肿瘤细胞膜免疫组化病理图像的自动分割  

Automatic Segmentation in Tumour Immunohistochemical Pathological Images

作　　者：林嘉雯[1] 刘勇[1] 刘秉瀚[1] 

机构地区：[1]福州大学数学与计算机科学学院,福建福州350002

出　　处：《莆田学院学报》2007年第5期45-47,50,共4页Journal of putian University

基　　金：福建省自然基金项目(A0610013);卫生部科学基金项目(WKJ2005-2-006);福州大学本科科研训练计划(05062)

摘　　要：针对肿瘤细胞膜着色的免疫组化病理图像中组织灰度在RGB颜色通道中的分布形态特点,采用阈值分割方法将阳性产物、细胞、胞浆、间质及空隙等各组织区域分离;针对细胞核和阳性目标的灰度存在重叠现象,依据区域的面积、圆形度等几何形态特征建立二级分割模型,进一步将细胞核和阳性产物区分开来。在上述基础上,提出并实现一种肿瘤免疫组化病理图像的自动分割方法。结果表明,此方法速度快,分割效果理想。To put forward a new segmentation method in tumour immunohistochemical pathological image, we set up the first segmentation model of gray distribution information and f＇md different coloration region according to the characteristics of tissue gray distribution in pathological images. Then the second segmentation model and separate different tissue （just as Positive Membrane and Nuclear） in same coloration region are established on the basis of the characteristics of geometry form in pathological images. Automatic segmentation of tumour immunohistochemical pathological images is achieved. The results attained can be as a ground work for further analysis of cancer.

关 键 词：免疫组化 病理图像 形态特征 二级分割 

分 类 号：R730.21[医药卫生—肿瘤]