alignment and reducing the flyway of the hair. The automatic quantification of the following hair image features was conducted: local and global hair volumes and hair alignment. These features were assessed at three time points: t₀ (no treatment), t₁ (two treatments), t₂ (three treatments). Classifier tests were applied to test the accuracy of the machine learning. A sensory test (paired comparison of t₀ vs t₂ ) and an online front-image based survey (paired comparison of t₀ vs t₁ , t₁ vs t₂ , t₀ vs t₂ ) were conducted to compare human assessment with that of the algorithms.

RESULTS : The automatic image analysis identified changes to hair volume and alignment which enabled the successful application of the classification tests, especially when the hair images were grouped into untreated and treated groups. The human assessment of hair presented in pairs confirmed the automatic image analysis. The image assessment for both virgin hair and bleached only partially agreed with the analysis of the subset of images used in the online survey. One hypothesis is that treatments changed somewhat the shape of the hair tress, with the effect being more pronounced in bleached hair. This made human assessment of flat images more challenging than when viewed directly in 3D. Overall, the bleached hair exhibited effects of higher magnitude than the virgin hair.

CONCLUSIONS : This study illustrated the capacity of artificial intelligence for hair image detection and classification, and for image analysis of hair assembly features following treatments. The human assessment partially confirmed the image analysis, and highlighted the challenges imposed by the presentation mode.

Daniels Gabriela, Tamburic Slobodanka, Benini Sergio, Randall Jane, Sanderson Tracey, Savardi Mattia

2021-Apr-13

artificial intelligence, bleached hair, hair detection, hair segmentation, machine learning, sensory assessment, virgin hair

In Brain : a journal of neurology

Gross Catharina C, Schulte-Mecklenbeck Andreas, Madireddy Lohith, Pawlitzki Marc, Strippel Christine, Räuber Saskia, Krämer Julia, Rolfes Leoni, Ruck Tobias, Beuker Carolin, Schmidt-Pogoda Antje, Lohmann Lisa, Schneider-Hohendorf Tilman, Hahn Tim, Schwab Nicholas, Minnerup Jens, Melzer Nico, Klotz Luisa, Meuth Sven G, Hörste Gerd Meyer Zu, Baranzini Sergio E, Wiendl Heinz

2021-Apr-12

CNS autoimmunity, CSF, differential diagnosis, immune profile, multiple sclerosis

In IEEE journal of biomedical and health informatics

Feng Yi, Wang Xianglin, Zhang Juan

2021-Apr-13

In IEEE journal of biomedical and health informatics

Kwon Hyun Bin, Choi Sang Ho, Lee Dongseok, Son Dongyeon, Yoon Heenam, Lee Mi Hyun, Lee Yu Jin, Park Kwang Suk

2021-Apr-13

In IEEE transactions on medical imaging ; h5-index 74.0

Hoppe Elisabeth, Wetzl Jens, Yoon Seung Su, Bacher Mario, Roser Philipp, Stimpel Bernhard, Preuhs Alexander, Maier Andreas

2021-Apr-13

In Seminars in cardiothoracic and vascular anesthesia ; h5-index 16.0

Wang Ryan F, Fagelman Erica J, Smith Natalie K, Sakai Tetsuro

2021-Apr-13

COVID-19, anesthesiology, intestine, kidney, liver, pancreas, transplantation

In NPJ digital medicine

Esteva Andre, Kale Anuprit, Paulus Romain, Hashimoto Kazuma, Yin Wenpeng, Radev Dragomir, Socher Richard

2021-Apr-12

In Statistical methods in medical research

Austin Peter C, Harrell Frank E, Steyerberg Ewout W

2021-Apr-13

Machine learning, Monte Carlo simulations, data-generating process, generalized boosting methods, logistic regression, random forests

In Advances in experimental medicine and biology

Yan Qi, Ding Ying, Weeks Daniel E, Chen Wei

2021

AMD genetics, GWAS, Machine learning, Prediction, Progression, Statistical methods

In Internal and emergency medicine ; h5-index 30.0

Gamboa-Antiñolo Fernando-Miguel

2021-Apr-13

In Journal of medical systems ; h5-index 48.0

Peng Junfeng, Zou Kaiqiang, Zhou Mi, Teng Yi, Zhu Xiongyong, Zhang Feifei, Xu Jun

2021-Apr-13

Hepatitis, Local Interpretable Model-agnostic Explanations, Model interpretation, Partial Dependence Plots, SHapley Additive exPlanations

In European radiology ; h5-index 62.0

OBJECTIVES : The interpretability of convolutional neural networks (CNNs) for classifying subsolid nodules (SSNs) is insufficient for clinicians. Our purpose was to develop CNN models to classify SSNs on CT images and to investigate image features associated with the CNN classification.

METHODS : CT images containing SSNs with a diameter of ≤ 3 cm were retrospectively collected. We trained and validated CNNs by a 5-fold cross-validation method for classifying SSNs into three categories (benign and preinvasive lesions [PL], minimally invasive adenocarcinoma [MIA], and invasive adenocarcinoma [IA]) that were histologically confirmed or followed up for 6.4 years. The mechanism of CNNs on human-recognizable CT image features was investigated and visualized by gradient-weighted class activation map (Grad-CAM), separated activation channels and areas, and DeepDream algorithm.

RESULTS : The accuracy was 93% for classifying 586 SSNs from 569 patients into three categories (346 benign and PL, 144 MIA, and 96 IA in 5-fold cross-validation). The Grad-CAM successfully located the entire region of image features that determined the final classification. Activated areas in the benign and PL group were primarily smooth margins (p < 0.001) and ground-glass components (p = 0.033), whereas in the IA group, the activated areas were mainly part-solid (p < 0.001) and solid components (p < 0.001), lobulated shapes (p < 0.001), and air bronchograms (p < 0.001). However, the activated areas for MIA were variable. The DeepDream algorithm showed the image features in a human-recognizable pattern that the CNN learned from a training dataset.

CONCLUSION : This study provides medical evidence to interpret the mechanism of CNNs that helps support the clinical application of artificial intelligence.

KEY POINTS : • CNN achieved high accuracy (93%) in classifying subsolid nodules on CT images into three categories: benign and preinvasive lesions, MIA, and IA. • The gradient-weighted class activation map (Grad-CAM) located the entire region of image features that determined the final classification, and the visualization of the separated activated areas was consistent with radiologists' expertise for diagnosing subsolid nodules. • DeepDream showed the image features that CNN learned from a training dataset in a human-recognizable pattern.

Jiang Beibei, Zhang Yaping, Zhang Lu, H de Bock Geertruida, Vliegenthart Rozemarijn, Xie Xueqian

2021-Apr-13

Adenocarcinoma of lung, Artificial intelligence, Deep learning, X-ray computed tomography

In European journal of nuclear medicine and molecular imaging ; h5-index 66.0

PURPOSE : The present scoping review aims to assess the non-inferiority of distributed learning over centrally and locally trained machine learning (ML) models in medical applications.

METHODS : We performed a literature search using the term "distributed learning" OR "federated learning" in the PubMed/MEDLINE and EMBASE databases. No start date limit was used, and the search was extended until July 21, 2020. We excluded articles outside the field of interest; guidelines or expert opinion, review articles and meta-analyses, editorials, letters or commentaries, and conference abstracts; articles not in the English language; and studies not using medical data. Selected studies were classified and analysed according to their aim(s).

RESULTS : We included 26 papers aimed at predicting one or more outcomes: namely risk, diagnosis, prognosis, and treatment side effect/adverse drug reaction. Distributed learning was compared to centralized or localized training in 21/26 and 14/26 selected papers, respectively. Regardless of the aim, the type of input, the method, and the classifier, distributed learning performed close to centralized training, but two experiments focused on diagnosis. In all but 2 cases, distributed learning outperformed locally trained models.

CONCLUSION : Distributed learning resulted in a reliable strategy for model development; indeed, it performed equally to models trained on centralized datasets. Sensitive data can get preserved since they are not shared for model development. Distributed learning constitutes a promising solution for ML-based research and practice since large, diverse datasets are crucial for success.

Kirienko Margarita, Sollini Martina, Ninatti Gaia, Loiacono Daniele, Giacomello Edoardo, Gozzi Noemi, Amigoni Francesco, Mainardi Luca, Lanzi Pier Luca, Chiti Arturo

2021-Apr-13

Clinical trial, Distributed learning, Ethics, Federated learning, Machine learning, Privacy

In G3 (Bethesda, Md.)

Kismiantini Montesinos-López, Osval Antonio Crossa, José Setiawan, Ezra Putranda Wutsqa

2021-Feb

count data, generalized poisson regression, genomic data, genomic selection, high-resolution images, plant breeding

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : Although the potential of big data analytics for health care is well recognized, evidence is lacking on its effects on public health.

OBJECTIVE : The aim of this study was to assess the impact of the use of big data analytics on people's health based on the health indicators and core priorities in the World Health Organization (WHO) General Programme of Work 2019/2023 and the European Programme of Work (EPW), approved and adopted by its Member States, in addition to SARS-CoV-2-related studies. Furthermore, we sought to identify the most relevant challenges and opportunities of these tools with respect to people's health.

METHODS : Six databases (MEDLINE, Embase, Cochrane Database of Systematic Reviews via Cochrane Library, Web of Science, Scopus, and Epistemonikos) were searched from the inception date to September 21, 2020. Systematic reviews assessing the effects of big data analytics on health indicators were included. Two authors independently performed screening, selection, data extraction, and quality assessment using the AMSTAR-2 (A Measurement Tool to Assess Systematic Reviews 2) checklist.

RESULTS : The literature search initially yielded 185 records, 35 of which met the inclusion criteria, involving more than 5,000,000 patients. Most of the included studies used patient data collected from electronic health records, hospital information systems, private patient databases, and imaging datasets, and involved the use of big data analytics for noncommunicable diseases. "Probability of dying from any of cardiovascular, cancer, diabetes or chronic renal disease" and "suicide mortality rate" were the most commonly assessed health indicators and core priorities within the WHO General Programme of Work 2019/2023 and the EPW 2020/2025. Big data analytics have shown moderate to high accuracy for the diagnosis and prediction of complications of diabetes mellitus as well as for the diagnosis and classification of mental disorders; prediction of suicide attempts and behaviors; and the diagnosis, treatment, and prediction of important clinical outcomes of several chronic diseases. Confidence in the results was rated as "critically low" for 25 reviews, as "low" for 7 reviews, and as "moderate" for 3 reviews. The most frequently identified challenges were establishment of a well-designed and structured data source, and a secure, transparent, and standardized database for patient data.

CONCLUSIONS : Although the overall quality of included studies was limited, big data analytics has shown moderate to high accuracy for the diagnosis of certain diseases, improvement in managing chronic diseases, and support for prompt and real-time analyses of large sets of varied input data to diagnose and predict disease outcomes.

TRIAL REGISTRATION : International Prospective Register of Systematic Reviews (PROSPERO) CRD42020214048; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=214048.

Borges do Nascimento Israel Júnior, Marcolino Milena Soriano, Abdulazeem Hebatullah Mohamed, Weerasekara Ishanka, Azzopardi-Muscat Natasha, Gonçalves Marcos André, Novillo-Ortiz David

2021-Apr-13

World Health Organization, big data, big data analytics, evidence-based medicine, health status, machine learning, overview, public health, secondary data analysis, systematic review

In Wiley interdisciplinary reviews. Nanomedicine and nanobiotechnology

Teunissen Abraham J P, Burnett Marianne E, Prévot Geoffrey, Klein Emma D, Bivona Daniel, Mulder Willem J M

2021-Apr-13

innate immunotherapy, nanomaterials, nanomedicine, nanotherapeutics

In The Journal of nutrition ; h5-index 61.0

BACKGROUND : Donor milk is the standard of care for hospitalized very low birth weight (VLBW) infants when mother's milk is unavailable; however, growth of donor milk-fed infants is frequently suboptimal. Variability in nutrient composition of donated milk complicates the production of a uniform pooled product and, subsequently, the provision of adequate nutrition to promote optimal growth and development of VLBW infants. We reasoned a machine learning approach to construct batches using characteristics of the milk donation might be an effective strategy in reducing the variability in donor milk product composition.

OBJECTIVE : The objective of this study was to identify whether machine learning models can accurately predict donor milk macronutrient content. We focused on predicting fat and protein, given their well-established importance in VLBW infant growth outcomes.

METHODS : Samples of donor milk, consisting of 272 individual donations and 61 pool samples, were collected from the Rogers Hixon Ontario Human Milk Bank and analyzed for macronutrient content. Four different machine learning models were constructed using independent variable groups associated with donations, donors, and donor-pumping practices. A baseline model was established using lactation stage and infant gestational status. Predictions were made for individual donations and resultant pools.

RESULTS : Machine learning models predicted protein of individual donations and pools with a mean absolute error (MAE) of 0.16 g/dL and 0.10 g/dL, respectively. Individual donation and pooled fat predictions had an MAE of 0.91 g/dL and 0.42 g/dL, respectively. At both the individual donation and pool levels, protein predictions were significantly more accurate than baseline, whereas fat predictions were competitive with baseline.

CONCLUSIONS : Machine learning models can provide accurate predictions of macronutrient content in donor milk. The macronutrient content of pooled milk had a lower prediction error, reinforcing the value of pooling practices. Future research should examine how macronutrient content predictions can be used to facilitate milk bank pooling strategies.

Wong Rachel K, Pitino Michael A, Mahmood Rafid, Zhu Ian Yihang, Stone Debbie, O’Connor Deborah L, Unger Sharon, Chan Timothy C Y

2021-Apr-13

donor human milk, human milk banking, machine learning, macronutrient analysis, macronutrient prediction

In Circulation ; h5-index 165.0

Segar Matthew W, Jaeger Byron C, Patel Kershaw V, Nambi Vijay, Ndumele Chiadi E, Correa Adolfo, Butler Javed, Chandra Alvin, Ayers Colby, Rao Shreya, Lewis Alana A, Raffield Laura M, Rodriguez Carlos J, Michos Erin D, Ballantyne Christie M, Hall Michael E, Mentz Robert J, de Lemos James A, Pandey Ambarish

2021-Apr-13

In European journal of radiology ; h5-index 47.0

PURPOSE : As of August 30th, there were in total 25.1 million confirmed cases and 845 thousand deaths caused by coronavirus disease of 2019 (COVID-19) worldwide. With overwhelming demands on medical resources, patient stratification based on their risks is essential. In this multi-center study, we built prognosis models to predict severity outcomes, combining patients' electronic health records (EHR), which included vital signs and laboratory data, with deep learning- and CT-based severity prediction.

METHOD : We first developed a CT segmentation network using datasets from multiple institutions worldwide. Two biomarkers were extracted from the CT images: total opacity ratio (TOR) and consolidation ratio (CR). After obtaining TOR and CR, further prognosis analysis was conducted on datasets from INSTITUTE-1, INSTITUTE-2 and INSTITUTE-3. For each data cohort, generalized linear model (GLM) was applied for prognosis prediction.

RESULTS : For the deep learning model, the correlation coefficient of the network prediction and manual segmentation was 0.755, 0.919, and 0.824 for the three cohorts, respectively. The AUC (95 % CI) of the final prognosis models was 0.85(0.77,0.92), 0.93(0.87,0.98), and 0.86(0.75,0.94) for INSTITUTE-1, INSTITUTE-2 and INSTITUTE-3 cohorts, respectively. Either TOR or CR exist in all three final prognosis models. Age, white blood cell (WBC), and platelet (PLT) were chosen predictors in two cohorts. Oxygen saturation (SpO2) was a chosen predictor in one cohort.

CONCLUSION : The developed deep learning method can segment lung infection regions. Prognosis results indicated that age, SpO2, CT biomarkers, PLT, and WBC were the most important prognostic predictors of COVID-19 in our prognosis model.

Gong Kuang, Wu Dufan, Arru Chiara Daniela, Homayounieh Fatemeh, Neumark Nir, Guan Jiahui, Buch Varun, Kim Kyungsang, Bizzo Bernardo Canedo, Ren Hui, Tak Won Young, Park Soo Young, Lee Yu Rim, Kang Min Kyu, Park Jung Gil, Carriero Alessandro, Saba Luca, Masjedi Mahsa, Talari Hamidreza, Babaei Rosa, Mobin Hadi Karimi, Ebrahimian Shadi, Guo Ning, Digumarthy Subba R, Dayan Ittai, Kalra Mannudeep K, Li Quanzheng

2021-Feb-05

COVID-19, Computed tomography, Deep learning, Electronic health records, Prognosis

ArXiv Preprint

Vikas Raunak, Arul Menezes, Marcin Junczys-Dowmunt

2021-04-14

In Journal of endourology

Houshyar Roozbeh, Glavis-Bloom Justin, Bui Thanh-Lan, Chahine Chantal, Bardis Michelle D, Ushinsky Alexander, Liu Hanna, Bhatter Param, Lebby Elliott, Fujimoto Dylann, Grant William, Tran-Harding Karen, Landman Jaime, Chow Daniel S, Chang Peter D

2021-Apr-13

In PsyCh journal

He Qiao, Wang Qiandong, Wu Yaxue, Yi Li, Wei Kunlin

2021-Apr-12

autism diagnosis, autism screening, autism spectrum disorder, automatic classification, machine learning, visual orienting

In Angewandte Chemie (International ed. in English)

Domulevicz Lucas, Jeong Hyunhak, Paul Nayan K, Gomez-Diaz Juan Sebastian, Hihath Joshua

2021-Apr-12

Density Functional Calculations, Molecular electronics, Picocavity, Raman spectroscopy, Single-molecule studies

In Environmental science and pollution research international

Kim Kisok, Park Hyejin

2021-Apr-12

KNHNES, Lead exposure, Machine-learning, Osteoarthritis, Post-menopausal women, Predictive model

In Graefe's archive for clinical and experimental ophthalmology = Albrecht von Graefes Archiv fur klinische und experimentelle Ophthalmologie

PURPOSE : To detect the leakage points of central serous chorioretinopathy (CSC) automatically from dynamic images of fundus fluorescein angiography (FFA) using a deep learning algorithm (DLA).

METHODS : The study included 2104 FFA images from 291 FFA sequences of 291 eyes (137 right eyes and 154 left eyes) from 262 patients. The leakage points were segmented with an attention gated network (AGN). The optic disk (OD) and macula region were segmented simultaneously using a U-net. To reduce the number of false positives based on time sequence, the leakage points were matched according to their positions in relation to the OD and macula.

RESULTS : With the AGN alone, the number of cases whose detection results perfectly matched the ground truth was only 37 out of 61 cases (60.7%) in the test set. The dice on the lesion level were 0.811. Using an elimination procedure to remove false positives, the number of accurate detection cases increased to 57 (93.4%). The dice on the lesion level also improved to 0.949.

CONCLUSIONS : Using DLA, the CSC leakage points in FFA can be identified reproducibly and accurately with a good match to the ground truth. This novel finding may pave the way for potential application of artificial intelligence to guide laser therapy.

Chen Menglu, Jin Kai, You Kun, Xu Yufeng, Wang Yao, Yip Chee-Chew, Wu Jian, Ye Juan

2021-Apr-12

Central serous chorioretinopathy, Deep learning, Fundus fluorescein angiography, Time sequence

In Chinese journal of academic radiology

Objective : This study aimed to compare quantifiable radiologic findings and their dynamic change throughout the clinical course of common and severe coronavirus disease 2019 (COVID-19), and to provide valuable evidence for radiologic classification of the two types of this disease.

Methods : 112 patients with laboratory-confirmed COVID-19 were retrospectively analyzed. Volumetric percentage of infection and density of the lung were measured by a computer-aided software. Clinical parameters were recorded to reflect disease progression. Baseline data and dynamic change were compared between two groups and a decision-tree algorithm was developed to determine the cut-off value for classification.

Results : 93 patients were finally included and were divided into common group (n = 76) and severe group (n = 17) based on current criteria. Compared with common patients, severe patients experienced shorter advanced stage, peak time and plateau, but longer absorption stage. The dynamic change of volume and density coincided with the clinical course. The interquartile range of volumetric percentage of the two groups were 1.0-7.2% and 11.4-31.2%, respectively. Baseline volumetric percentage of infection was significantly higher in severe group, and the cut-off value of it was 10.10%.

Conclusions : Volumetric percentage between severe and common patients was significantly different. Because serial CT scans are systemically performed in patients with COVID-19 pneumonia, this quantitative analysis can simultaneously provide valuable information for physicians to evaluate their clinical course and classify common and severe patients accurately.

Yang Chongtu, Cao Guijuan, Liu Fen, Liu Jiacheng, Huang Songjiang, Xiong Bin

2021-Apr-08

Artificial intelligence, Computer-assisted, Coronavirus disease 2019, Decision trees, Multidetector computed tomography, Numerical analysis

In Biomedical signal processing and control

Sharifrazi Danial, Alizadehsani Roohallah, Roshanzamir Mohamad, Joloudari Javad Hassannataj, Shoeibi Afshin, Jafari Mahboobeh, Hussain Sadiq, Sani Zahra Alizadeh, Hasanzadeh Fereshteh, Khozeimeh Fahime, Khosravi Abbas, Nahavandi Saeid, Panahiazar Maryam, Zare Assef, Islam Sheikh Mohammed Shariful, Acharya U Rajendra

2021-Apr-08

CNN., Covid-19, Data Mining, Deep Learning, Feature Extraction, Image Processing, Machine Learning, SVM, Sobel operator

In Information sciences

Ouyang Liwei, Yuan Yong, Cao Yumeng, Wang Fei-Yue

2021-Apr-08

blockchain, collaborative early warning, federated learning, learning markets, smart contracts

In Nature metabolism

Shamsi Farnaz, Piper Mary, Ho Li-Lun, Huang Tian Lian, Gupta Anushka, Streets Aaron, Lynes Matthew D, Tseng Yu-Hua

2021-Apr-12

In NPJ digital medicine

Esteva Andre, Kale Anuprit, Paulus Romain, Hashimoto Kazuma, Yin Wenpeng, Radev Dragomir, Socher Richard

2021-Apr-12

In F1000Research

Thomas Laurent S V, Schaefer Franz, Gehrig Jochen

2020

Fiji, ImageJ, KNIME, bioimage analysis, ground-truth labelling, image annotation, image classification, qualitative analysis

In Seminars in cardiothoracic and vascular anesthesia ; h5-index 16.0

Wang Ryan F, Fagelman Erica J, Smith Natalie K, Sakai Tetsuro

2021-Apr-13

COVID-19, anesthesiology, intestine, kidney, liver, pancreas, transplantation

ArXiv Preprint

Jessica Morley, Caroline Morton, Kassandra Karpathakis, Mariarosaria Taddeo, Luciano Floridi

2021-04-14

In Scientific reports ; h5-index 158.0

Shim Miseon, Lee Seung-Hwan, Hwang Han-Jeong

2021-Apr-12

In Scientific reports ; h5-index 158.0

Kim Jeoung Kun, Choo Yoo Jin, Shin Hyunkwang, Choi Gyu Sang, Chang Min Cheol

2021-Apr-12

In Scientific reports ; h5-index 158.0

Schouten Jens P E, Matek Christian, Jacobs Luuk F P, Buck Michèle C, Bošnački Dragan, Marr Carsten

2021-Apr-12

In Scientific reports ; h5-index 158.0

Jo Yong-Yeon, Choi Young Sang, Park Hyun Woo, Lee Jae Hyeok, Jung Hyojung, Kim Hyo-Eun, Ko Kyounglan, Lee Chan Wha, Cha Hyo Soung, Hwangbo Yul

2021-Apr-12

In Scientific reports ; h5-index 158.0

Andersson Sam, Bathula Deepti R, Iliadis Stavros I, Walter Martin, Skalkidou Alkistis

2021-Apr-12

In Scientific data

Ahmed Shahzad, Wang Dingyang, Park Junyoung, Cho Sung Ho

2021-Apr-12

In Scientific data

Tong Kangkang, Nagpure Ajay Singh, Ramaswami Anu

2021-Apr-12

In Nature communications ; h5-index 260.0

Feng Xiaohua, Gao Liang

2021-04-12

In European journal of radiology ; h5-index 47.0

PURPOSE : As of August 30th, there were in total 25.1 million confirmed cases and 845 thousand deaths caused by coronavirus disease of 2019 (COVID-19) worldwide. With overwhelming demands on medical resources, patient stratification based on their risks is essential. In this multi-center study, we built prognosis models to predict severity outcomes, combining patients' electronic health records (EHR), which included vital signs and laboratory data, with deep learning- and CT-based severity prediction.

METHOD : We first developed a CT segmentation network using datasets from multiple institutions worldwide. Two biomarkers were extracted from the CT images: total opacity ratio (TOR) and consolidation ratio (CR). After obtaining TOR and CR, further prognosis analysis was conducted on datasets from INSTITUTE-1, INSTITUTE-2 and INSTITUTE-3. For each data cohort, generalized linear model (GLM) was applied for prognosis prediction.

RESULTS : For the deep learning model, the correlation coefficient of the network prediction and manual segmentation was 0.755, 0.919, and 0.824 for the three cohorts, respectively. The AUC (95 % CI) of the final prognosis models was 0.85(0.77,0.92), 0.93(0.87,0.98), and 0.86(0.75,0.94) for INSTITUTE-1, INSTITUTE-2 and INSTITUTE-3 cohorts, respectively. Either TOR or CR exist in all three final prognosis models. Age, white blood cell (WBC), and platelet (PLT) were chosen predictors in two cohorts. Oxygen saturation (SpO2) was a chosen predictor in one cohort.

CONCLUSION : The developed deep learning method can segment lung infection regions. Prognosis results indicated that age, SpO2, CT biomarkers, PLT, and WBC were the most important prognostic predictors of COVID-19 in our prognosis model.

Gong Kuang, Wu Dufan, Arru Chiara Daniela, Homayounieh Fatemeh, Neumark Nir, Guan Jiahui, Buch Varun, Kim Kyungsang, Bizzo Bernardo Canedo, Ren Hui, Tak Won Young, Park Soo Young, Lee Yu Rim, Kang Min Kyu, Park Jung Gil, Carriero Alessandro, Saba Luca, Masjedi Mahsa, Talari Hamidreza, Babaei Rosa, Mobin Hadi Karimi, Ebrahimian Shadi, Guo Ning, Digumarthy Subba R, Dayan Ittai, Kalra Mannudeep K, Li Quanzheng

2021-Feb-05

COVID-19, Computed tomography, Deep learning, Electronic health records, Prognosis

In ISA transactions

Yang Chuangyan, Ma Jun, Wang Xiaodong, Li Xiang, Li Zhuorui, Luo Ting

2021-Mar-31

Feature fusion, GRNN, Health indicators, RUL prediction, Rolling bearing

In Current opinion in allergy and clinical immunology

PURPOSE OF REVIEW : To perform a nonsystematic review of the literature on the possible role of probiotics for food allergy (FA).

RECENT FINDINGS : Animal model and in vitro evidence suggest that the gut microbiome could protect against FA and that probiotics could be a valid instrument. There is no consistent evidence in identifying the specific species, the dosage, and the optimal duration to obtain the correct immunomodulation. Early life supplementation with specific 'missing' immunomodulatory microbes - derived from machine learning approach to birth cohort studies - might represent a novel approach to the primary prevention of multiple human atopic diseases. However, further studies are needed.

SUMMARY : Currently, there is no positive recommendation from the main scientific societies to use probiotics neither for the treatment nor for the prevention of FA.

Mennini Maurizio, Arasi Stefania, Artesani Maria Cristina, Fiocchi Alessandro Giovanni

2021-Mar-31

In Journal of computational social science

Caliskan Cantay

2021-Apr-05

Awareness, COVID-19, Emotion classification, Twitter

In Computers in biology and medicine

INTRODUCTION : We investigate the contribution of demographic, socio-economic, and geographic characteristics as determinants of physical health and well-being to guide public health policies and preventative behavior interventions (e.g., countering coronavirus).

METHODS : We use machine learning to build predictive models of overall well-being and physical health among veterans as a function of these three sets of characteristics. We link Gallup's U.S. Daily Poll between 2014 and 2017 over a range of demographic and socio-economic characteristics with zipcode characteristics from the Census Bureau to build predictive models of overall and physical well-being.

RESULTS : Although the predictive models of overall well-being have weak performance, our classification of low levels of physical well-being performed better. Gradient boosting delivered the best results (80.2% precision, 82.4% recall, and 80.4% AUROC) with perceptions of purpose in the workplace and financial anxiety as the most predictive features. Our results suggest that additional measures of socio-economic characteristics are required to better predict physical well-being, particularly among vulnerable groups, like veterans.

CONCLUSION : Socio-economic characteristics explain large differences in physical and overall well-being. Effective predictive models that incorporate socio-economic data will provide opportunities to create real-time and personalized feedback to help individuals improve their quality of life.

Makridis Christos A, Zhao David Y, Bejan Cosmin A, Alterovitz Gil

2021-Apr-03

Health informatics, Machine learning, Socioeconomics, Subjective well-being, Veterans

In Computers in biology and medicine

Sauvat Allan, Cerrato Giulia, Humeau Juliette, Leduc Marion, Kepp Oliver, Kroemer Guido

2021-Apr-04

Cellular imaging, Convolutional neural network, Immunogenic cell death, Semantic segmentation, Systems biology, Transmitted light microscopy, UNET

In Contemporary clinical trials

Sverdlov Oleksandr, Ryeznik Yevgen, Wong Weng Kee

2021-Apr-09

Adaptive designs, Cardiovascular trials, Machine learning, Master protocols, Nature-inspired metaheuristics, Rare diseases, Seamless designs

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for multiple evanescent white dot syndrome (MEWDS).

DESIGN : Machine learning of cases with MEWDS and 8 other posterior uveitides.

METHODS : Cases of posterior uveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the infectious posterior/panuveitides. The resulting criteria were evaluated on the validation set.

RESULTS : One thousand sixty-eight cases of posterior uveitides, including 51 cases of MEWDS, were evaluated by machine learning. Key criteria for MEWDS included: 1) multifocal gray white chorioretinal spots with foveal granularity; 2) characteristic imaging on fluorescein angiography ("wreath-like" hyperfluorescent lesions) and/or optical coherence tomography (hyper-reflective lesions extending from retinal pigment epithelium through ellipsoid zone into the retinal outer nuclear layer); and 3) absent to mild anterior chamber and vitreous inflammation. Overall accuracy for posterior uveitides was 93.9% in the training set and 98.0% (95% confidence interval 94.3, 99.3) in the validation set. The misclassification rates for MEWDS were 7% in the training set and 0% in the validation set.

CONCLUSIONS : The criteria for MEWDS had a low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

Jabs Douglas A, Brezin Antoine P, Dick Andrew D, Levinson Ralph D, Lim Lyndell L, McCluskey Peter, Oden Neal, Palestine Alan G, Thorne Jennifer E, Trusko Brett E, Vitale Albert, Wittenberg Susan E

2021-Apr-09

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for acute posterior multifocal placoid pigment epitheliopathy (APMPPE).

DESIGN : Machine learning of cases with APMPPE and 8 other posterior uveitides.

METHODS : Cases of posterior uveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the infectious posterior/panuveitides. The resulting criteria were evaluated on the validation set.

RESULTS : One thousand sixty-eight cases of posterior uveitides, including 82 cases of APMPPE, were evaluated by machine learning. Key criteria for APMPPE included: 1) choroidal lesions with a plaque-like or placoid appearance and 2) characteristic imaging on fluorescein angiography (lesions "block early and stain late diffusely"). Overall accuracy for posterior uveitides was 92.7% in the training set and 98.0% (95% confidence interval 94.3, 99.3) in the validation set. The misclassification rates for APMPPE were 5% in the training set and 0% in the validation set.

CONCLUSIONS : The criteria for APMPPE had a low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

The Standardization Of Uveitis Nomenclature Sun Working Group Jabs, Douglas A Brezin, Antoine P Dick, Andrew D Levinson, Ralph D Lim, Lyndell L McCluskey, Peter Oden, Neal Palestine, Alan G Thorne, Jennifer E Trusko, Brett E Vitale, Albert Wittenberg

2021-Apr-09

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for tubulointerstitial nephritis with uveitis (TINU) DESIGN: Machine learning of cases with TINU and 8 other anterior uveitides.

METHODS : Cases of anterior uveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on the diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the anterior uveitides. The resulting criteria were evaluated on the validation set.

RESULTS : One thousand eighty-three cases of anterior uveitides, including 94 cases of TINU, were evaluated by machine learning. The overall accuracy for anterior uveitides was 97.5% in the training set and 96.7% in the validation set (95% confidence interval 92.4, 98.6). Key criteria for TINU included anterior chamber inflammation and evidence of tubulointerstitial nephritis with either 1) a positive renal biopsy or 2) evidence of nephritis (elevated serum creatinine and/or abnormal urine analysis) and an elevated urine β-2 microglobulin. The misclassification rates for TINU were 1.2% in the training set and 0% in the validation set, respectively.

CONCLUSIONS : The criteria for TINU had a low misclassification rate and appeared to perform well enough for use in clinical and translational research.

Jabs Douglas A, Denniston Alastair K, Galor Anat, Lightman Susan, McCluskey Peter, Oden Neal, Palestine Alan G, Rosenbaum James T, Saleem Sophia M, Thorne Jennifer E, Trusko Brett E

2021-Apr-09

TINU-associated anterior uveitis

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for multiple sclerosis-associated intermediate uveitis.

DESIGN : Machine learning of cases with multiple sclerosis-associated intermediate uveitis and 4 other intermediate uveitides.

METHODS : Cases of intermediate uveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on the diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the intermediate uveitides. The resulting criteria were evaluated on the validation set.

RESULTS : Five hundred eighty-nine of cases of intermediate uveitides, including 112 cases of multiple sclerosis-associated intermediate uveitis, were evaluated by machine learning. The overall accuracy for intermediate uveitides was 99.8% in the training set and 99.3% in the validation set (95% confidence interval 96.1, 99.9). Key criteria for multiple sclerosis-associated intermediate uveitis included unilateral or bilateral intermediate uveitis and a diagnosis of multiple sclerosis by the McDonald Criteria. Key exclusions included syphilis and sarcoidosis. The misclassification rates for multiple sclerosis-associated intermediate uveitis were 0 % in the training set and 0% in the validation set, respectively.

CONCLUSIONS : The criteria for multiple sclerosis-associated intermediate uveitis had a low misclassification rate and appeared to perform sufficiently well enough for use in clinical and translational research.

**

2021-Apr-09

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for juvenile idiopathic arthritis (JIA)-associated chronic anterior uveitis (CAU).

DESIGN : Machine learning of cases with JIA CAU and 8 other anterior uveitides.

METHODS : Cases of anterior uveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on the diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the anterior uveitides. The resulting criteria were evaluated on the validation set.

RESULTS : One thousand eighty-three cases of anterior uveitides, including 202 cases of JIA CAU, were evaluated by machine learning. The overall accuracy for anterior uveitides was 97.5% in the training set and 96.7% in the validation set (95% confidence interval 92.4, 98.6). Key criteria for JIA CAU included 1) chronic anterior uveitis (or if newly diagnosed insidious onset) and 2) JIA, except for the systemic, rheumatoid factor-positive polyarthritis, and enthesitis related arthritis variants. The misclassification rates for JIA CAU were 2.4% in the training set and 0% in the validation set, respectively.

CONCLUSIONS : The criteria for JIA CAU had a low misclassification rate and appeared to perform well enough for use in clinical and translational research.

Jabs Douglas A, Acharya Nisha R, Chee Soon Phaik, Goldstein Debra, McCluskey Peter, Murray Philip I, Oden Neal, Palestine Alan G, Rosenbaum James T, Thorne Jennifer E, Trusko Brett E

2021-Apr-09

JIA-associated chronic anterior uveitis

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for syphilitic uveitis DESIGN: Machine learning of cases with syphilitic uveitis and 24 other uveitides.

METHODS : Cases of anterior, intermediate, posterior, and panuveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on the diagnosis, using formal consensus techniques. Cases were analyzed by anatomic class, and each class was split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the different uveitic classes. The resulting criteria were evaluated on the validation set.

RESULTS : Two hundred twenty-two cases of syphilitic uveitis were evaluated by machine learning with cases evaluated against other uveitides in the relevant uveitic class. Key criteria for syphilitic uveitis included a compatible uveitic presentation, (1) anterior uveitis, 2) intermediate uveitis, or 3) posterior or panuveitis with retinal, retinal pigment epithelial, or retinal vascular inflammation) and evidence of syphilis infection with a positive treponemal test. The Centers for Disease Control and Prevention reverse screening algorithm for syphilis testing is recommended. The misclassification rates for syphilitic uveitis in the training sets were: anterior uveitides 0%, intermediate uveitides 6.0%, posterior uveitides 0%, panuveitides 0%, and infectious posterior/panuveitides 8.6%. The overall accuracy of the diagnosis of syphilitic uveitis in the validation set was 100% (99% CI 99.5, 100) - i.e. the validation sets misclassification rates were 0% for each uveitic class.

CONCLUSIONS : The criteria for syphilitic uveitis had a low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

**

2021-Apr-09

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for cytomegalovirus (CMV) anterior uveitis DESIGN: : Machine learning of cases with CMV anterior uveitis and 8 other anterior uveitides.

METHODS : Cases of anterior uveitides were collected in an informatics-designed preliminary database, and a final datafubase was constructed of cases achieving supermajority agreement on the diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the anterior uveitides. The resulting criteria were evaluated on the validation set.

RESULTS : One thousand eighty-three of cases of anterior uveitides, including 89 cases of CMV anterior uveitis, were evaluated by machine learning. The overall accuracy for anterior uveitides was 97.5% in the training set and 96.7% in the validation set (95% confidence interval 92.4, 98.6). Key criteria for CMV anterior uveitis included unilateral anterior uveitis with a positive aqueous humor polymerase chain reaction assay for CMV. No clinical features reliably diagnosed CMV anterior uveitis. The misclassification rates for CMV anterior uveitis were 1.3 % in the training set and 0% in the validation set, respectively.

CONCLUSIONS : The criteria for CMV anterior uveitis had a low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

Jabs Douglas A, Caspers Laure, Chee Soon-Phaik, Galor Anat, Goldstein Debra, McCluskey Peter, Murray Philip I, Oden Neal, Palestine Alan G, Rosenbaum James T, Thorne Jennifer E, Trusko Brett E

2021-Apr-09

In EuroIntervention : journal of EuroPCR in collaboration with the Working Group on Interventional Cardiology of the European Society of Cardiology

BACKGROUND : Delayed or misdiagnosis of acute myocardial infarction (AMI) is not unusual in the daily practice. Since 12- lead electrocardiogram (ECG) is crucial for the detection of AMI, the systematic algorithm to strengthen ECG interpretation may have important implications for improving diagnosis.

AIMS : We aimed to develop a deep learning model (DLM) as a diagnostic support tool based on a 12-lead electrocardiogram.

METHODS : This retrospective cohort study included 1,051/697 ECGs from 737/287 coronary angiogram (CAG)-validated STEMI/NSTEMI patients and 140,336 ECGs from 76,775 not-AMI patients at the emergency department. The DLM was trained and validated by 80% and 20% of these ECGs. A human-machine competition was conducted. The area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were used to evaluate the performance of the DLM.

RESULTS : The AUC of the DLM for STEMI detection was 0.976 in the human-machine competition, which was significantly better than that of the best physicians. Furthermore, the DLM independently demonstrated sufficient diagnostic capacity for STEMI detection (AUC=0.997; sensitivity, 98.4%; specificity, 96.9%). Regarding NSTEMI detection, the AUC of the combined DLM and conventional cardiac troponin I (cTnI) increased to 0.978, which was better than that of either the DLM (0.877) or cTnI (0.950).

CONCLUSIONS : The DLM may serve as a timely, objective and precise diagnostic decision support tool to assist emergency medical system-based networks and frontline physicians in detecting AMI and subsequently initiating reperfusion therapy.

Liu Wen-Cheng, Lin Chin-Sheng, Tsai Chien-Sung, Tsao Tien-Ping, Cheng Cheng-Chung, Liou Jun-Ting, Lin Wei-Shiang, Cheng Shu-Meng, Lou Yu-Sheng, Lee Chia-Cheng, Lin Chin

2021-Apr-13

In Informatics in medicine unlocked

Ghany Kareem Kamal A, Zawbaa Hossam M, Sabri Heba M

2021-Apr-06

Artificial Intelligence, COVID-19, Deep Learning, LSTM, Prediction

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for Vogt-Koyanagi-Harada (VKH) disease DESIGN: Machine learning of cases with VKH disease and 5 other panuveitides.

METHODS : Cases of panuveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on the diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the panuveitides. The resulting criteria were evaluated on the validation set.

RESULTS : One thousand twelve cases of panuveitides, including 156 cases of early-stage VKH and 103 cases of late-stage VKH, were evaluated. Overall accuracy for panuveitides was 96.3% in the training set and 94.0% in the validation set (95% confidence interval 89.0, 96.8). Key criteria for early-stage VKH included: 1) exudative retinal detachment with characteristic appearance on fluorescein angiogram or optical coherence tomography or 2) panuveitis with ≥2 of 5 neurologic symptoms/signs. Key criteria for late-stage VKH included history of early-stage VKH and either: 1) sunset glow fundus or 2) uveitis and ≥1 of 3 cutaneous signs. The misclassification rates in the learning and validation sets for early-stage VKH were 8.0% and 7.7%, respectively, and for late-stage VKH 1.0% and 12%, respectively.

CONCLUSIONS : The criteria for VKH had a reasonably low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

Jabs Douglas A, Denniston Alastair K, Dick Andrew D, Dunn James P, Kramer Michal, McCluskey Peter, Oden Neal, Okada Annabelle A, Palestine Alan G, Read Russell W, Thorne Jennifer E, Trusko Brett E

2021-Apr-09

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for multifocal choroiditis with panuveitis (MFCPU) DESIGN: : Machine learning of cases with MFCPU and 8 other posterior uveitides.

METHODS : Cases of posterior uveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the posterior uveitides. The resulting criteria were evaluated on the validation set.

RESULTS : One thousand sixty-eight cases of posterior uveitides, including 138 cases of MFCPU, were evaluated by machine learning. Key criteria for MFCPU included: 1) multifocal choroiditis with the predominant lesions size >125 µm in diameter; 2) lesions outside the posterior pole (with or without posterior involvement); and either 3) punched-out atrophic chorioretinal scars or 4) more than minimal mild anterior chamber and/or vitreous inflammation. Overall accuracy for posterior uveitides was 93.9% in the training set and 98.0% (95% confidence interval 94.3, 99.3) in the validation set. The misclassification rates for MFCPU were 15% in the training set and 0% in the validation set.

CONCLUSIONS : The criteria for MFCPU had a reasonably low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

Jabs Douglas A, Brezin Antoine P, Levinson Ralph D, McCluskey Peter, Oden Neal, Palestine Alan G, Read Russell W, Thorne Jennifer E, Trusko Brett E, Vitale Albert, Wittenberg Susan E

2021-Apr-09

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for cytomegalovirus (CMV) retinitis.

DESIGN : Machine learning of cases with CMV retinitis and 4 other infectious posterior/ panuveitides.

METHODS : Cases of infectious posterior/panuveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the infectious posterior/panuveitides. The resulting criteria were evaluated on the validation set.

RESULTS : Eight hundred three cases of infectious posterior/panuveitides, including 211 cases of CMV retinitis, were evaluated by machine learning. Key criteria for CMV retinitis included: 1) necrotizing retinitis with indistinct borders due to numerous small satellites; 2) evidence of immune compromise; and either 3) a characteristic clinical appearance or 4) positive polymerase chain assay for CMV from an intraocular specimen. Characteristic appearances for CMV retinitis included: 1) wedge-shaped area of retinitis; 2) hemorrhagic retinitis; or 3) granular retinitis. Overall accuracy for infectious posterior/panuveitides was 92.1% in the training set and 93.3% (95% confidence interval 88.2, 96.3) in the validation set. The misclassification rates for CMV retinitis were 6.9% in the training set and 6.3% in the validation set.

CONCLUSIONS : The criteria for CMV retinitis had a low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

Jabs Douglas A, Belfort Rubens, Bodaghi Bahram, Graham Elizabeth, Holland Gary N, Lightman Susan L, Oden Neal, Palestine Alan G, Smith Justine R, Thorne Jennifer E, Trusko Brett E, Van Gelder Russell N

2021-Apr-09

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for tubercular uveitis DESIGN: Machine learning of cases with tubercular uveitis and 14 other uveitides.

METHODS : Cases of non-infectious posterior or panuveitis, and of infectious posterior or panuveitis were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on the diagnosis, using formal consensus techniques. Cases were analyzed by anatomic class, and each class was split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the intermediate uveitides. The resulting criteria were evaluated on the validation sets.

RESULTS : Two hundred seventy-seven cases of tubercular uveitis were evaluated by machine learning against other uveitides. Key criteria for tubercular uveitis were a compatible uveitic syndrome, including: 1) anterior uveitis with iris nodules, 2) serpiginous-like tubercular choroiditis, 3) choroidal nodule (tuberculoma), 4) occlusive retinal vasculitis, and 5) in hosts with evidence of active systemic tuberculosis, multifocal choroiditis; and evidence of tuberculosis, including: 1) histologically- or microbiologically-confirmed infection, 2) positive interferon-Ɣ release assay test, or 3) positive tuberculin skin test. The overall accuracy of the diagnosis of tubercular uveitis versus other uveitides in the validation set was 98.2% (95% CI 96.5, 99.1). The misclassification rates for tubercular uveitis were: training set 3.4%; and validation set 3.6%.

CONCLUSIONS : The criteria for tubercular uveitis had a low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

The Standardization Of Uveitis Nomenclature Sun Working Group Jabs, Douglas A Belfort, Rubens Bodaghi, Bahram Graham, Elizabeth Gupta, Vishali Holland, Gary N Lightman, Susan L Oden, Neal Palestine, Alan G Smith, Justine R Thorne, Jennifer E Trusko

2021-Apr-09

Tubercular uveitis

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for serpiginous choroiditis.

DESIGN : Machine learning of cases with serpiginous choroiditis and 8 other posterior uveitides.

METHODS : Cases of posterior uveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the infectious posterior/panuveitides. The resulting criteria were evaluated on the validation set.

RESULTS : One thousand sixty-eight cases of posterior uveitides, including 122 cases of serpiginous choroiditis, were evaluated by machine learning. Key criteria for serpiginous choroiditis included: 1) choroiditis with an ameboid or serpentine shape; 2) characteristic imaging on fluorescein angiography or fundus autofluorescence; 3) absent to mild anterior chamber and vitreous inflammation; and 4) the exclusion of tuberculosis. Overall accuracy for posterior uveitides was 93.9% in the training set and 98.0% (95% confidence interval 94.3, 99.3) in the validation set. The misclassification rates for serpiginous choroiditis were 0% in both the training set and the validation set.

CONCLUSIONS : The criteria for serpiginous choroiditis had a low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

Jabs Douglas A, Brezin Antoine P, Levinson Ralph D, Oden Neal, Palestine Alan G, Rao Narsing A, Thorne Jennifer E, Trusko Brett E, Vitale Albert, Wittenberg Susan E

2021-Apr-09

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for acute retinal necrosis (ARN).

DESIGN : Machine learning of cases with ARN and 4 other infectious posterior/ panuveitides.

METHODS : Cases of infectious posterior/panuveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the infectious posterior/panuveitides. The resulting criteria were evaluated on the validation set.

RESULTS : Eight hundred three cases of infectious posterior/panuveitides, including 186 cases of ARN, were evaluated by machine learning. Key criteria for ARN included: 1) peripheral necrotizing retinitis; and either 2) polymerase chain reaction assay of an intraocular fluid specimen positive for either herpes simplex virus or varicella zoster virus; or 3) a characteristic clinical appearance with circumferential or confluent retinitis, retinal vascular sheathing and/or occlusion, and more than minimal vitritis. Overall accuracy for infectious posterior/panuveitides was 92.1% in the training set and 93.3% (95% confidence interval 88.2, 96.3) in the validation set. The misclassification rates for ARN were 15% in the training set and 11.5% in the validation set.

CONCLUSIONS : The criteria for ARN had a reasonably low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

Jabs Douglas A, Belfort Rubens, Bodaghi Bahram, Graham Elizabeth, Holland Gary N, Lightman Susan L, Margolis Todd P, Oden Neal, Palestine Alan, Smith Justine R, Thorne Jennifer E, Trusko Brett E, Van Gelder Russell N

2021-Apr-09

Acute retinal necrosis

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for punctate inner choroiditis (PIC).

DESIGN : Machine learning of cases with PIC and 8 other posterior uveitides.

METHODS : Cases of posterior uveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the posterior uveitides. The resulting criteria were evaluated on the validation set.

RESULTS : One thousand sixty-eight cases of posterior uveitides, including 144 cases of PIC, were evaluated by machine learning. Key criteria for PIC included: 1) "punctate" appearing choroidal spots <250 µm in diameter; 2) absent to minimal anterior chamber and vitreous inflammation; and 3) involvement of the posterior pole with or without mid-periphery. Overall accuracy for posterior uveitides was 93.9% in the training set and 98.0% (95% confidence interval 94.3, 99.3) in the validation set. The misclassification rates for PIC were 15% in the training set and 9% in the validation set.

CONCLUSIONS : The criteria for PIC had a reasonably low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

The Standardization Of Uveitis Nomenclature Sun Working Group Jabs, Douglas A Brezin, Antoine P Levinson, Ralph D Lightman, Susan L McCluskey, Peter Oden, Neal Palestine, Alan G Rao, Narsing A Thorne, Jennifer E Trusko, Brett E Vitale, Albert Wittenberg

2021-Apr-09

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for varicella zoster virus (VZV) anterior uveitis DESIGN: Machine learning of cases with VZV anterior uveitis and 8 other anterior uveitides.

METHODS : Cases of anterior uveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on the diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the anterior uveitides. The resulting criteria were evaluated on the validation set.

RESULTS : One thousand eighty-three cases of anterior uveitides, including 123 cases of VZV anterior uveitis, were evaluated by machine learning. The overall accuracy for anterior uveitides was 97.5% in the training set and 96.7% in the validation set (95% confidence interval 92.4, 98.6). Key criteria for VZV anterior uveitis included unilateral anterior uveitis with either 1) positive aqueous humor polymerase chain reaction assay for VZV; 2) sectoral iris atrophy in a patient ≥60 years of age; or 3) concurrent or recent dermatomal herpes zoster. The misclassification rates for VZV anterior uveitis were 0.9% in the training set and 0% in the validation set, respectively.

CONCLUSIONS : The criteria for VZV anterior uveitis had a low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

Jabs Douglas A, Acharya Nisha R, Caspers Laure, Chee Soon-Phaik, Goldstein Debra, McCluskey Peter, Murray Philip I, Oden Neal, Palestine Alan G, Rosenbaum James T, Thorne Jennifer E, Trusko Brett E

2021-Apr-09

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for herpes simplex virus (HSV) anterior uveitis DESIGN: : Machine learning of cases with HSV anterior uveitis and 8 other anterior uveitides.

METHODS : Cases of anterior uveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on the diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the anterior uveitides. The resulting criteria were evaluated on the validation set.

RESULTS : One thousand eighty-three of cases of anterior uveitides, including 101 cases of HSV anterior uveitis, were evaluated by machine learning. The overall accuracy for anterior uveitides was 97.5% in the training set and 96.7% in the validation set (95% confidence interval 92.4, 98.6). Key criteria for HSV anterior uveitis included unilateral anterior uveitis with either 1) positive aqueous humor polymerase chain reaction assay for HSV; 2) sectoral iris atrophy in a patient ≤50 years of age; or 3) HSV keratitis. The misclassification rates for HSV anterior uveitis were 8.3% in the training set and 17% in the validation set, respectively.

CONCLUSIONS : The criteria for HSV anterior uveitis had a reasonably low misclassification rate and appeared to perform well enough for use in clinical and translational research.

Jabs Douglas A, Acharya Nisha R, Caspers Laure, Chee Soon-Phaik, Goldstein Debra, McCluskey Peter, Murray Philip I, Oden Neal, Palestine Alan G, Rosenbaum James T, Thorne Jennifer E, Trusko Brett E

2021-Apr-09

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for Behçet disease uveitis.

DESIGN : Machine learning of cases with Behçet disease and 5 other panuveitides.

METHODS : Cases of panuveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on the diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the intermediate uveitides. The resulting criteria were evaluated on the validation set.

RESULTS : One thousand twelve of cases panuveitides, including 194 cases of Behçet disease with uveitis, were evaluated by machine learning. The overall accuracy for panuveitides was 96.3% in the training set and 94.0% in the validation set (95% confidence interval 89.0, 96.8). Key criteria for Behçet disease uveitis were a diagnosis of Behçet disease using the International Study Group for Behçet Disease criteria and a compatible uveitis, including: 1) anterior uveitis; 2) anterior chamber and vitreous inflammation; 3) posterior uveitis with retinal vasculitis and/or focal infiltrates; or 4) panuveitis with retinal vasculitis and/or focal infiltrates. The misclassification rates for Behçet disease uveitis were 0.6 % in the training set and 0% in the validation set, respectively.

CONCLUSIONS : The criteria for Behçet disease uveitis had a low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

Jabs Douglas A, Dick Andrew D, Dunn James P, Kramer Michal, McCluskey Peter, Oden Neal, Okada Annabelle A, Palestine Alan G, Read Russell W, Thorne Jennifer E, Trusko Brett E, Yeh Steven

2021-Apr-09

In Pancreatology : official journal of the International Association of Pancreatology (IAP) ... [et al.]

OBJECTIVE : Quality gaps in medical imaging datasets lead to profound errors in experiments. Our objective was to characterize such quality gaps in public pancreas imaging datasets (PPIDs), to evaluate their impact on previously published studies, and to provide post-hoc labels and segmentations as a value-add for these PPIDs.

METHODS : We scored the available PPIDs on the medical imaging data readiness (MIDaR) scale, and evaluated for associated metadata, image quality, acquisition phase, etiology of pancreas lesion, sources of confounders, and biases. Studies utilizing these PPIDs were evaluated for awareness of and any impact of quality gaps on their results. Volumetric pancreatic adenocarcinoma (PDA) segmentations were performed for non-annotated CTs by a junior radiologist (R1) and reviewed by a senior radiologist (R3).

RESULTS : We found three PPIDs with 560 CTs and six MRIs. NIH dataset of normal pancreas CTs (PCT) (n = 80 CTs) had optimal image quality and met MIDaR A criteria but parts of pancreas have been excluded in the provided segmentations. TCIA-PDA (n = 60 CTs; 6 MRIs) and MSD(n = 420 CTs) datasets categorized to MIDaR B due to incomplete annotations, limited metadata, and insufficient documentation. Substantial proportion of CTs from TCIA-PDA and MSD datasets were found unsuitable for AI due to biliary stents [TCIA-PDA:10 (17%); MSD:112 (27%)] or other factors (non-portal venous phase, suboptimal image quality, non-PDA etiology, or post-treatment status) [TCIA-PDA:5 (8.5%); MSD:156 (37.1%)]. These quality gaps were not accounted for in any of the 25 studies that have used these PPIDs (NIH-PCT:20; MSD:1; both: 4). PDA segmentations were done by R1 in 91 eligible CTs (TCIA-PDA:42; MSD:49). Of these, corrections were made by R3 in 16 CTs (18%) (TCIA-PDA:4; MSD:12) [mean (standard deviation) Dice: 0.72(0.21) and 0.63(0.23) respectively].

CONCLUSION : Substantial quality gaps, sources of bias, and high proportion of CTs unsuitable for AI characterize the available limited PPIDs. Published studies on these PPIDs do not account for these quality gaps. We complement these PPIDs through post-hoc labels and segmentations for public release on the TCIA portal. Collaborative efforts leading to large, well-curated PPIDs supported by adequate documentation are critically needed to translate the promise of AI to clinical practice.

Suman Garima, Patra Anurima, Korfiatis Panagiotis, Majumder Shounak, Chari Suresh T, Truty Mark J, Fletcher Joel G, Goenka Ajit H

2021-Apr-02

Benchmarking, Bias, Deep learning, Metadata, Pancreatic carcinoma

In Informatics in medicine unlocked

Alballa Norah, Al-Turaiki Isra

2021-Apr-03

COVID-19, Machine learning, artificial intelligence, diagnosis, feature selection, prognosis

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for Fuchs uveitis syndrome.

DESIGN : Machine learning of cases with Fuchs uveitis syndrome and 8 other anterior uveitides.

METHODS : Cases of anterior uveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on the diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the anterior uveitides. The resulting criteria were evaluated on the validation set.

RESULTS : One thousand eighty-three cases of anterior uveitides, including 146 cases of Fuchs uveitis syndrome, were evaluated by machine learning. The overall accuracy for anterior uveitides was 97.5% in the training set and 96.7% in the validation set (95% confidence interval 92.4, 98.6). Key criteria for Fuchs uveitis syndrome included unilateral anterior uveitis with or without vitritis and either: 1) heterochromia or 2) unilateral diffuse iris atrophy and stellate keratic precipitates. The overall accuracy for anterior uveitides was 97.5% in the training set (95% confidence interval [CI] 96.3, 98.4) and 96.7% in the validation set (95% CI 92.4, 98.6). The misclassification rates for FUS were 4.7% in the training set and 5.5% in the validation set, respectively.

CONCLUSIONS : The criteria for Fuchs uveitis syndrome had a low misclassification rate and appeared to perform well enough for use in clinical and translational research.

Working Group The Standardization Of Uveitis Nomenclature Sun, Jabs Douglas A, Acharya Nisha R, Chee Soon-Phaik, Goldstein Debra, McCluskey Peter, Murray Philip I, Oden Neal, Palestine Alan G, Rosenbaum James T, Thorne Jennifer E, Trusko Brett E

2021-Apr-09

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for pars planitis DESIGN: Machine learning of cases with pars planitis and 4 other intermediate uveitides.

METHODS : Cases of intermediate uveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on the diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the intermediate uveitides. The resulting criteria were evaluated on the validation set.

RESULTS : Five hundred eighty-nine cases of intermediate uveitides, including 226 cases of pars planitis, were evaluated by machine learning. The overall accuracy for intermediate uveitides was 99.8% in the training set and 99.3% in the validation set (95% confidence interval 96.1, 99.9). Key criteria for pars planitis included unilateral or bilateral intermediate uveitis with either 1) snowballs in the vitreous or 2) snowbanks on the pars plana. Key exclusions included: 1) multiple sclerosis, 2) sarcoidosis, and 3) syphilis. The misclassification rates for pars planitis were 0 % in the training set and 1.7% in the validation set, respectively.

CONCLUSIONS : The criteria for pars planitis had a low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

The Standardization Of Uveitis Nomenclature Sun Working Group Jabs, Douglas A Denniston, Alastair K Dick, Andrew D Dunn, James P Kramer, Michal Oden, Neal Okada, Annabelle A Palestine, Alan G Read, Russell W Thorne, Jennifer E Trusko, Brett E Yeh

2021-Apr-09

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for sympathetic ophthalmia DESIGN: Machine learning of cases with sympathetic ophthalmia and 5 other panuveitides.

METHODS : Cases of panuveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on the diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the panuveitides. The resulting criteria were evaluated on the validation set.

RESULTS : One thousand twelve cases of panuveitides, including 110 cases of sympathetic ophthalmia, were evaluated by machine learning. The overall accuracy for panuveitides was 96.3% in the training set and 94.0% in the validation set (95% confidence interval 89.0, 96.8). Key criteria for sympathetic ophthalmia included bilateral uveitis with 1) a history of unilateral ocular trauma or surgery and 2) an anterior chamber and vitreous inflammation or a panuveitis with choroidal involvement. The misclassification rates for sympathetic ophthalmia were 4.2 % in the training set and 6.7% in the validation set, respectively.

CONCLUSIONS : The criteria for sympathetic ophthalmia had a low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

Jabs Douglas A, Dick Andrew, Kramer Michal, Muccioli Cristina, Oden Neal, Okada Annabelle A, Palestine Alan G, Rao Narsing A, Read Russell W, Thorne Jennifer E, Trusko Brett E

2021-Apr-09

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for spondyloarthritis/HLA-B27-associated anterior uveitis DESIGN: Machine learning of cases with spondyloarthritis/HLA-B27-associated anterior uveitis and 8 other anterior uveitides.

METHODS : Cases of anterior uveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on the diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the anterior uveitides. The resulting criteria were evaluated on the validation set.

RESULTS : One thousand eighty-three cases of anterior uveitides, including 184 cases of spondyloarthritis/HLA-B27-associated anterior uveitis, were evaluated by machine learning. The overall accuracy for anterior uveitides was 97.5% in the training set (95% confidence interval [CI] 96.3, 98.4) and 96.7% in the validation set (95% CI 92.4, 98.6). Key criteria for spondyloarthritis/HLA-B27-associated anterior uveitis included 1) acute or recurrent acute unilateral or unilateral alternating anterior uveitis with either spondyloarthritis or a positive test for HLA-B27 or 2) chronic anterior uveitis with a history of the classic course and either spondyloarthritis or HLA-B27 or 3) anterior uveitis with both spondyloarthritis and HLA-B27. The misclassification rates for spondyloarthritis/HLA-B27-associated anterior uveitis were 0% in the training set and 3.6% in the validation set, respectively.

CONCLUSIONS : The criteria for spondyloarthritis/HLA-B27-associated anterior uveitis had a low misclassification rate and appeared to perform well enough for use in clinical and translational research.

Jabs Douglas A, Acharya Nisha R, Chee Soon-Phaik, Goldstein Debra, McCluskey Peter, Murray Philip I, Oden Neal, Palestine Alan G, Rosenbaum James T, Thorne Jennifer E, Trusko Brett E

2021-Apr-09

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for birdshot chorioretinitis.

DESIGN : Machine learning of cases with birdshot chorioretinitis and 8 other posterior uveitides.

METHODS : Cases of posterior uveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the infectious posterior/panuveitides. The resulting criteria were evaluated on the validation set.

RESULTS : One thousand sixty-eight cases of posterior uveitides, including 207 cases of birdshot chorioretinitis, were evaluated by machine learning. Key criteria for birdshot chorioretinitis included a multifocal choroiditis with: 1) the characteristic appearance a bilateral multifocal choroiditis with cream-colored or yellow-orange, oval or round choroidal spots ("birdshot" spots); 2) absent to mild anterior chamber inflammation; and 3) absent to moderate vitreous inflammation; or multifocal choroiditis with positive HLA-A29 testing and either: 1) classic "birdshot spots" or 2) characteristic imaging on indocyanine green angiography. Overall accuracy for posterior uveitides was 93.9% in the training set and 98.0% (95% confidence interval 94.3, 99.3) in the validation set. The misclassification rates for birdshot chorioretinitis were 10% in the training set and 0% in the validation set.

CONCLUSIONS : The criteria for birdshot chorioretinitis had a low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

The Standardization Of Uveitis Nomenclature Sun Working Group Jabs, Douglas A Brezin, Antoine P Levinson, Ralph D McCluskey, Peter Oden, Neal Palestine, Alan G Read, Russell W Thorne, Jennifer E Trusko, Brett E Vitale, Albert Wittenberg

2021-Apr-09

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for toxoplasmic retinitis.

DESIGN : Machine learning of cases with toxoplasmic retinitis and 4 other infectious posterior/ panuveitides.

METHODS : Cases of infectious posterior/panuveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on diagnosis, using formal consensus techniques. Cases were split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training set to determine a parsimonious set of criteria that minimized the misclassification rate among the infectious posterior/panuveitides. The resulting criteria were evaluated on the validation set.

RESULTS : Eight hundred three cases of infectious posterior/panuveitides, including 174 cases of toxoplasmic retinitis, were evaluated by machine learning. Key criteria for toxoplasmic retinitis included: 1) focal or paucifocal necrotizing retinitis and either; 2) positive polymerase chain reaction assay for Toxoplasma gondii from an intraocular specimen or 3) the characteristic clinical picture of a round or oval retinitis lesion proximal to a hyperpigmented and/or atrophic chorioretinal scar. Overall accuracy for infectious posterior/panuveitides was 92.1% in the training set and 93.3% (95% confidence interval 88.2, 96.3) in the validation set. The misclassification rates for toxoplasmic retinitis were 8.2% in the training set and 10% in the validation set.

CONCLUSIONS : The criteria for toxoplasmic retinitis had a low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

**

2021-Apr-09

In American journal of ophthalmology ; h5-index 67.0

PURPOSE : To determine classification criteria for sarcoidosis-associated uveitis DESIGN: Machine learning of cases with sarcoid uveitis and 15 other uveitides.

METHODS : Cases of anterior, intermediate, and panuveitides were collected in an informatics-designed preliminary database, and a final database was constructed of cases achieving supermajority agreement on the diagnosis, using formal consensus techniques. Cases were analyzed by anatomic class, and each class was split into a training set and a validation set. Machine learning using multinomial logistic regression was used on the training sets to determine a parsimonious set of criteria that minimized the misclassification rate among the intermediate uveitides. The resulting criteria were evaluated on the validation sets.

RESULTS : One thousand eighty-three anterior uveitides, 589 intermediate uveitides, and 1012 panuveitides, including 278 cases of sarcoidosis-associated uveitis, were evaluated by machine learning. Key criteria for sarcoidosis-associated uveitis included a compatible uveitic syndrome of any anatomic class and evidence of sarcoidosis, either 1) a tissue biopsy demonstrating non-caseating granulomata or 2) bilateral hilar adenopathy on chest imaging. The overall accuracy of the diagnosis of sarcoidosis-associated uveitis in the validation set was 99.7% (95% confidence interval 98.8, 99.9).The misclassification rates for sarcoidosis-associated uveitis in the training sets were: anterior uveitis 3.2%, intermediate uveitis 2.6%, and panuveitis 1.2%; in the validation sets the misclassification rates were: anterior uveitis 0%, intermediate uveitis 0%, and panuveitis 0%, respectively.

CONCLUSIONS : The criteria for sarcoidosis-associated uveitis had a low misclassification rate and appeared to perform sufficiently well for use in clinical and translational research.

Jabs Douglas A, Acharya Nisha R, Denniston Alastair K, Lightman Susan L, McCluskey Peter, Oden Neal, Okada Annabelle A, Palestine Alan G, Thorne Jennifer E, Trusko Brett E, Vitale Albert

2021-Apr-09

In PloS one ; h5-index 176.0

Schafer Katherine M, Kennedy Grace, Gallyer Austin, Resnik Philip

2021

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : National governments have implemented non-pharmaceutical interventions to control and mitigate against the COVID-19 pandemic.

OBJECTIVE : We investigate the prediction of future daily national Confirmed Infection Growths - the percentage change in total cumulative cases across 14 days for 114 countries using non-pharmaceutical intervention metrics and cultural dimension metrics, which are metrics indicative of specific national sociocultural norms.

METHODS : We combine the OxCGRT dataset, Hofstede's cultural dimensions, and COVID-19 daily reported infection case numbers to train and evaluate five non-time series machine learning models in predicting Confirmed Infection Growth. We use three validation methods - in-distribution, out-of-distribution, and country-based cross-validation - for evaluation, each applicable to a different use case of the models.

RESULTS : Our results demonstrate high R2 values between the labels and predictions for the in-distribution method (0.959), and moderate R2 values for the out-of-distribution and country-based cross-validation methods (0.513 and 0.574, respectively) using random forest and AdaBoost regression. While these models may be used to predict the Confirmed Infection Growth, the differing accuracies obtained from the three tasks suggest a strong influence of the use case.

CONCLUSIONS : This work provides new considerations in using machine learning techniques with non-pharmaceutical interventions and cultural dimensions data for predicting the national growth of confirmed infections of COVID-19.

CLINICALTRIAL :

Yeung Arnold Ys, Roewer-Despres Francois, Rosella Laura, Rudzicz Frank

2021-Mar-23

In IEEE transactions on medical imaging ; h5-index 74.0

Huang Yixing, Preuhs Alexander, Manhart Michael, Lauritsch Guenter, Maier Andreas

2021-Apr-12

In The Journal of prosthetic dentistry ; h5-index 51.0

STATEMENT OF PROBLEM : Artificial intelligence (AI) applications are increasing in restorative procedures. However, the current development and performance of AI in restorative dentistry applications has not yet been systematically documented and analyzed.

PURPOSE : The purpose of this systematic review was to identify and evaluate the ability of AI models in restorative dentistry to diagnose dental caries and vertical tooth fracture, detect tooth preparation margins, and predict restoration failure.

MATERIAL AND METHODS : An electronic systematic review was performed in 5 databases: MEDLINE/PubMed, EMBASE, World of Science, Cochrane, and Scopus. A manual search was also conducted. Studies with AI models were selected based on 4 criteria: diagnosis of dental caries, diagnosis of vertical tooth fracture, detection of the tooth preparation finishing line, and prediction of restoration failure. Two investigators independently evaluated the quality assessment of the studies by applying the Joanna Briggs Institute (JBI) Critical Appraisal Checklist for Quasi-Experimental Studies (nonrandomized experimental studies). A third investigator was consulted to resolve lack of consensus.

RESULTS : A total of 34 articles were included in the review: 29 studies included AI techniques for the diagnosis of dental caries or the elaboration of caries and postsensitivity prediction models, 2 for the diagnosis of vertical tooth fracture, 1 for the tooth preparation finishing line location, and 2 for the prediction of the restoration failure. Among the studies reviewed, the AI models tested obtained a caries diagnosis accuracy ranging from 76% to 88.3%, sensitivity ranging from 73% to 90%, and specificity ranging from 61.5% to 93%. The caries prediction accuracy among the studies ranged from 83.6% to 97.1%. The studies reported an accuracy for the vertical tooth fracture diagnosis ranging from 88.3% to 95.7%. The article using AI models to locate the finishing line reported an accuracy ranging from 90.6% to 97.4%.

CONCLUSIONS : AI models have the potential to provide a powerful tool for assisting in the diagnosis of caries and vertical tooth fracture, detecting the tooth preparation margin, and predicting restoration failure. However, the dental applications of AI models are still in development. Further studies are required to assess the clinical performance of AI models in restorative dentistry.

Revilla-León Marta, Gómez-Polo Miguel, Vyas Shantanu, Barmak Basir A, Özcan Mutlu, Att Wael, Krishnamurthy Vinayak R

2021-Apr-09

In Frontiers in robotics and AI

Ahmed Syed Nizamuddin

2021

AI, COVID-19, artificial intelligence, neurologist, neurology, robotics, telemedicine

In Biotechnology and bioengineering

Ren Tingwei, Tan Zhijun, Ehamparanathan Vivekh, Lewandowski Angela, Ghose Sanchayita, Li Zheng Jian

2021-Apr-12

Antibody, Disulfide bond, Process development, Reduction/Oxidation

In Molecular diversity

Gupta Rohan, Srivastava Devesh, Sahu Mehar, Tiwari Swati, Ambasta Rashmi K, Kumar Pravir

2021-Apr-12

Artificial intelligence, Artificial neural networks, Computer-aided drug design, Deep learning, Drug design and discovery, Drug repurposing, Machine learning, Quantitative structure–activity relationship, Virtual screening

In Pediatric radiology

Dillman Jonathan R, Somasundaram Elan, Brady Samuel L, He Lili

2021-Apr-12

Abdomen, Artificial intelligence, Children, Computed tomography, Deep learning, Machine learning, Magnetic resonance imaging

In Water science and technology : a journal of the International Association on Water Pollution Research

Hejabi Nasim, Saghebian Seyed Mahdi, Aalami Mohammad Taghi, Nourani Vahid

2021-Apr

In Journal of X-ray science and technology

OBJECTIVE : In order to solve the blurred structural details and over-smoothing effects in sparse representation dictionary learning reconstruction algorithm, this study aims to test sparse angle CT reconstruction with weighted dictionary learning algorithm based on adaptive Group-Sparsity Regularization (AGSR-SART).

METHODS : First, a new similarity measure is defined in which Covariance is introduced into Euclidean distance, Non-local image patches are adaptively divided into groups of different sizes as the basic unit of sparse representation. Second, the weight factor of the regular constraint terms is designed through the residuals represented by the dictionary, so that the algorithm takes different smoothing effects on different regions of the image during the iterative process. The sparse reconstructed image is modified according to the difference between the estimated value and the intermediate image. Last, The SBI (Split Bregman Iteration) iterative algorithm is used to solve the objective function. An abdominal image, a pelvic image and a thoracic image are employed to evaluate performance of the proposed method.

RESULTS : In terms of quantitative evaluations, experimental results show that new algorithm yields PSNR of 48.20, the maximum SSIM of 99.06% and the minimum MAE of 0.0028.

CONCLUSIONS : This study demonstrates that new algorithm can better preserve structural details in reconstructed CT images. It eliminates the effect of excessive smoothing in sparse angle reconstruction, enhances the sparseness and non-local self-similarity of the image, and thus it is superior to several existing reconstruction algorithms.

Yang Tiejun, Tang Lu, Tang Qi, Li Lei

2021-Apr-02

Adaptive group-sparsity regularization, CT reconstruction, dictionary learning, spares angle

In Journal of Alzheimer's disease : JAD

BACKGROUND : Olfactory dysfunction (OD) is an early symptom of Alzheimer's disease (AD). However, olfactory testing is not commonly performed to test OD in the setting of AD.

OBJECTIVE : This work investigates objective OD as a non-invasive biomarker for accurately classifying subjects as cognitively unimpaired (CU), mild cognitive impairment (MCI), and AD.

METHODS : Patients with MCI (n = 24) and AD (n = 24), and CU (n = 33) controls completed two objective tests of olfaction (Affordable, Rapid, Olfactory Measurement Array -AROMA; Sniffin' Sticks Screening 12 Test -SST12). Demographic and subjective sinonasal and olfaction symptom information was also obtained. Analyses utilized traditional statistics and machine learning to determine olfactory variables, and combinations of variables, of importance for differentiating normal and disease states.

RESULTS : Inability to correctly identify a scent after detection was a hallmark of MCI/AD. AROMA was superior to SST12 for differentiating MCI from AD. Performance on the clove scent was significantly different between all three groups. AROMA regression modeling yielded six scents with AUC of the ROC of 0.890 (p <  0.001). Random forest model machine learning algorithms considering AROMA olfactory data successfully predicted MCI versus AD disease state. Considering only AROMA data, machine learning algorithms were 87.5%accurate (95%CI 0.4735, 0.9968). Sensitivity and specificity were 100%and 75%, respectively with ROC of 0.875. When considering AROMA and subject demographic and subjective data, the AUC of the ROC increased to 0.9375.

CONCLUSION : OD differentiates CUs from those with MCI and AD and can accurately predict MCI versus AD. Leveraging OD data may meaningfully guide management and research decisions.

Li Jennifer, Bur Andres M, Villwock Mark R, Shankar Suraj, Palmer Gracie, Sykes Kevin J, Villwock Jennifer A

2021-Apr-05

Alzheimer’s disease, machine learning, mild cognitive impairment, olfaction, olfactory dysfunction

In Journal of clinical sleep medicine : JCSM : official publication of the American Academy of Sleep Medicine

STUDY OBJECTIVES : The aim of the study was to inspect acoustic properties and sleep characteristics of pre-apneic snoring sound. The feasibility of forecasting upcoming respiratory events by snoring sound was also investigated.

METHODS : Participants with habitual snoring or heavy breathing sound during sleep were recruited consecutively. Polysomnography was conducted and snoring related breathing sound was recorded simultaneously. Acoustic features and sleep features were extracted from 30-second samples and a machine learning algorithm was used to establish two prediction models.

RESULTS : A total of 74 eligible participants were included. Model 1 tested by five-fold cross validation achieved the accuracy of 0.92 and area under the curve of 0.94 for respiratory event prediction. model 2 with acoustic features and sleep information tested by Leave-One-Out cross validation had the accuracy of 0.78 and area under the curve of 0.80. Sleep position was found to be the most important amongst all sleep features contributing to the performance.

CONCLUSIONS : Pre-apneic sound presented unique acoustic characteristics and snoring related breathing sound could be deployed as a real-time apneic event predictor. The model combined with sleep information served as a promising tool for an early warning system to forecast apneic events.

Wang Bochun, Yi Xuanyu, Gao Jiandong, Li Yanru, Xu Wen, Wu Ji, Han Demin

2021-Apr-12

acoustic features, early warning system, obstructive sleep apnea, real-time prediction, snoring related breathing sound

In Cardiology in the young

McMahon Colin J, Tretter Justin T, Redington Andrew N, Bu’Lock Frances, Zühlke Liesl, Heying Ruth, Mattos Sandra, Kumar R Krishna, Jacobs Jeffrey P, Windram Jonathan D

2021-Apr-12

Adult Congenital Heart Disease, Congenital Cardiology, Congenital Heart Disease, Education, Paediatric Cardiology, Training

In Expert opinion on drug discovery ; h5-index 34.0

INTRODUCTION : Knowledge graphs have proven to be promising systems of information storage and retrieval. Due to the recent explosion of heterogeneous multimodal data sources generated in the biomedical domain, and an industry shift toward a systems biology approach, knowledge graphs have emerged as attractive methods of data storage and hypothesis generation.

AREAS COVERED : In this review, the author summarizes the applications of knowledge graphs in drug discovery. They evaluate their utility; differentiating between academic exercises in graph theory, and useful tools to derive novel insights, highlighting target identification and drug repurposing as two areas showing particular promise. They provide a case study on COVID-19, summarizing the research that used knowledge graphs to identify repurposable drug candidates. They describe the dangers of degree and literature bias, and discuss mitigation strategies.

EXPERT OPINION : Whilst knowledge graphs and graph-based machine learning have certainly shown promise, they remain relatively immature technologies. Many popular link prediction algorithms fail to address strong biases in biomedical data, and only highlight biological associations, failing to model causal relationships in complex dynamic biological systems. These problems need to be addressed before knowledge graphs reach their true potential in drug discovery.

MacLean Finlay

2021-Apr-12

Biomedical knowledge graphs, drug repositioning, drug repurposing, graph machine learning, heterogeneous information networks, knowledge graph embedding, network embeddings, network medicine, network pharmacology

In Analytical chemistry

Cheng Hao, Rao Bing, Liu Lei, Cui Lizhen, Xiao Guobao, Su Ran, Wei Leyi

2021-Apr-12

In Australian health review : a publication of the Australian Hospital Association

Zhou Huaqiong, Albrecht Matthew A, Roberts Pamela A, Porter Paul, Della Philip R

2021-Apr-12

In International journal of disaster risk reduction : IJDRR

Reddin Karen, Bang Henry, Miles Lee

2021-Apr-05

Emergency Exercise, Epidemic, Lessons learnt, Pandemic, Simulation

In European heart journal cardiovascular Imaging

AIMS : To evaluate the impact of coronary artery calcium (CAC) score, minimal lumen area (MLA), and length of coronary artery stenosis on the diagnostic performance of the machine-learning-based computed tomography-derived fractional flow reserve (ML-FFR).

METHODS AND RESULTS : In 471 patients with coronary artery disease, computed tomography angiography (CTA) and invasive coronary angiography were performed with fractional flow reserve (FFR) in 557 lesions at a single centre. Diagnostic performances of ML-FFR, computational fluid dynamics-based CT-FFR (CFD-FFR), MLA, quantitative coronary angiography (QCA), and visual stenosis grading were evaluated using invasive FFR as a reference standard. Diagnostic performances were analysed according to lesion characteristics including the MLA, length of stenosis, CAC score, and stenosis degree. ML-FFR was obtained by automated feature selection and model building from quantitative CTA. A total of 272 lesions showed significant ischaemia, defined by invasive FFR ≤0.80. There was a significant correlation between CFD-FFR and ML-FFR (r = 0.99, P < 0.001). ML-FFR showed moderate sensitivity and specificity in the per-patient analysis. Diagnostic performances of CFD-FFR and ML-FFR did not decline in patients with high CAC scores (CAC > 400). Sensitivities of CFD-FFR and ML-FFR showed a downward trend along with the increase in lesion length and decrease in MLA. The area under the curve (AUC) of ML-FFR (0.73) was higher than those of QCA and visual grading (AUC = 0.65 for both, P < 0.001) and comparable to those of MLA (AUC = 0.71, P = 0.21) and CFD-FFR (AUC = 0.73, P = 0.86).

CONCLUSION : ML-FFR showed comparable results to MLA and CFD-FFR for the prediction of lesion-specific ischaemia. Specificities and accuracies of CFD-FFR and ML-FFR decreased with smaller MLA and long lesion length.

Koo Hyun Jung, Kang Joon-Won, Kang Soo-Jin, Kweon Jihoon, Lee June-Goo, Ahn Jung-Min, Park Duk-Woo, Lee Seung Whan, Lee Cheol Whan, Park Seong-Wook, Park Seung-Jung, Kim Young-Hak, Yang Dong Hyun

2021-Apr-11

CT fractional flow reserve, coronary artery disease, coronary calcium, machine-learning

In Briefings in bioinformatics

  : Essential genes are critical for the growth and survival of any organism. The machine learning approach complements the experimental methods to minimize the resources required for essentiality assays. Previous studies revealed the need to discover relevant features that significantly classify essential genes, improve on the generalizability of prediction models across organisms, and construct a robust gold standard as the class label for the train data to enhance prediction. Findings also show that a significant limitation of the machine learning approach is predicting conditionally essential genes. The essentiality status of a gene can change due to a specific condition of the organism. This review examines various methods applied to essential gene prediction task, their strengths, limitations and the factors responsible for effective computational prediction of essential genes. We discussed categories of features and how they contribute to the classification performance of essentiality prediction models. Five categories of features, namely, gene sequence, protein sequence, network topology, homology and gene ontology-based features, were generated for Caenorhabditis elegans to perform a comparative analysis of their essentiality prediction capacity. Gene ontology-based feature category outperformed other categories of features majorly due to its high correlation with the genes' biological functions. However, the topology feature category provided the highest discriminatory power making it more suitable for essentiality prediction. The major limiting factor of machine learning to predict essential genes conditionality is the unavailability of labeled data for interest conditions that can train a classifier. Therefore, cooperative machine learning could further exploit models that can perform well in conditional essentiality predictions.

SHORT ABSTRACT : Identification of essential genes is imperative because it provides an understanding of the core structure and function, accelerating drug targets' discovery, among other functions. Recent studies have applied machine learning to complement the experimental identification of essential genes. However, several factors are limiting the performance of machine learning approaches. This review aims to present the standard procedure and resources available for predicting essential genes in organisms, and also highlight the factors responsible for the current limitation in using machine learning for conditional gene essentiality prediction. The choice of features and ML technique was identified as an important factor to predict essential genes effectively.

Aromolaran Olufemi, Aromolaran Damilare, Isewon Itunuoluwa, Oyelade Jelili

2021-Apr-10

conditional essentiality, conditionally essential genes, essential genes, essential proteins, feature selection, supervised learning

In Biophysics reviews

Yang Zhenzhong, Francisco Jack, Reese Alexandra S, Spriggs David R, Im Hyungsoon, Castro Cesar M

2021-Mar

In IBRO neuroscience reports

Wang Ridey H, Dai Lulin, Okamura Jun-Ya, Fuchida Takayasu, Wang Gang

2021-Jun

Discrimination, Inferotemporal cortex, Learning, Monkey, Object recognition, View-invariance

In Radiology. Artificial intelligence

Purpose : To generate and assess an algorithm combining eye tracking and speech recognition to extract brain lesion location labels automatically for deep learning (DL).

Materials and Methods : In this retrospective study, 700 two-dimensional brain tumor MRI scans from the Brain Tumor Segmentation database were clinically interpreted. For each image, a single radiologist dictated a standard phrase describing the lesion into a microphone, simulating clinical interpretation. Eye-tracking data were recorded simultaneously. Using speech recognition, gaze points corresponding to each lesion were obtained. Lesion locations were used to train a keypoint detection convolutional neural network to find new lesions. A network was trained to localize lesions for an independent test set of 85 images. The statistical measure to evaluate our method was percent accuracy.

Results : Eye tracking with speech recognition was 92% accurate in labeling lesion locations from the training dataset, thereby demonstrating that fully simulated interpretation can yield reliable tumor location labels. These labels became those that were used to train the DL network. The detection network trained on these labels predicted lesion location of a separate testing set with 85% accuracy.

Conclusion : The DL network was able to locate brain tumors on the basis of training data that were labeled automatically from simulated clinical image interpretation.© RSNA, 2020.

Stember Joseph N, Celik Haydar, Gutman David, Swinburne Nathaniel, Young Robert, Eskreis-Winkler Sarah, Holodny Andrei, Jambawalikar Sachin, Wood Bradford J, Chang Peter D, Krupinski Elizabeth, Bagci Ulas

2021-Jan

In EClinicalMedicine

Background : Mycoplasma genitalium (MG) causes a sexually transmitted infection (STI) with a rising rate of antimicrobial resistance. Currently, guidelines do not recommend screening asymptomatic men who have sex with men (MSM). We developed a mathematical model of MG transmission to examine the impact of various screening strategies on the incidence and prevalence of MG among MSM attending a sexual health clinic.

Methods : A compartmental mathematical model of MG transmission among MSM was constructed and calibrated using data from the Melbourne Sexual Health center, where resistance-guided therapy provides high treatment effectiveness (92-95%). The model stratified men by symptom status, sexual risk behaviours and whether or not they had MG with macrolide resistance. We simulated the impact on endemic steady-state MG prevalence and incidence of the following screening scenarios, namely screening: 1) no MSM; 2) only symptomatic MSM (the current recommendation); 3) all symptomatic and high-risk asymptomatic MSM; and 4) all MSM. Our base case analysis assumed a treatment effectiveness of 92-95% using resistance-guided therapy. We also examined the impact of treatment effectiveness (i.e. the proportion of detected MG that were cured) and screening coverage (i.e. testing rate) on MG prevalence.

Findings : The model predicts that the overall endemic MG prevalence is 9.1% (95% CI: 7.9-10.0) in the current situation where screening is only offered to symptomatic MSM (base-case). This would increase to 11·4% (95% confidence intervals (CI): 10.2-13.7) if no MSM are offered screening, but would decrease to 7.3% (95% CI: 5.7-8.4) if all symptomatic and high-risk asymptomatic MSM were offered screening and 6.4% (95% CI: 4.7-7·7) if all MSM were offered screening. Increasing coverage of MSM screening strategies shows a similar effect on decreasing endemic MG incidence. When evaluating the simultaneous impact of treatment effectiveness and screening coverage, we found that offering screening to more MSM may reduce the overall prevalence but leads to a higher proportion of macrolide-resistant MG, particularly when using treatment regimens with lower effectiveness.

Interpretation : Based on the available treatment options, offering screening for MG to other MSM (beyond the currently recommended group of symptomatic MSM) could slightly reduce the prevalence and incidence of MG. However, further increasing screening coverage must be weighed against the impact of lower treatment effectiveness (i.e. when not using resistance-guided therapy), increasing the selection of macrolide resistance, and other negative consequences related to AMR and management (e.g. unnecessary psychological morbidity from infections that do not need treatment).

Ong Jason J, Ruan Luanqi, Lim Aaron G, Bradshaw Catriona S, Taylor-Robinson David, Unemo Magnus, Horner Patrick J, Vickerman Peter, Zhang Lei

2021-Mar

In ACS omega

Jung Hyo Gi, Lee Dongtak, Lee Sang Won, Kim Insu, Kim Yonghwan, Jang Jae Won, Lee Jeong Hoon, Lee Gyudo, Yoon Dae Sung

2021-Apr-06

In Journal of computational social science

Caliskan Cantay

2021-Apr-05

Awareness, COVID-19, Emotion classification, Twitter

In International journal of data science and analytics

Buczynski Wojtek, Cuzzolin Fabio, Sahakian Barbara