Investigating Robot Control Policy Learning for Autonomous X-ray-guided Spine Procedures

2025-11-05

AI Summary

This study addresses the challenges of visual interpretation and robot control strategy design in multi-view X-ray-guided spinal puncture procedures. We propose a vision-only imitation learning framework that learns end-to-end open-loop control policies directly from real-time biplanar X-ray sequences. To overcome data scarcity, we develop a high-fidelity phantom-based simulator and synthesize a large-scale biplanar X-ray dataset. The method integrates dual-plane image processing, cross-domain transfer learning, and anatomically constrained trajectory modeling. It demonstrates strong generalization and robustness across complex anatomies and fracture cases. Experimental results show a 68.5% first-pass puncture success rate, consistent maintenance of safe pedicle-confined trajectories throughout insertion, and clinically interpretable path generation on real X-ray images. This work presents the first empirical validation of fully X-ray-driven robotic control for spinal intervention—without requiring preoperative image registration or additional intraoperative modalities.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Multi-modal Vision
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Imitation learning-based robot control policies are enjoying renewed interest in video-based robotics. However, it remains unclear whether this approach applies to X-ray-guided procedures, such as spine instrumentation. This is because interpretation of multi-view X-rays is complex. We examine opportunities and challenges for imitation policy learning in bi-plane-guided cannula insertion. We develop an in silico sandbox for scalable, automated simulation of X-ray-guided spine procedures with a high degree of realism. We curate a dataset of correct trajectories and corresponding bi-planar X-ray sequences that emulate the stepwise alignment of providers. We then train imitation learning policies for planning and open-loop control that iteratively align a cannula solely based on visual information. This precisely controlled setup offers insights into limitations and capabilities of this method. Our policy succeeded on the first attempt in 68.5% of cases, maintaining safe intra-pedicular trajectories across diverse vertebral levels. The policy generalized to complex anatomy, including fractures, and remained robust to varied initializations. Rollouts on real bi-planar X-rays further suggest that the model can produce plausible trajectories, despite training exclusively in simulation. While these preliminary results are promising, we also identify limitations, especially in entry point precision. Full closed-look control will require additional considerations around how to provide sufficiently frequent feedback. With more robust priors and domain knowledge, such models may provide a foundation for future efforts toward lightweight and CT-free robotic intra-operative spinal navigation.

Problem

• Developing imitation learning policies for X-ray-guided spine cannula insertion
• Creating realistic simulation environment for automated spinal procedure training
• Evaluating policy generalization across diverse vertebral anatomy and fractures

Innovation

• Simulation sandbox for X-ray-guided spine procedures
• Imitation learning policy for cannula trajectory planning
• Visual-based iterative alignment using bi-planar X-rays