Exploring Wavelets Toward an Automatic Microcalcification Detection in Breast Cancer

2D wavelet transform decomposition is widely used in computer aided detection of microcalcifications in mammograms. The aim of this paper is to investigate the better type of wavelet and its optimal potential level of decomposition that gives us better detection. Our algorithm consists of four steps: first, dimension reduction is performed on the mammography images to delimitate the ROI (region of interest). Second, microcalcification profiles are extracted from digital mammograms. Next, a 1D WT with different families of wavelet is applied on the signal up to the sixth level. Finally, comparison between details coefficients of each level is done to carry out the optimal level. To validate our result, 2D wavelet transform decomposition and reconstruction with the better wavelet and up to its optimal level is applied on digital mammograms from the DDSM (digital database for screening mammography) to carry out microcalcifications