高技术通讯2011，Vol.21Issue(11)：1164-1170,7.DOI:10.3772/j.issn.1002-0470.2011.11.009
改进的肝脏软组织分割算法及实时绘制
An improved method for medical liver segmentation and real-time rendering
摘要
Abstract
This paper presents a medical CT image segmentation algorithm based on the Graph Cut algorithm for separating the liver from other soft tissues, and realizes the effective real-time rendering of soft tissue using the open computing language (OpenCL). The average segmentation accuracy of the improved Graph Cut algorithm can be increased to 96. 2% , and the Octree-based RayCasting algorithm improved using the OpenCL can greatly raise the parallel rendering speed, which makes it possible to implement a real-time cross-platform virtual surgery system. Moreover, this method can make doctor's diagnosis more accurate, so it has a very good practical prospect.关键词
Graph Cut算法/OpenCL语言/医学肝脏分割/RayCasting算法/八叉树/实时绘制Key words
Graph Cut/open computing language (OpenCL)/medical liver segmentation/RayCasting/Octree/real-time rendering引用本文复制引用
康飞龙,杨杰..改进的肝脏软组织分割算法及实时绘制[J].高技术通讯,2011,21(11):1164-1170,7.基金项目
国家自然科学基金(31100672)和上海交大医工合作基金资助项目. （31100672）
