Adaptive CAD modules for mass detection in digital mammography

Abstract

The development and evaluation of a new class of algorithms for computer assisted diagnostic (CAD) methods for segmentation and detection of masses in digitized mammograms is reported. Both non-adaptive and adaptive CAD MODULES are reported that employ three key novel image processing CAD modules, specifically tailored for digital mammography, namely: (a) a tree-structured non-linear filter for noise suppression, (b) a multiorientation directional wavelet transform (DWT) for removal of directional features and for the direct detection of spiculations for spiculated lesions, and (c) a multiresolution wavelet transform for image enhancement to improve the segmentation of suspicious areas. The aim of the work is to provide a brief overview of both the non-adaptive and adaptive methods and comparison of their performance using computer ROC curves. The results confirmed the importance of using adaptive CAD methods, where progressive improvement in Az values was observed for each adaptive CAD module. The methods proposed here are useful for other CAD applications such as the detection of microcalcifications and lung nodules.