Sample-Efficient Learning with Online Expert Correction for Autonomous Catheter Steering in Endovascular Bifurcation Navigation

2026-02-23

AI Summary

This work proposes a sample-efficient reinforcement learning framework incorporating online expert corrections to address the challenges of sparse rewards and reliance on static vascular models in autonomous catheter navigation, which typically lead to poor sample efficiency and limited generalization. The approach uniquely integrates segmentation-driven real-time pose estimation, a fuzzy control-based directional adjustment mechanism, and a structured reward shaped by expert priors, substantially enhancing navigation performance at vascular bifurcations. Built upon the Soft Actor-Critic algorithm, the method achieves superior efficiency and robustness: in transparent vascular phantom experiments, it requires only 123 training episodes—25.9% fewer than the baseline—and reduces the average positional error to 83.8% of the baseline level.

Technology Category

• Intelligent Robots: State Estimation
• Search and Optimization: Sampling/Simulation-based Search
• Reasoning under Uncertainty: Sequential Decision Making

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• 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

Robot-assisted endovascular intervention offers a safe and effective solution for remote catheter manipulation, reducing radiation exposure while enabling precise navigation. Reinforcement learning (RL) has recently emerged as a promising approach for autonomous catheter steering; however, conventional methods suffer from sparse reward design and reliance on static vascular models, limiting their sample efficiency and generalization to intraoperative variations. To overcome these challenges, this paper introduces a sample-efficient RL framework with online expert correction for autonomous catheter steering in endovascular bifurcation navigation. The proposed framework integrates three key components: (1) A segmentation-based pose estimation module for accurate real-time state feedback, (2) A fuzzy controller for bifurcation-aware orientation adjustment, and (3) A structured reward generator incorporating expert priors to guide policy learning. By leveraging online expert correction, the framework reduces exploration inefficiency and enhances policy robustness in complex vascular structures. Experimental validation on a robotic platform using a transparent vascular phantom demonstrates that the proposed approach achieves convergence in 123 training episodes -- a 25.9% reduction compared to the baseline Soft Actor-Critic (SAC) algorithm -- while reducing average positional error to 83.8% of the baseline. These results indicate that combining sample-efficient RL with online expert correction enables reliable and accurate catheter steering, particularly in anatomically challenging bifurcation scenarios critical for endovascular navigation.

Problem

• sample-efficient learning
• autonomous catheter steering
• endovascular bifurcation navigation
• reinforcement learning
• online expert correction

Innovation

• sample-efficient reinforcement learning
• online expert correction
• autonomous catheter steering
• bifurcation navigation
• structured reward design