Fig. 9: The proposed integrated framework.
From: Connected-UNets: a deep learning architecture for breast mass segmentation
a Original mammogram with ground truth of mass (red). b Detected ROI of mass (yellow) superimposed on the original mammogram. c Detected ROI (i.e., input mass) obtained with ground truth (red) (Domain X). d Detected ROI obtained from a different mammography dataset (Domain Y). e Original ROI (Domain X) and Synthetic ROI (transferred from Domain Y to Domain X). f Output segmented binary mask of input mass. g Segmented output mass where tissue is masked.
