Detection of Microcalcifications Using Wavelet-Based Thresholding and Filling Dilation

Microcalcifications (MCs) are the main symptoms of breast cancer in the mammograms. This paper proposed a new computer-aided diagnosis (CAD) algorithm to detect the MCs. At first, discrete wavelet transform (DWT) was applied to extract the high-frequency signal, and thresholding with hysteresis was used to locate the suspicious MCs. Then, filling dilation was utilized to segment the desired regions. During the detection process, ANFIS was applied for auto-adjustment, making the CAD more adaptive. Finally, the segmented MCs were classified with MLP, and a satisfying result validated this method.

[1]  Weidong Xu,et al.  Application of CMAC-Based Networks on Medical Image Classification , 2004, ISNN.

[2]  Wei Zhang,et al.  Optimizing wavelet transform based on supervised learning for detection of microcalcifications in digital mammograms , 1995, Proceedings., International Conference on Image Processing.

[3]  Robert M. Nishikawa,et al.  Computer-aided detection of clustered microcalcifications , 1992, [Proceedings] 1992 IEEE International Conference on Systems, Man, and Cybernetics.

[4]  John Håkon Husøy,et al.  Optimal filter-based detection of microcalcifications , 2001, IEEE Transactions on Biomedical Engineering.

[5]  A. K. Chan,et al.  Microcalcification cluster detection in digitized mammograms using multiscale techniques , 1998, 1998 IEEE Southwest Symposium on Image Analysis and Interpretation (Cat. No.98EX165).