DBT-Bleed: Dual-Branch Temporal Modeling with Key-Frame Selection for Surgical Bleeding Detection

2026-06-22

AI Summary

This study addresses the challenge of distinguishing intraoperative adverse bleeding events from visually similar residual blood in long-duration surgical videos, where existing methods struggle to balance computational efficiency with fine-grained temporal modeling. To this end, the authors propose a dual-branch multi-scale temporal modeling framework that integrates a hierarchical entropy-driven keyframe selection mechanism (HiRED) with temporal adapters, effectively decoupling representations of bleeding and normal states while preserving discriminative temporal information and eliminating redundancy. The work introduces EndoPit-IAE, the first annotated neurosurgical dataset for intraoperative adverse events, and demonstrates significant performance gains: on MultiBypass, it achieves absolute improvements of 6.53% in F1, 5.62% in Recall, and 9% in MCC; under cross-procedure zero-shot settings, F1 and MCC improve by 6% and 8%, respectively, substantially outperforming current approaches.

Abstract

Intraoperative Adverse Events (IAEs) detection is critical for improving surgical safety, with bleeding being among the most frequent events across many surgery types. Existing methods struggle to distinguish bleeding IAE from visually similar residual blood due to limited temporal reasoning. Moreover, modeling long surgical videos while preserving fine-grained temporal dynamics remains computationally challenging. We propose DBT-Bleed, a dual-branch multi-scale temporal modeling framework disentangling bleeding and normal representations using layer-wise temporal adapters for short- and long-term bleeding progression. To efficiently process long surgical videos without sacrificing fine-grained temporal information, we introduce HiRED, a Hierarchical Entropy-Driven frame selection strategy that retains temporally informative segments while removing redundancy. Experiments on the MultiBypass dataset demonstrate gains of 6.53% in F1, 5.62% in Recall and 9% in MCC values for bleeding IAE detection, consistently outperforming video-level baselines. Additionally, we evaluate cross-procedure generalization on a newly curated dataset from a different surgical procedure type, where DBT-Bleed demonstrates robust transferability by achieving gain of 6% in F1 and 8% in MCC under zero-shot setting. To support this evaluation, we introduce EndoPit-IAE, an Endonasal Pituitary Surgery dataset annotated for IAEs, representing the first IAE-annotated dataset in neurosurgery. Code will be made publicly available upon acceptance.

Problem

• Surgical Bleeding Detection
• Intraoperative Adverse Events
• Temporal Modeling
• Long Video Processing
• Key-Frame Selection

Innovation

• Dual-Branch Temporal Modeling
• Key-Frame Selection
• Hierarchical Entropy-Driven
• Surgical Bleeding Detection
• Cross-Procedure Generalization