EndoARSS: Adapting Spatially-Aware Foundation Model for Efficient Activity Recognition and Semantic Segmentation in Endoscopic Surgery

TL;DR

EndoARSS is a multi-task learning framework built on DINOv2, designed to improve activity recognition and semantic segmentation in endoscopic surgery by leveraging spatially-aware multi-scale attention and efficient fine-tuning techniques.

Contribution

The paper introduces EndoARSS, a novel multi-task learning approach that combines Low-Rank Adaptation and spatially-aware multi-scale attention for enhanced surgical scene understanding.

Findings

Significantly outperforms existing models in accuracy and robustness.

Introduces three new datasets for endoscopic surgery analysis.

Demonstrates effective multi-task learning for complex surgical environments.

Abstract

Endoscopic surgery is the gold standard for robotic-assisted minimally invasive surgery, offering significant advantages in early disease detection and precise interventions. However, the complexity of surgical scenes, characterized by high variability in different surgical activity scenarios and confused image features between targets and the background, presents challenges for surgical environment understanding. Traditional deep learning models often struggle with cross-activity interference, leading to suboptimal performance in each downstream task. To address this limitation, we explore multi-task learning, which utilizes the interrelated features between tasks to enhance overall task performance. In this paper, we propose EndoARSS, a novel multi-task learning framework specifically designed for endoscopy surgery activity recognition and semantic segmentation. Built upon the DINOv2 foundation model, our approach integrates Low-Rank Adaptation to facilitate efficient fine-tuning while incorporating Task Efficient Shared Low-Rank Adapters to mitigate gradient conflicts across diverse tasks. Additionally, we introduce the Spatially-Aware Multi-Scale Attention that enhances feature representation discrimination by enabling cross-spatial learning of global information. In order to evaluate the effectiveness of our framework, we present three novel datasets, MTLESD, MTLEndovis and MTLEndovis-Gen, tailored for endoscopic surgery scenarios with detailed annotations for both activity recognition and semantic segmentation tasks. Extensive experiments demonstrate that EndoARSS achieves remarkable performance across multiple benchmarks, significantly improving both accuracy and robustness in comparison to existing models. These results underscore the potential of EndoARSS to advance AI-driven endoscopic surgical systems, offering valuable insights for enhancing surgical safety and efficiency.