A method of three-dimensional non-rigid localization of liver tumors based on structured light

In this paper, a new localization method of liver tumors based on structured light is proposed to achieve the accurate resection of liver tumors. Firstly, the preoperative liver model and intraoperative liver surface are completed respectively based on CT sections and structured light. Secondly, the 3D reconstruction of liver surface based on structured light is used as the reference surface for registration. A common ICP algorithm and an improved ICP algorithm based on statistics are proposed to complete the rough and fine registration of liver surface. In this paper, the theory of PBD is introduced, and the registration results are used to guide the preoperative deformation of the liver model. Finally, the method of 3D non-rigid localization of liver tumors proposed in this paper is verified by experiments. In addition, the effectiveness of the proposed method is compared with that of the tumor localization obtained by the finite element analysis and Coherent point drift methods. The experimental results show that the average localization error of liver tumor is 1.24 mm. The time spent during the entire process of tumor localization is kept under fifteen seconds. The method presented in this paper meets the needs of surgeons in the intraoperative process.