红细胞形态特征的计算机图像处理研究(英文)  

Study on erythrocyte morphological features by computerized image processing

作　　者：郝葆青[1] 尹光富[2] 秦天莺[1] 王燕[1] 

机构地区：[1]西南民族大学生命科学与技术学院,成都610041 [2]四川大学生物医学工程中心,成都610065

出　　处：《西南民族大学学报（自然科学版）》2005年第3期390-395,共6页Journal of Southwest Minzu University(Natural Science Edition)

基　　金：国家自然科学基金(No.39870232).

摘　　要：比较传统人工细胞形态观察和定量的操作方式,介绍了基于计算机图像处理和分析技术的有关红细胞形态特征定量的研究成果.并且展现了一套适用范围广、可作为红细胞图像的形态描述或细胞分类的图形特征参数--圆周率.同时,基于圆周率提供了一些较为适用的细胞形态特征值的算法(比如,多定位的图像获取、红细胞胞体形态的自动识别,以及对其进行的分类算法).红细胞形态的算法或程序都基于VC++,不过该程序也可运用于其他用途.红细胞的染色采用瑞氏法(亚甲基蓝-伊红).介绍的红细胞形态计算机处理方法可提供有效的半自动红细胞形态特征的统计分析,可描述出群体细胞形态特征的分布曲线或不同细胞形状比率的数据.In this article, computer-based image processing and analysis techniques are studied for quantitative analysis of the erythrocyte morphological features, and the performance is compared with traditional non-image processing methods. A wide set of specific morphological parameters are defined, the circularity ratio. That is a description of the erythrocytes morphological features is used for the shape identifying and classification of erythrocytes. Some significant algorithms have been implemented (i.e. one for multiple focus image acquisition, automatic erythrocyte body shape recognition, and classification), based on the cell circularity ratio. The algorithms of erythrocytes image segmentation, cell body's shape recognition and measurement can be implemented through the VC++, but the algorithms and the methods can be employed elsewhere. Bodies of erythrocytes are stained with Wright-Stain (Methylene blue + Eosin). The present method can provide efficiently and semi-automatically the morphological statistical analysis of erythrocyte, and can give the distribution of erythrocyte morphological features or the proportion data of different cell bodies' shapes.

关 键 词：红细胞 图像处理 圆周率 分割 识别 

分 类 号：R318[医药卫生—生物医学工程]