利用图像处理技术对胃管状肿瘤结构异常进行形态计量分析。
Morphometrical analysis of structural abnormality of tubular tumors of the stomach with image processing.
作者信息
Ishido T, Yamaguchi H, Yoshida S, Tonouchi S, Nakamura K
机构信息
Division of Internal Medicine, National Cancer Center Hospital, Tokyo.
出版信息
Jpn J Cancer Res. 1991 Sep;82(9):1015-21. doi: 10.1111/j.1349-7006.1991.tb01937.x.
Four indices, index of tubular density, index of dispersion of tubular size, degree of complexity of tubular shape and frequency of complex-shaped tubuli, were defined to quantify the structural abnormalities of gastric tumors in morphometrical analysis by image processing. The values of each index corresponded well with the degree of each structural abnormality found pathologically. These indices were considered to be valid for representing their respective structural factors. There were significant differences among the mean values of the scores calculated by a formula using all 4 indices of benign tubular adenoma, borderline lesion and well-differentiated type tubular adenocarcinoma (P less than 0.05). Therefore, the discriminant formula was considered to be valid for integrating these indices and for representing structural abnormality of gastric tubular tumors.
定义了四个指标，即肾小管密度指数、肾小管大小离散指数、肾小管形状复杂程度以及复杂形状肾小管的频率，通过图像处理在形态计量分析中量化胃肿瘤的结构异常。每个指标的值与病理发现的每种结构异常程度高度对应。这些指标被认为对代表其各自的结构因素是有效的。使用良性管状腺瘤、交界性病变和高分化型管状腺癌的所有4个指标通过公式计算的得分平均值之间存在显著差异（P小于0.05）。因此，判别公式被认为对于整合这些指标以及代表胃管状肿瘤的结构异常是有效的。
Zotero 插件