Background-incorporated volumetric model for patient-specific surgical simulation: a segmentation-free, modeling-free framework
Background-incorporated volumetric model for patient-specific surgical simulation: a segmentation-free, modeling-free framework
复制标题
用于患者特定手术模拟的背景合并体积模型：无分割、无建模框架
DOI：
10.1007/s11548-010-0456-1
复制
发表时间：
2010
影响因子：
3
通讯作者：
K. Minato
中科院分区：
文献类型：
--
作者：
K. W. Hung;M. Nakao;Koji Yoshimura;K. Minato
目的针对特定患者的手术模拟带来了实际和技术上的挑战。提出了一种无需分割、无需建模的框架，用于在特定患者的手术模拟中创建直观的体变形和操作的医学体模型。方法基于一种新的网格结构--体代理网格(Volume Proxy Mesh，VPM)创建体模型。模型的生成分为两个阶段：顶点放置阶段和网格改进阶段。通过基于曲率的顶点放置方法将VPM的顶点分配到初始位置，然后使用粒子群优化算法(PSO)进行网格改进。使用该框架，所得到的模型被紧密地定制为数据集的详细特征。此外，生成的VPM网格可以支持操纵器官与其周围组织之间更广泛的频谱变形。在网格质量方面的进展也表明粒子群算法用于网格改进是可行的。结论该框架应用于多个肾脏CT体积数据集。使用该框架，所得到的模型被紧密地定制为数据集的详细特征。此外，生成的VPM网格可以支持操纵器官与其周围组织之间更广泛的频谱变形。对最终网格质量的评估表明，粒子群算法用于网格改进是可行的。
PurposePatient-specific surgical simulation imposes both practical and technical challenges. We propose a segmentation-free, modeling-free framework that creates medical volumetric models for intuitive volume deformation and manipulation in patient-specific surgical simulation.MethodsThe proposed framework creates a volumetric model based upon a new form of mesh structure, a Volume Proxy Mesh (VPM). The model can be generated in two phases: the vertex placement phase and mesh improvement phase. Vertices of a VPM are assigned to an initial location by curvature-based vertex placement method, and followed by mesh improvement performed by Particle Swarm Optimization (PSO).ResultsThe framework is applied to several kidney CT volume data. Using the framework, the resulting models are closely tailored to the detailed features of the datasets. Moreover, the resulting VPM meshes can support broader spectrum deformation between the manipulated organ and its surrounding tissues. Progress in the mesh quality of the final mesh also shows that PSO is feasible for mesh improvement.ConclusionThe framework was applied to several kidney CT volume datasets. Using the framework, the resulting models are closely tailored to the detailed features of the datasets. Moreover, the resulting VPM meshes can support broader spectrum deformation between the manipulated organ and its surrounding tissues. Evaluation of final mesh quality shows that PSO is feasible for mesh improvement.