SurgLAT: Surgical Latent Attention Tracking for Depth-Aware Robotic Laparoscope Control

2026-08-07

AI Summary

This work addresses the challenge of autonomous laparoscope control by interpreting surgeons’ implicit and time-varying operational intent in dynamic surgical environments, rather than tracking static targets. To this end, the authors propose SurgLAT, a novel framework that models implicit surgical attention as a causally evolving latent state. SurgLAT integrates memory-guided spatial priors with a dynamic retrieval mechanism to robustly track regions of interest. The method employs a frozen DINOv2 encoder, a state-conditioned spatial token mixer, a selective causal latent memory module, and a probabilistic attention heatmap decoder, combined with redundancy-aware null-space control respecting remote center of motion (RCM) constraints to ensure smooth and stable endoscope motion. Experiments demonstrate that the system exhibits strong robustness to occlusions, rapid movements, and target switches on both real surgical videos and robotic platforms.

Abstract

Autonomous laparoscopic camera control requires continuous understanding of the surgeon's operative intent in dynamic surgical scenes, where the target operative region is not a stable physical object but a latent and temporally evolving attention state. In this work, we present Surgical Latent Attention Tracking (SurgLAT), a causal online framework for latent surgical attention modeling and autonomous laparoscopic view control. SurgLAT uses a frozen DINOv3 encoder and a state-conditioned spatial token mixer to extract operative evidence under a memory-guided spatial prior, while a selective causal latent memory module jointly models short-term motion continuity and long-horizon surgical intent evolution through dynamic retrieval of current, recent, and historical latent states. The learned latent surgical attention state is decoded into a probabilistic attention heatmap and operative region for downstream endoscope guidance. Beyond perception, we further introduce a robotic deployment framework with explicit laparoscopic Remote Center of Motion (RCM) constrained control based on virtual-axis formulation, together with redundancy-aware null-space initialization for stable and smooth manipulator motion. We validate the full system on real laparoscopic surgical videos and a physical robotic laparoscope platform. Experimental results demonstrate robust online operative-region tracking and stable autonomous endoscopy adjustment under occlusion, rapid motion, and target transitions, highlighting the effectiveness of latent surgical intent modeling for surgical autonomy.

Problem

• surgical intent
• latent attention
• autonomous laparoscopy
• operative region tracking
• robotic endoscopy

Innovation

• latent attention tracking
• causal online modeling
• surgical intent estimation
• RCM-constrained control
• memory-guided perception