ArthroCut: Autonomous Policy Learning for Robotic Bone Resection in Knee Arthroplasty

TL;DR

ArthroCut is an autonomous robotic framework for knee bone resection in arthroplasty, integrating multimodal data and token-based decision-making to improve safety, reliability, and interpretability in surgical automation.

Contribution

The paper introduces ArthroCut, a novel autonomous policy learning method that combines multimodal data and token-based encoding for context-aware surgical robot control.

Findings

Achieves 86% success rate in bench-top knee resections

Outperforms baseline methods in reliability and stability

Demonstrates effective integration of preoperative and intraoperative data

Abstract

Despite rapid commercialization of surgical robots, their autonomy and real-time decision-making remain limited in practice. To address this gap, we propose ArthroCut, an autonomous policy learning framework that upgrades knee arthroplasty robots from assistive execution to context-aware action generation. ArthroCut fine-tunes a Qwen--VL backbone on a self-built, time-synchronized multimodal dataset from 21 complete cases (23,205 RGB--D pairs), integrating preoperative CT/MR, intraoperative NDI tracking of bones and end effector, RGB--D surgical video, robot state, and textual intent. The method operates on two complementary token families -- Preoperative Imaging Tokens (PIT) to encode patient-specific anatomy and planned resection planes, and Time-Aligned Surgical Tokens (TAST) to fuse real-time visual, geometric, and kinematic evidence -- and emits an interpretable action grammar under grammar/safety-constrained decoding. In bench-top experiments on a knee prosthesis across seven trials, ArthroCut achieves an average success rate of 86% over the six standard resections, significantly outperforming strong baselines trained under the same protocol. Ablations show that TAST is the principal driver of reliability while PIT provides essential anatomical grounding, and their combination yields the most stable multi-plane execution. These results indicate that aligning preoperative geometry with time-aligned intraoperative perception and translating that alignment into tokenized, constrained actions is an effective path toward robust, interpretable autonomy in orthopedic robotic surgery.