Computer-Aided Diagnosis Using Computerized BI-RADS Features

スポンサーリンク

概要

• 論文の詳細を見る

• Based on the descriptions of masses category in BI-RADS developed by American College of Radiology, dominant characteristics of a mass is quantified into ten computerized features. The regression of pathological result on these features is utilized to construct a computer-aided diagnosis system using binary logistic regression model. For constructed CAD system, the performance indices, accuracy, sensitivity, specificity, PPV, and NPV, are 91.7%, 91.4%, 91.9%, 85.1%, and 95.5% respectively and the diagnostic performance estimated by the Az value of ROC curve is 0.954. The angular and spiculated feature in margin class is the most important characteristic for differentiating abnormality from benign.

• 社団法人電子情報通信学会の論文
• 2007-01-20

著者

• Shen Wei-chih

  Department Of Computer Science And Information Engineering National Chung Cheng University

• Chang Ruey-feng

  Department Of Computer Science And Information Engineering Graduate Institute Of Biomedical Electron

• Chang Ruey-feng

  Department Of Computer Science And Information Engineering National Taiwan University

関連論文

• コンピュータによる全乳房連続超音波断層像の腫瘍存在診断
• コンピュータによる全乳房連続超音波断層像の腫瘍存在診断
• Computer-aided Diagnosis of Breast Elastography
• Computer-Aided Diagnosis Using Computerized BI-RADS Features
• Computer-aided diagnosis for 2D, 3D, and whole breast ultrasound (医用画像)
• Computer-aided Diagnosis for 2D, 3D, and Whole Breast Ultrasound(International Forum on Medical Imaging in Asia 2009 (IFMIA 2009))

スポンサーリンク