Detección de patologías de la mama mediante filtros multirresolución

Abstract

Mammography is considered as the gold standard examination for early detection of breast cancer. However, in the techniques of medical image analysis 3 aspects are yet an issue: noise removal, segmentation of images, and extracting the region of interest (ROI). This work presents a survey of breast mammography and the preliminary results corresponding to a method for the automatic detection of breast anomalies in mammography images using histogram transformation, contrast enhancement and Quadrature Mirror Filter (QMF) banks. The QMF filters are preceded by an advanced tree structured nonlinear filter for noise suppression. Several multiresolution methods are reported for feature extraction in breast cancer using digital mammography. The initial approach is directed to malignant diseases, such as microcalcification clusters, present in mammograms, using initially two or three channels for the multiresolution decomposition.