Improvement of a Computerized Automatic Classification Method of Digital Mammograms Based on Assessment of Fibroglandular Breast Tissue Density

概要

論文の詳細を見る
We have been developing an automated classification method based on the analysis of fibroglandular breast tissue density for digital mammograms. In this study, the definition method of points for approximation by second curve on thoracic wall side of Dmg region and that of threshold for extraction of Dc region based on the variance histogram analysis were improved. The agreement rate between physicians and the improved computerized method was increased from 80% to 91%. This result indicates that our improved method is more effective for the computerized classification of mammograms based on fibroglandular breast tissue density.
Medical Imaging and Information Sciencesの論文