Detection of architectural distortion in mammogram images using contourlet transform

This paper presents the detection of architectural distortion a common classification of cancer, using countourlet transform and the phase portrait methods. The architectural distortion is a very important finding in interpreting breast cancers as well as microcalcification and mass on mammograms. However, it is more difficult for physicians to detect architectural distortion than microcalcification and mass. The proposed detection method consists of five steps. Initially, the Otsu technique was performed for segmentation. In order to smooth the edges, the top-hat processing was performed. The contourlet decomposing is performed so that the image can be filtered in multidirections; the phase portrait analysis is done for the orientation analysis purpose. The false positives were eliminated by their concentration of white spaces in the sliding window. Our image database consisted 16 cases from MIAS with architectural distortions. As a result, it was convincing that our methods were effective to detect architectural distortions and further need to be increased with the textural analysis.