Robot-Assisted TMS with Enhanced Positioning Accuracy and Efficiency: A Hybrid Hand-Eye Calibration Framework

Co-optimizing the positioning accuracy and operational efficiency of robot-assisted transcranial magnetic stimulation (TMS) systems is a key challenge in clinical applications. In this paper, a hybrid hand-eye calibration strategy that incorporates robot base-camera offline global calibration (BCGC) and end-coil precision calibration (ECPC) is presented. It employs a hierarchical motion control approach based on the different stages of the treatment process to achieve progressive control from wide-range fast localization to sub-millimeter precision adjustment. The performance of the method was validated by registration error (RE) and total system error (TSE) evaluations. The results show that BCGC provides good repeatability in the initial positioning phase with an average FRE of 2.93 mm. In comparison, the ECPC method reduces the RE to 0.46 mm for fine positioning, significantly improving accuracy. Measured by a coordinate measuring machine (CMM), the overall system error is 0.997 mm, confirming the high accuracy of the system in practical applications. This hybrid approach improves the positioning accuracy of the TMS coil while maintaining the stability and flexibility of the system, providing a reliable solution for neuronal navigation and precise regulation of brain functions.