Breast cancer detection in mammogram with AM-FM modeling and Gabor filtering

In this paper, we propose to apply an integrated method to the mammography images for breast cancer detection. This method combines AM-FM modeling techniques with image filtering by a bank of specially designed Gabor filters. Cancer features will be extracted from the decomposed or filtered images and projected back to the reconstructed mammography images. We seek to validate our approach by showing that a mammogram can be accurately decomposed, analyzed and reconstructed. Experimental results with a number of mammography images and cancer cases have confirmed our expectations.