Human-Inspired Framework for Robotic Craniotomy: Integrating Multimodal Fusion and Adaptive Trajectory Adjustment

2026-07-23

AI Summary

This study addresses the limitations of existing open-loop robotic craniotomy systems, which rely solely on preoperative imaging and are vulnerable to intraoperative registration errors and bone deformation, posing risks of dural injury. To enhance safety, this work proposes a human-inspired closed-loop autonomous craniotomy framework that integrates preoperative planning with intraoperative execution through adaptive trajectory adjustment. The core innovations include a novel force-acoustic multimodal sensing mechanism combined with a CMA-TCN-Transformer network and adaptive Bayesian filtering for high-precision, real-time detection of bone penetration onset. Additionally, an adaptive dual-contour fusion algorithm and an in-situ projected trajectory correction strategy enable dynamic path optimization. Experimental results demonstrate 97% prediction accuracy on bovine ribs, a detection latency of 0.048 ± 0.097 seconds, and a maximum overshoot of only 0.29 mm; notably, no dural injuries occurred in four ex vivo human skull trials.

Abstract

Manual craniotomy is a high-risk, skill-dependent procedure associated with surgeon fatigue and potential dural injury. While robotic approaches have improved safety, existing open-loop systems rely solely on preoperative images and cannot compensate for intraoperative registration errors or tissue deformation. To address this, we propose a human-inspired closed-loop robotic craniotomy framework that intelligently integrates preoperative planning with intraoperative execution. An adaptive dual-contour fusion algorithm is employed to generate trajectories that conform to complex cranial geometries while maintaining a consistent tool-bone relative pose. For intraoperative perception, a multimodal two-stage cross-modal attention block (CMA)-temporal convolutional network (TCN)-Transformer network combined with an adaptive Bayesian filter fuses force and acoustic signals to achieve robust breakthrough detection under varying bone conditions. Upon detection, an in-situ projection-based trajectory adjustment strategy dynamically compensates for depth deviations, enabling safe residual bone isolation. Experiments on bovine ribs show a breakthrough prediction accuracy of 97%, a detection latency of 0.048 +/- 0.097 s, and a maximum overshoot of 0.29 mm. All four ex vivo cranial experiments were successfully completed without dural injury. These results demonstrate that the proposed cybernetic framework enables safe and autonomous craniotomy with highly effective closed-loop control.

Problem

• robotic craniotomy
• intraoperative registration errors
• tissue deformation
• closed-loop control
• breakthrough detection

Innovation

• closed-loop robotic craniotomy
• multimodal fusion
• adaptive trajectory adjustment
• breakthrough detection
• dual-contour fusion