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

Surgical action triplet recognition provides a better understanding of thesurgical scene. This task is of high relevance as it provides the surgeon withcontext-aware support and safety. The current go-to strategy for improvingperformance is the development of new network mechanisms. However, theperformance of current state-of-the-art techniques is substantially lower thanother surgical tasks. Why is this happening? This is the question that weaddress in this work. We present the first study to understand the failure ofexisting deep learning models through the lens of robustness andexplainability. Firstly, we study current existing models under weak and strong$\delta-$perturbations via an adversarial optimisation scheme. We then analysethe failure modes via feature based explanations. Our study reveals that thekey to improving performance and increasing reliability is in the core andspurious attributes. Our work opens the door to more trustworthy and reliabledeep learning models in surgical data science.

Why Deep Surgical Models Fail?: Revisiting Surgical Action Triplet Recognition through the Lens of Robustness