图像引导下肝脏形变矫正仿真

Image-guided Liver Deformation Correction Simulation

为解决腹腔镜肝切除手术中气腹压引起的肝脏形变问题,提出一种在图像引导下对肝脏形变进行矫正的流程。首先,对术前肝脏CT图像分割并网格划分成有限元网格,根据生物力学参数生成生物力学模型;其次,提取术中状态下肝脏表面与术前生物力学模型表面做基于ICP算法的初始配准,将配准结果作为生物力学边界条件求解变形场;然后,对于采集的术前CT图像与术中CT图像使用B样条算法进行非刚性配准,将得到配准后的变形场作为金标准;最后使用变形场作为约束条件求解力学方程,驱动生物力学模型变形。使用真实实验获取的七头猪的肝脏数据作为实验数据进行实验研究,结果表明上述方法在肝脏形变矫正问题的精确性和鲁棒性上具有一定的优越性。

A procedure of image-guided liver deformation correction under pneumoperitoneum is proposed to deal with the liver deformation problem caused by gas injection.First,we segmented the CT image of the liver and obtain the biomechanical model after mesh generation.Second,we extracted the intra operative liver surface and did registration with surface of the pre-operative biomechanical model based on Iterative Closest Point algorithm.The registration result was served as our boundary condition to solve deformation vector field.Then,we did non-rigid registration based on B spline algorithm to get deformation field as golden standard.At last,we used the deformation field to solve the mechanical equation,driving our model to deform.The liver data from real pig were used as the experimental data to carry the research and the results show that the method has certain advantages of the accuracy and robustness in the problem of liver deformation correction.