Space Registration and Experiments for Electromagnetic Navigation in Bronchoscopic Surgery

Space registration is a fundamental and critical component in image-guided surgery and navigation systems. It permits the precise alignment of preoperative imaging data with the patient's anatomy and enables the accurate display of the position of surgical instruments within the patient's body. We propose a markers-based space registration framework which utilizes a quaternion-based iterative closest point (ICP) algorithm to investigate the mapping relationships between the patient space and the virtual image space. The performance of the proposed approach is evaluated through model-based and in vivo porcine lung experiments. In a static experiment, the registration error ranged from 0.8 mm to 1.05 mm, while in a dynamic experiment, the average error is 1.14 ± 0.16 mm. Furthermore, in vivo experiments with pig lungs, the navigation average errors 1.95 ± 0.24 mm. The experimental results demonstrate that our method is effective and feasible with potential clinical applications.