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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Deep learning models are routinely employed in computational pathology(CPath) for solving problems of diagnostic and prognostic significance.Typically, the generalization performance of CPath models is analyzed usingevaluation protocols such as cross-validation and testing on multi-centriccohorts. However, to ensure that such CPath solutions are robust and safe foruse in a clinical setting, a critical analysis of their predictive performanceand vulnerability to adversarial attacks is required, which is the focus ofthis paper. Specifically, we show that a highly accurate model forclassification of tumour patches in pathology images (AUC > 0.95) can easily beattacked with minimal perturbations which are imperceptible to lay humans andtrained pathologists alike. Our analytical results show that it is possible togenerate single-instance white-box attacks on specific input images with highsuccess rate and low perturbation energy. Furthermore, we have also generated asingle universal perturbation matrix using the training dataset only which,when added to unseen test images, results in forcing the trained neural networkto flip its prediction labels with high confidence at a success rate of > 84%.We systematically analyze the relationship between perturbation energy of anadversarial attack, its impact on morphological constructs of clinicalsignificance, their perceptibility by a trained pathologist and saliency mapsobtained using deep learning models. Based on our analysis, we stronglyrecommend that computational pathology models be critically analyzed using theproposed adversarial validation strategy prior to clinical adoption.

Now You See It, Now You Dont: Adversarial Vulnerabilities in Computational Pathology