节点文献

计算机辅助显微镜评估乳腺癌Ki-67指数的一致性探究

Consistency of computer-aided assessment for Ki-67 index in breast cancer

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 吴小艳陈杰傅静向旭辉石乐飞步宏

【Author】 WU Xiao-yan;CHEN Jie;FU Jing;XIANG Xu-hui;SHI Le-fei;BU Hong;Department of Pathology, West China Hospital, Sichuan University;Institute of Clinical Pathology, West China Hospital, Sichuan University;Department of Pathology, Sichuan Provincial People’s Hospital;

【通讯作者】 步宏;

【机构】 四川大学华西医院病理科四川大学华西医院临床病理研究所四川省人民医院病理科

【摘要】 目的 探索在计算机辅助显微镜下评估Ki-67指数的方法以提高医生间的一致性。方法 医生用显微镜下平均值法和计算机辅助显微镜下多视野评估100例浸润性乳腺癌的Ki-67指数,计算ICC分析一致性。结果 显微镜下3位医生的ICC为0.836;计算机辅助下3位医生评估视野数为1、3、5、7时ICC分别为0.743、0.878、0.917和0.934,一致性随视野数增加而递增。在同、异质性,组织学分级等不同分类中,当视野数≥5时,计算机辅助下医生对Ki-67指数评估的一致性均高于显微镜下医生间的一致性。结论 计算机辅助显微镜评估Ki-67指数的方法能提高医生间的一致性。该方法临床可操作性强,有望早日应用于病理医生常规工作中。

【Abstract】 Objective To propose a method of computer-aided assessment for Ki-67 index to improve inter-physician consistency. Methods A total of 100 cases of invasive breast cancer were evaluated by two methods: the average method under microscope and multi-field assessment for computer-aided. Intraclass correlation coefficient(ICC) was calculated for analysis consistency. Results The ICC of three doctors was 0.836. The ICC of the doctors with computer-aided were 0.743, 0.878, 0.917 and 0.934 at 1, 3, 5 and 7 fields, respectively. It showed that the more fields, the better consistency. For different classification, including homogeneity or heterogeneity, histological grade and so on, when the number of assessed fields ≥ 5, the consistency of the computer-aided doctors was higher than that of doctors under the microscope. Conclusion The method of computer-aided assessment for Ki-67 index can improve the consistency among doctors. The method has strongly clinical operability and is expected to be applied in the routine work of pathologists as soon as possible.

【关键词】 乳腺癌计算机辅助Ki-67指数一致性
【Key words】 Breast cancerComputer-aidedKi-67 indexConsistency
【基金】 四川大学华西医院学科卓越发展1·3·5工程项目(ZYGD18012);成都市新型产业技术研究院科技创新项目(2017-CY02-00026-GX)
  • 【文献出处】 诊断病理学杂志 ,Chinese Journal of Diagnostic Pathology , 编辑部邮箱 ,2023年02期
  • 【分类号】R737.9
  • 【下载频次】9
节点文献中: