RGA-Net: A Vision Enhancement Framework for Robotic Surgical Systems Using Reciprocal Attention Mechanisms

2026-02-14

AI Summary

This study addresses the critical challenge posed by surgical smoke, which severely degrades visibility in robotic endoscopic videos and compromises both surgical safety and human–robot interaction efficiency. To tackle this issue, the authors propose RGA-Net, a novel hierarchical encoder–decoder framework that integrates dual-stream hybrid attention, axis-decomposed attention, and reciprocal cross-gating mechanisms. By further incorporating shifted-window attention and frequency-domain processing, the model enables bidirectional modulation of encoder–decoder features for efficient smoke removal. Extensive experiments on the DesmokeData and LSD3K datasets demonstrate that the proposed method significantly enhances de-smoking performance and intraoperative visual quality, offering a robust technical solution to reduce surgeons’ cognitive load and mitigate procedural risks.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Adversarial Attacks & Robustness
• Natural Language Processing: Safety and Robustness

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Robotic surgical systems rely heavily on high-quality visual feedback for precise teleoperation; yet, surgical smoke from energy-based devices significantly degrades endoscopic video feeds, compromising the human-robot interface and surgical outcomes. This paper presents RGA-Net (Reciprocal Gating and Attention-fusion Network), a novel deep learning framework specifically designed for smoke removal in robotic surgery workflows. Our approach addresses the unique challenges of surgical smoke-including dense, non-homogeneous distribution and complex light scattering-through a hierarchical encoder-decoder architecture featuring two key innovations: (1) a Dual-Stream Hybrid Attention (DHA) module that combines shifted window attention with frequency-domain processing to capture both local surgical details and global illumination changes, and (2) an Axis-Decomposed Attention (ADA) module that efficiently processes multi-scale features through factorized attention mechanisms. These components are connected via reciprocal cross-gating blocks that enable bidirectional feature modulation between encoder and decoder pathways. Extensive experiments on the DesmokeData and LSD3K surgical datasets demonstrate that RGA-Net achieves superior performance in restoring visual clarity suitable for robotic surgery integration. Our method enhances the surgeon-robot interface by providing consistently clear visualization, laying a technical foundation for alleviating surgeons'cognitive burden, optimizing operation workflows, and reducing iatrogenic injury risks in minimally invasive procedures. These practical benefits could be further validated through future clinical trials involving surgeon usability assessments. The proposed framework represents a significant step toward more reliable and safer robotic surgical systems through computational vision enhancement.

Problem

• surgical smoke
• visual degradation
• robotic surgery
• endoscopic video
• human-robot interface

Innovation

• Reciprocal Attention
• Smoke Removal
• Robotic Surgery
• Dual-Stream Hybrid Attention
• Axis-Decomposed Attention