Deep reinforcement learning for guidewire navigation in coronary artery phantom

TL;DR

This paper presents a deep reinforcement learning framework for autonomous guidewire navigation in coronary artery phantom models, improving precision and efficiency in robot-assisted cardiac interventions.

Contribution

It introduces a novel RL approach using Rainbow with human demonstrations and subgoals for guidewire navigation, advancing automation in minimally invasive cardiac procedures.

Findings

RL agent successfully navigates to all targets in the phantom model

Enhanced training efficiency through transfer learning and demonstrations

Framework demonstrates potential for automation in robot-assisted interventions

Abstract

In percutaneous intervention for treatment of coronary plaques, guidewire navigation is a primary procedure for stent delivery. Steering a flexible guidewire within coronary arteries requires considerable training, and the non-linearity between the control operation and the movement of the guidewire makes precise manipulation difficult. Here, we introduce a deep reinforcement learning(RL) framework for autonomous guidewire navigation in a robot-assisted coronary intervention. Using Rainbow, a segment-wise learning approach is applied to determine how best to accelerate training using human demonstrations with deep Q-learning from demonstrations (DQfD), transfer learning, and weight initialization. `State' for RL is customized as a focus window near the guidewire tip, and subgoals are placed to mitigate a sparse reward problem. The RL agent improves performance, eventually enabling the guidewire to reach all valid targets in `stable' phase. Our framework opens anew direction in the automation of robot-assisted intervention, providing guidance on RL in physical spaces involving mechanical fatigue.