Breast Cancer Research
RISK PREDICTION
Mammographic Parenchymal Pattern (MPP) and Percent Density (PD) as radiomics biomarkers of breast cancer risk
Track changes in MPP and PD over time in screening mammograms
Automated segmentation and estimation of percent density (PD)
Parenchymal pattern analysis
Pectoral Muscle Segmentation – Deep CNN
Architecture of DCNN for automated pectoral muscle identification on MLO-view mammograms. Each box corresponds to a multi-channel feature map. The number of channels shown as parameters a to m, is adjustable to optimize performance. The size of each feature map is shown at the lower left edge of the box.
Parenchymal pattern analysis by deep belief network (DBN)
Learn to extract a deep hierarchical representation from the training data
Deep features classified by random forest classifier
Mammographic parenchymal pattern (MPP) as risk predictor over time