English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The novel 2019 Coronavirus disease (COVID-19) global pandemic is a defininghealth crisis. Recent efforts have been increasingly directed towards achievingquick and accurate detection of COVID-19 across symptomatic patients tomitigate the intensity and spread of the disease. Artificial intelligence (AI)algorithms applied to chest X-ray (CXR) images have emerged as promisingdiagnostic tools, and previous work has demonstrated impressive classificationperformances. However, such methods have faced criticisms from physicians dueto their black-box reasoning process and unpredictable nature. In contrast toprofessional radiologist diagnosis, AI systems often lack generalizability,explainability, and robustness in the clinical decision making process. In ourwork, we address these issues by first proposing an extensive baseline study,training and evaluating 21 convolutional neural network (CNN) models on adiverse set of 33,000+ CXR images to classify between healthy, COVID-19, andnon-COVID-19 pneumonia CXRs. Our resulting models achieved a 3-wayclassification accuracy, recall, and precision of up to 97.03\%, 97.97\%, and99.95\%, respectively. Next, we investigate the effectiveness of adversarialtraining on model robustness and explainability via Gradient-weighted ClassActivation Mapping (Grad-CAM) heatmaps. We find that adversarially trainedmodels not only significantly outperform their standard counterparts onclassifying perturbed images, but also yield saliency maps that 1) betterspecify clinically relevant features, 2) are robust against extraneousartifacts, and 3) agree considerably more with expert radiologist findings.

Robust and Interpretable COVID-19 Diagnosis on Chest X-ray Images using Adversarial Training