Classifying Clusters of Microcalcifications in Digitized Mammograms by Artificial Neural Network

Computer-Aided Diagnosis (CAD) schemes have presented good results in aiding the early diagnosis of breast cancer. The detected signals classification demands multi-works investigations, since cytological characteristics concerning the mammographic findings have to be investigated in addition to computer techniques. Artificial neural networks (ANN) have been successfully used in CAD classifiers. with success in the classification in CAD. For example, the classification of clustered microcalcifications has been made from individual microcalcifications analysis. In this work, regarding characteristics determined only from the cluster itself, and discarding the characteristics analysis and extraction from individual microcalcifications, such a classification was made in two classes: “nonsuspect” and “suspect”. Dismissing microcalcifications individual features for the network input has allowed to eliminate procedures intended to separate each structure from the whole image. The classifier using ANN has shown the geometric descriptors efficiency for characterizing microcalcifications clusters as well as the influence of features extracted from images reports, as “age” and “density”. The best data have shown 92% of correct results, with A, = 0.96.