Digital mammography: hybrid M-channel wavelet transform for microcalcification segmentation

A novel algorithm is proposed for the automatic segmentation of microcalcification clusters (MCCs) in digital mammography. The hybrid method involves the use of a nonlinear filter for image noise suppression, coupled with wavelet transforms for image decomposition and an adaptive method for selective subimage reconstruction as a basis for segmentation of MCCs. The use of M=2,3,4 and 8 channel wavelet transforms are evaluated to determine if the sensitivity of detection of MCCs can be improved and if the selective reconstruction of the higher order M² subimages allows better preservation of the segmented MCCs as required for their classification. The M=2,3,4 and 8 wavelet transforms are implemented on different filter bank structures to determine if their computational efficiency can be improved while retaining properties such as near perfect reconstruction, namely polyphase quadrature mirror filters (QMF), tree structure and lattice structure