Ipsilateral-mammogram computer-aided detection of breast cancer.

摘要: In this paper, an ipsilateral multi-view computer-aided detection (CAD) scheme is presented for mass in digital mammograms by exploiting correlative information of suspicious lesions between the same breast. After nonlinear tree-structured filtering image noise suppression, two wavelet-based methods, directional wavelet transform and enhancement, adaptive fuzzy C-means algorithm segmentation are employed on each breast, respectively, concurrent analysis developed iterative inter-projective feature matching analysis. A supervised artificial neural network as a classifier, which back-propagation combined with Kalman used training algorithm, free-response receiver operating characteristic to test performance unilateral CAD system. Performance comparison has been conducted final system our previously single-mammogram-based The study results demonstrate advantages method over current single-view false positive reduction.