Automatic detection of clustered microcalcifications in digital mammograms using an SVM classifier Armando Bazzani † , Alessandro Bevilacqua ‡ , Dante Bollini † , Rosa Brancaccio †, Renato Campanini † , Nico Lanconelli , Alessandro Riccardi †, Davide Romani and Gianluca Zamboni † † Department of Physics, University of Bologna, Italy ‡ Department of Electronics, Computer Science and Systems, University of Bologna, Italy National Institute for Nuclear Physics, Bologna, Italy Abstract. In this paper we investigate the performance of a Computer Aided Diagnosis (CAD) system for the detection of clustered microcalcifications in mammograms. Our detection algorithm consists on the combination of two different methods. The first one, based on difference-image techniques and gaussianity statistical tests, finds out the most obvious signals. The second one is able to discover more subtle microcalcifications by exploiting a multiresolution analysis by means of the wavelet transform. In the false- positive reduction step we separate false signals from microcalcifications by means of an SVM classifier. Our algorithm yields a sensitivity of 94.6% with 0.6 false positive cluster per image on the 40 images of the Nijmegen database. 1. Introduction Breast cancer is the most common form of cancer among women. The presence of microcalcifications in breast tissues is one of the main features considered by radiologists for its diagnosis. Several techniques developed for the automated detection of microcalcifications can mainly be grouped in three different categories: multiresolution analyses [1,2], filtering methods [3] and statistical methods [4,5]. By comparing the different methods it turns out that some microcalcifications are detected by one method but missed by others. In this paper we propose an approach based on the combination of different detection methods in order to get optimal performances. Yoshida et al pointed out that the simultaneous use of two or more techniques might improve the results of an optimized single method [6]. The basic idea of our method is to combine a multiresolution analysis based on wavelet transform with a difference-image method and a gaussianity statistical test and to perform a logical OR operation on the detected microcalcifications before clustering. In the false-positive reduction (fpr) step we try to separate false signals from microcalcifications by using a classifier based on a Support Vector Machine (SVM). The detection scheme has been tested on 40 digitized mammograms coming from Nijmegen Hospital. D-Facto public., ISBN 2-930307-00-5, pp. 195-200 B orks 0, ES Netw r 0 A l 0 ug ra 2 NN Neu e l '2 l s i 000 icia pr Artif ( A p on B 8 ro m e 2 ce iu l - edi pos g 6 ngs ym i 2 - S u E an m, urope )