基于移动最小二乘法的无网格肝脏形变研究  

Research of Liver Deformation Based on Moving Least Square of Meshless Method

作　　者：许少剑 陈国栋[1] 

机构地区：[1]福州大学物理与信息工程学院,福建福州350000

出　　处：《计算机仿真》2015年第4期408-413,共6页Computer Simulation

基　　金：福建省自然科学基金(2013J05090);福建省科技计划重点项目(2011H0027)

摘　　要：在手术虚拟仿真对软组织形变研究中,采用传统的网格形变方法,由于出现网格畸变、网格重构会影响形变效果的问题。针对手术虚拟仿真要求尽可能逼真地模拟形变以达到良好的真实感,提出了一种基于移动最小二乘法的无网格方法对以肝脏为对象的软组织进行形变研究,考虑肝脏软组织生物力学特性,采用Voigt模型对肝脏软组织进行物理建模。同时,针对无网格肝脏模型的碰撞检测,提出AABB包围盒与八叉树空间分割相结合的碰撞检测算法,在肝脏模型体节点的合理布局基础上可以对碰撞进行有效准确的检测。实验结果表明,无网络方法实现简单且具有较好的形变真实感,对虚拟手术中软组织形变研究具有重要的指导意义。This paper proposed a meshless method based on moving least square method to study the soft tissue deformation and selected liver as the object. Taking into account the biomechanical properties of liver soft tissue, the Voigt model was selected to make the physical model of liver. Meanwhile, for the collision detection of meshlcss liver model, we proposed a collision detection algorithm based on the AABB bounding volume and the octree spatial subdi- vision, and the collision detection is effective and accurate on the basis of rational placement of volumetric node. The experimental results show that the method of deformation is simple and has better realistic, also has important guiding significance to the study on soft tissue deformation in virtual surgery.

关 键 词：肝脏形变 无网格法 移动最小乘法 物理建模 碰撞检测 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]