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 work tackles a central machine learning problem of performancedegradation on out-of-distribution (OOD) test sets. The problem is particularlysalient in medical imaging based diagnosis system that appears to be accuratebut fails when tested in new hospitals/datasets. Recent studies indicate thesystem might learn shortcut and non-relevant features instead of generalizablefeatures, so-called good features. We hypothesize that adversarial training caneliminate shortcut features whereas saliency guided training can filter outnon-relevant features; both are nuisance features accounting for theperformance degradation on OOD test sets. With that, we formulate a novel modeltraining scheme for the deep neural network to learn good features forclassification and/or detection tasks ensuring a consistent generalizationperformance on OOD test sets. The experimental results qualitatively andquantitatively demonstrate the superior performance of our method using thebenchmark CXR image data sets on classification tasks.

Saliency Guided Adversarial Training for Learning Generalizable Features with Applications to Medical Imaging Classification System