DSTED: Decoupling Temporal Stabilization and Discriminative Enhancement for Surgical Workflow Recognition

2025-12-22

AI Summary

Surgical workflow recognition faces two key challenges: inter-frame prediction jitter and difficulty in distinguishing ambiguous procedural phases. To address these, we propose DSTED, a dual-path temporal enhancement framework. One path employs Reliable Memory Propagation (RMP) to model temporal stability and ensure consistency across frames; the other integrates Uncertainty-aware Prototype Retrieval (UPR) to enhance discriminability for ambiguous phases, coupled with a confidence-driven gating mechanism for adaptive feature fusion. This decoupled dual-path paradigm is the first of its kind for surgical phase recognition. Evaluated on the AutoLaparo-hysterectomy dataset, DSTED achieves 84.36% accuracy and 65.51% F1-score—significantly suppressing temporal jitter and improving robustness in phase transition detection compared to prior methods.

Technology Category

• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Purpose: Surgical workflow recognition enables context-aware assistance and skill assessment in computer-assisted interventions. Despite recent advances, current methods suffer from two critical challenges: prediction jitter across consecutive frames and poor discrimination of ambiguous phases. This paper aims to develop a stable framework by selectively propagating reliable historical information and explicitly modeling uncertainty for hard sample enhancement. Methods: We propose a dual-pathway framework DSTED with Reliable Memory Propagation (RMP) and Uncertainty-Aware Prototype Retrieval (UPR). RMP maintains temporal coherence by filtering and fusing high-confidence historical features through multi-criteria reliability assessment. UPR constructs learnable class-specific prototypes from high-uncertainty samples and performs adaptive prototype matching to refine ambiguous frame representations. Finally, a confidence-driven gate dynamically balances both pathways based on prediction certainty. Results: Our method achieves state-of-the-art performance on AutoLaparo-hysterectomy with 84.36% accuracy and 65.51% F1-score, surpassing the second-best method by 3.51% and 4.88% respectively. Ablations reveal complementary gains from RMP (2.19%) and UPR (1.93%), with synergistic effects when combined. Extensive analysis confirms substantial reduction in temporal jitter and marked improvement on challenging phase transitions. Conclusion: Our dual-pathway design introduces a novel paradigm for stable workflow recognition, demonstrating that decoupling the modeling of temporal consistency and phase ambiguity yields superior performance and clinical applicability.

Problem

• Reduces temporal jitter in surgical video recognition
• Improves discrimination of ambiguous surgical phases
• Enhances stability and accuracy for clinical workflow analysis

Innovation

• Reliable Memory Propagation filters high-confidence historical features
• Uncertainty-Aware Prototype Retrieval refines ambiguous frame representations
• Confidence-driven gate dynamically balances dual-pathway framework