Towards Markerless Intraoperative Tracking of Deformable Spine Tissue

2025-06-30

AI Summary

This study addresses the inaccuracy and clinical translatability challenges in intraoperative soft-tissue deformation tracking during spinal surgery, which arises from overreliance on bony landmarks. We propose a markerless, real-time, low-cost soft-tissue dynamic tracking framework. Leveraging the first real-world clinical RGB-D dataset specifically acquired for spinal surgery, we develop SpineAlign—a robust anatomical registration system—and CorrespondNet—a multi-task deep learning architecture—jointly enabling preoperative-intraoperative alignment of key anatomical regions, soft-tissue deformation estimation, and pixel-level intraoperative segmentation. Our approach eliminates the need for invasive bone-anchored fiducials and operates solely with an off-the-shelf RGB-D camera, substantially reducing hardware dependency and procedural complexity. Experimental evaluation on authentic surgical data demonstrates sub-centimeter registration accuracy and robust real-time performance. The work establishes a clinically viable, generalizable technical pathway and foundational dataset for intraoperative spinal navigation.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Consumer-grade RGB-D imaging for intraoperative orthopedic tissue tracking is a promising method with high translational potential. Unlike bone-mounted tracking devices, markerless tracking can reduce operating time and complexity. However, its use has been limited to cadaveric studies. This paper introduces the first real-world clinical RGB-D dataset for spine surgery and develops SpineAlign, a system for capturing deformation between preoperative and intraoperative spine states. We also present an intraoperative segmentation network trained on this data and introduce CorrespondNet, a multi-task framework for predicting key regions for registration in both intraoperative and preoperative scenes.

Problem

• Markerless tracking of deformable spine tissue during surgery
• Developing RGB-D dataset and system for spine surgery
• Predicting key regions for registration in surgical scenes

Innovation

• Consumer-grade RGB-D imaging for spine tracking
• SpineAlign system captures deformation states
• CorrespondNet predicts key registration regions