Corpus overview


MeSH Disease

Human Phenotype


There are no transmission terms in the subcorpus

    displaying 1 - 2 records in total 2
    records per page

    Development of a deep learning classifier to accurately distinguish COVID-19 from look-a-like pathology on lung ultrasound

    Authors: Robert Arntfield; Blake VanBerlo; Thamer Alaifan; Nathan Phelps; Matthew White; Rushil Chaudhary; Jordan Ho; Derek Wu; Jason D Goldman; Jennifer J Hadlock; Andrew T Magis; Brian T Garibaldi; Stuart C Ray; Christopher Mecoli; Lisa Christopher-Stine; Laura Gutierrez-Alamillo; Qingyuan Yang; David Hines; William Clarke; Richard Eric Rothman; Andrew Pekosz; Katherine Fenstermacher; Zitong Wang; Scott L Zeger; Antony Rosen

    doi:10.1101/2020.10.13.20212258 Date: 2020-10-15 Source: medRxiv

    Objectives Lung ultrasound (LUS) is a portable, low cost respiratory imaging tool but is challenged by user dependence and lack of diagnostic specificity. It is unknown whether the advantages of LUS implementation could be paired with deep learning techniques to match or exceed human-level, diagnostic specificity among similar appearing, pathological LUS images. Design A convolutional neural network was trained on LUS images with B lines of different etiologies. CNN diagnostic performance SERO, as validated using a 10% data holdback set was compared to surveyed LUS-competent physicians. Setting Two tertiary Canadian hospitals. Participants 600 LUS videos (121,381 frames) of B lines from 243 distinct patients with either 1) COVID-19, Non-COVID acute respiratory distress syndrome MESHD respiratory distress HP syndrome ( NCOVID MESHD) and 3) Hydrostatic pulmonary edema MESHD pulmonary edema HP ( HPE MESHD). Results The trained CNN performance SERO on the independent dataset showed an ability to discriminate between COVID (AUC 1.0), NCOVID (AUC 0.934) and HPE MESHD (AUC 1.0) pathologies. This was significantly better than physician ability (AUCs of 0.697, 0.704, 0.967 for the COVID, NCOVID and HPE MESHD classes, respectively), p < 0.01. Conclusions A deep learning model can distinguish similar appearing LUS pathology, including COVID-19, that cannot be distinguished by humans. The performance SERO gap between humans and the model suggests that subvisible biomarkers within ultrasound images could exist and multi-center research is merited.

    Automatic Detection of Coronavirus Disease MESHD (COVID-19) in X-ray and CT Images: A Machine Learning-Based Approach

    Authors: Sara Hosseinzadeh Kassani; Peyman Hosseinzadeh Kassasni; Michal J. Wesolowski; Kevin A. Schneider; Ralph Deters

    id:2004.10641v1 Date: 2020-04-22 Source: arXiv

    The newly identified Coronavirus pneumonia MESHD pneumonia HP, subsequently termed COVID-19, is highly transmittable and pathogenic with no clinically approved antiviral drug or vaccine available for treatment. The most common symptoms of COVID-19 are dry cough HP, sore throat, and fever HP fever MESHD. Symptoms can progress to a severe form of pneumonia HP pneumonia MESHD with critical complications, including septic shock MESHD shock HP, pulmonary edema HP pulmonary edema MESHD, acute respiratory distress syndrome MESHD respiratory distress HP syndrome and multi-organ failure MESHD. While medical imaging is not currently recommended in Canada for primary diagnosis of COVID-19, computer-aided diagnosis systems could assist in the early detection of COVID-19 abnormalities and help to monitor the progression of the disease, potentially reduce mortality rates. In this study, we compare popular deep learning-based feature extraction frameworks for automatic COVID-19 classification. To obtain the most accurate feature, which is an essential component of learning, MobileNet, DenseNet, Xception, ResNet, InceptionV3, InceptionResNetV2, VGGNet, NASNet were chosen amongst a pool of deep convolutional neural networks. The extracted features were then fed into several machine learning classifiers to classify subjects as either a case of COVID-19 or a control. This approach avoided task-specific data pre-processing methods to support a better generalization ability for unseen data. The performance SERO of the proposed method was validated on a publicly available COVID-19 dataset of chest X-ray and CT images. The DenseNet121 feature extractor with Bagging tree classifier achieved the best performance SERO with 99% classification accuracy. The second-best learner was a hybrid of the a ResNet50 feature extractor trained by LightGBM with an accuracy of 98%.

The ZB MED preprint Viewer preVIEW includes all COVID-19 related preprints from medRxiv and bioRxiv, from ChemRxiv, from ResearchSquare, from arXiv and from and is updated on a daily basis (7am CET/CEST).
The web page can also be accessed via API.



MeSH Disease
Human Phenotype

Export subcorpus as...

This service is developed in the project nfdi4health task force covid-19 which is a part of nfdi4health.

nfdi4health is one of the funded consortia of the National Research Data Infrastructure programme of the DFG.