Automatic Segmentation of Breast Lesions in Dynamic Contrast-Enhanced MR Images

Keywords

achieves approaches assisted becoming benign boundary breast build called chest class classifier classifiers clinical combining competing complex computational computing confirmed construct contrast dataset datasets depicted develop diagnosis ding dynamic efficacy empirical enhancement equal evaluated examinations extracted extracting fall feature features field fitted form forward gists growing help herein homepage identified involved kale labeled last least lesion lesions linear logistic machine malignant manually matrix medical medicine model models near necessary needs neighborhood numb occurrence open operating parametric peak pooled post preferably probability processing produce proposed quantitative question radiologist random randomly rapid recent remains resolution review sampled scientific score segmentation segmented selected selection series several short simple slice slices slop slope smallest society spacing spatial speed squares status stratified suffices support surprising suspicious suspiciously suspiciousness system systems table task technology temporal terms tissue train trained training truncate turn types validation variables variation vector versus volumes wall window wise women years yield yields