Detection of architectural distortion using coherence in relation to the expected orientation of breast tissue

We propose methods for the detection of architectural distortion in prior mammograms. The expected orientation of the breast tissue at each pixel was derived by using Gabor filters, the automatically detected breast boundary, the nipple, and the pectoral muscle (in mediolateral-oblique views). The angular deviation of the oriented structures as compared to the expected orientation was analyzed, using a measure of coherence, to detect sites of architectural distortion. With 106 prior mammograms of 56 interval-cancer cases and 52 mammograms of 13 normal cases, a sensitivity of 80% was obtained at 10.3 false positives/ image.