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 neural networks are discovered to be non-robust when attacked byimperceptible adversarial examples, which is dangerous for it applied intomedical diagnostic system that requires high reliability. However, the defensemethods that have good effect in natural images may not be suitable for medicaldiagnostic tasks. The preprocessing methods (e.g., random resizing,compression) may lead to the loss of the small lesions feature in the medicalimage. Retraining the network on the augmented data set is also not practicalfor medical models that have already been deployed online. Accordingly, it isnecessary to design an easy-to-deploy and effective defense framework formedical diagnostic tasks. In this paper, we propose a Robust and Retrain-LessDiagnostic Framework for Medical pretrained models against adversarial attack(i.e., MedRDF). It acts on the inference time of the pertained medical model.Specifically, for each test image, MedRDF firstly creates a large number ofnoisy copies of it, and obtains the output labels of these copies from thepretrained medical diagnostic model. Then, based on the labels of these copies,MedRDF outputs the final robust diagnostic result by majority voting. Inaddition to the diagnostic result, MedRDF produces the Robust Metric (RM) asthe confidence of the result. Therefore, it is convenient and reliable toutilize MedRDF to convert pre-trained non-robust diagnostic models into robustones. The experimental results on COVID-19 and DermaMNIST datasets verify theeffectiveness of our MedRDF in improving the robustness of medical diagnosticmodels.

MedRDF: A Robust and Retrain-Less Diagnostic Framework for Medical Pretrained Models Against Adversarial Attack