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

This paper addresses the new problem of automated screening of coronavirusdisease 2019 (COVID-19) based on chest X-rays, which is urgently demandedtoward fast stopping the pandemic. However, robust and accurate screening ofCOVID-19 from chest X-rays is still a globally recognized challenge because oftwo bottlenecks: 1) imaging features of COVID-19 share some similarities withother pneumonia on chest X-rays, and 2) the misdiagnosis rate of COVID-19 isvery high, and the misdiagnosis cost is expensive. While a few pioneering workshave made much progress, they underestimate both crucial bottlenecks. In thispaper, we report our solution, discriminative cost-sensitive learning (DCSL),which should be the choice if the clinical needs the assisted screening ofCOVID-19 from chest X-rays. DCSL combines both advantages from fine-grainedclassification and cost-sensitive learning. Firstly, DCSL develops aconditional center loss that learns deep discriminative representation.Secondly, DCSL establishes score-level cost-sensitive learning that canadaptively enlarge the cost of misclassifying COVID-19 examples into otherclasses. DCSL is so flexible that it can apply in any deep neural network. Wecollected a large-scale multi-class dataset comprised of 2,239 chest X-rayexamples: 239 examples from confirmed COVID-19 cases, 1,000 examples withconfirmed bacterial or viral pneumonia cases, and 1,000 examples of healthypeople. Extensive experiments on the three-class classification show that ouralgorithm remarkably outperforms state-of-the-art algorithms. It achieves anaccuracy of 97.01%, a precision of 97%, a sensitivity of 97.09%, and anF1-score of 96.98%. These results endow our algorithm as an efficient tool forthe fast large-scale screening of COVID-19.

Robust Screening of COVID-19 from Chest X-ray via Discriminative Cost-Sensitive Learning