新疆地方性肝包虫病CT图像的灰度共生矩阵分析  

Feature Analysis of Endemic Liver Hydatid CT Images in Xinjiang with Gray Level Co-occrence Matrix

作　　者：木拉提·哈米提[1] 李莉[1] 艾克热木·阿西木[2] 孙静[1] 周晶晶[1] 毕雪华[1] 宫新芳[1] 

机构地区：[1]新疆医科大学医学工程技术学院,乌鲁木齐830011 [2]新疆医科大学网控中心,乌鲁木齐830011

出　　处：《科技通报》2012年第3期75-80,共6页Bulletin of Science and Technology

基　　金：国家自然科学基金资助项目(No.30960097);新疆医科大学科研创新基金(XJC201089);新疆医科大学科研创新基金(2007-16)

摘　　要：采用灰度共生矩阵法对新疆地方性肝包虫CT图像进行了统计纹理特征提取,并对各个特征参数进行了计算和分析。结果表明,肝包虫病CT图像在不同方向的相关性、熵、对比度与正常肝脏相比较具有显著的差异,能有效的区分单囊型肝包虫CT图像和多囊型肝包虫CT图像,该项研究可为新疆地方性肝包虫病的临床诊断提供依据。In this paper,extract the statistical texture feature of Xinjiang local Liver hydatid CT images using Gray level co-occurrence matrix ,and various parameters are calculated and analyzed. The results show that correlation ,entropy, contrast of CT images of liver hydatid disease in different directions,have significant differences compared with normal liver,which can effectively distinguish mono-hydatid cyst and multiple daughter hydatid cyst CT images,The study provide the basis for hepatic hydatid disease in Xinjiang local clinical diagnosis.

关 键 词：纹理分析 灰度共生矩阵 肝包虫 

分 类 号：R319[医药卫生—基础医学]