Preoperative path planning algorithm for lung puncture biopsy based on path constraint and multidimensional space distance optimization

Abstract

The quality of the clinician-planned puncture path significantly affects the success rate of percutaneous lung biopsy procedures. Therefore, a preoperative path planning algorithm for lung puncture biopsy is proposed to assist clinicians in planning the needle path of lung biopsy. First, lung tissues and organs are segmented based on CT images. Then, a multiple puncture target point approach is proposed to screen the target locations of suspected tumors. And the puncture needle path is determined after selecting the entry points on the skin. According to lung biopsy clinical requirements, three hard constraints are proposed to remove paths that do not meet clinical requirements. Additionally, six soft constraints and the Pareto optimization based on multidimensional space distance are proposed to obtain the optimal puncture paths. The experimental data were collected from 33 patients with suspected pulmonary nodules in the Zigong First People’s Hospital. In the experiments, physicians labeled 98.46 % of the paths planned by the algorithm as clinically feasible, indicating that the proposed algorithm can be applied to nodule cases of different sizes and that it follows the clinical constraints of lung puncture. The average distance between the optimal algorithmic path and the surgical path was (5.77 ± 3.85) mm, which indicates that the proposed algorithm could simulate the surgical path according to the clinical guidelines of the physicians. The experimental results demonstrated that the proposed algorithm could provide an objective and effective path planning method for lung puncture biopsy of different nodule sizes.