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

TL;DR

This paper investigates why current deep learning models for surgical action triplet recognition underperform by analyzing their robustness and explainability, revealing core and spurious attributes as key factors affecting reliability.

Contribution

First study to analyze surgical action triplet models' failure modes through robustness and explainability, highlighting the importance of core and spurious features.

Findings

Models are sensitive to adversarial perturbations.

Failure modes are linked to core and spurious attribute reliance.

Enhancing robustness can improve model reliability.

Abstract

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