Clinical application of computer-aided detection in breast cancers on mammograms
- VernacularTitle:计算机辅助检测在乳腺癌X线诊断中的应用
- Author:
Yong CUI
;
Renyou ZHAI
;
Xiaojuan LIU
;
Yajie WANG
- Publication Type:Journal Article
- Keywords:
Breast neoplasms;
Mammography;
Image interpretation, computer-assisted
- From:
Chinese Journal of Radiology
2001;0(09):-
- CountryChina
- Language:Chinese
-
Abstract:
Objective To evaluate the clinical value of computer-aided detection (CAD) system in the diagnosis of breast cancer on mammograms. Methods The mammograms of 136 pathology-proved breast cancer cases were analyzed by CAD system. The results of CAD were compared with that made by radiologists. Results The sensitivity, specificity, and accuracy of CAD system to microcalcification lesions were 93.2% (41/44), 22.8%(21/92), and 45.6%(62/136), respectively. The sensitivity, specificity, and accuracy of CAD system to mass lesions were 82.9%(107/129), 71.4%(5/7), and 82.4%(112/136), respectively. There was middle level agreement between the detection of CAD system and that of the radiologists (Kappa_(mass)=0.424, P_(mass)=0.000; Kappa_(cal)=0.365, P_(cal)=0.013). Conclusion CAD system has the characteristics of high sensitivity and low specificity. It could be the aid of radiologists in breast cancer detection on mammograms.
