Surgical Navigation System for Low-Dose-Rate Brachytherapy Based on Mixed Reality

This article presents a personalized mixed reality (MR) surgical assistance system for brachytherapy. Using a novel, modified multi-information fusion method, the fusion of virtual organs and a preoperative plan for an actual patient and the real-time tracking of surgical tools were achieved. Using the quaternion-based iterative closest point (QICP) algorithm and a hand-eye calibration method, the preoperative plan can be fused into individual patients. Using the electromagnetic (EM) tracker, users can track the surgery tools in real time, without multiple CT scans, and doctors can immediately perform the surgery. We performed a series of experiments, including phantom and animal experiments, to test the accuracy and efficiency of the system. In the phantom experiment, the average needle location error was 0.957 mm. Based on the results of animal experiments, the needle insertion error was 2.416 mm. All experimental results indicated that the procedure could be applied in further clinical studies.