Suppr超能文献

细胞核形态测量法在结直肠腺瘤性息肉发育异常评估中的应用:一项初步研究

Cytonuclear morphometry in the assessment of dysplasia in colorectal adenomatous polyps. A pilot study.

作者信息

Meijer G A, Baak J P

机构信息

Institute for Pathology, Free University Hospital, Amsterdam, The Netherlands.

出版信息

Pathol Res Pract. 1992 Feb;188(1-2):148-56. doi: 10.1016/s0344-0338(11)81172-6.

Abstract

In order to objectify the degree of dysplasia, results of nuclear and cellular morphometry were compared with assessed grades of dysplasia in 32 colorectal adenomatous polyps. Of these adenomas 8 showed mild, 17 moderate and 7 severe dysplasia (according to blind duplicate assessments by two pathologists). Using a microscope video-overlay interactive digitizing measurement system, 100 nuclei were measured in each specimen according to a strict measurement protocol. In a stepwise discriminant analysis the best discriminating features appeared to be mean stratification index, N/C ratio, mean contour ratio and mean form AR. Overall with these features 65.6% correct jackknifed classification of the 32 cases could be achieved. On the other hand, a clear three-group distinction could not be obtained, even with the best set of discriminating features. Similarly to gastric dysplasia, the morphometric features might be used to design a two-grade (low, high) rather than a three-grade system to assess the degree of dysplasia. These findings make clear that objectifying the degree of dysplasia in adenomatous polyps is possible by means of interactive morphometric analysis.

摘要

为了客观量化发育异常程度,将32例大肠腺瘤性息肉的细胞核和细胞形态测量结果与评估的发育异常分级进行了比较。在这些腺瘤中,8例显示轻度发育异常,17例显示中度发育异常,7例显示重度发育异常(根据两位病理学家的双盲重复评估)。使用显微镜视频叠加交互式数字化测量系统,根据严格的测量方案,在每个标本中测量100个细胞核。在逐步判别分析中,最佳判别特征似乎是平均分层指数、核质比、平均轮廓比和平均形态AR。总体而言,利用这些特征,32例病例的留一法交叉验证正确分类率可达65.6%。另一方面,即使使用最佳的判别特征集,也无法获得明确的三组区分。与胃发育异常类似,形态测量特征可用于设计一个两级(低、高)而非三级系统来评估发育异常程度。这些发现表明,通过交互式形态测量分析可以客观量化腺瘤性息肉的发育异常程度。

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验