In this paper we present a novel approach to mass detection in digital mammograms. The great variability of the masses appearance is the main obstacle of building a mass detection method. It is indeed demanding to characterize all the varieties of masses with a reduced set of features. Hence, in our approach we decide not to extract any feature, for the detection of the region of interest; on the contrary we exploit all the information available on the image. No a priori knowledge and no appearance model are used. A multiresolution overcomplete wavelet representation is achieved, in order to codify the image with redundancy of information. The vectors of the very-large space obtained are classified by means of an SVM classifier. Training, validation and test are accomplished on images coming from USF DDSM database. The sensitivity of the presented system is 84% with a false-positive rate of 3.1 marks per image.
Campanini, R., Bazzani, A., Bevilacqua, A., Bollini, D., Dongiovanni, D., Iampieri, E., et al. (2003). A novel approach to mass detection in digital mammography based on Support Vector Machines (SVM). HEIDELBERGER PLATZ 3, D-14197 BERLIN, GERMANY : SPRINGER-VERLAG BERLIN.
A novel approach to mass detection in digital mammography based on Support Vector Machines (SVM)
Campanini, R;Bazzani, A;Bevilacqua, A;Bollini, D;Dongiovanni, D;Lanconelli, N;Roffilli, M;
2003
Abstract
In this paper we present a novel approach to mass detection in digital mammograms. The great variability of the masses appearance is the main obstacle of building a mass detection method. It is indeed demanding to characterize all the varieties of masses with a reduced set of features. Hence, in our approach we decide not to extract any feature, for the detection of the region of interest; on the contrary we exploit all the information available on the image. No a priori knowledge and no appearance model are used. A multiresolution overcomplete wavelet representation is achieved, in order to codify the image with redundancy of information. The vectors of the very-large space obtained are classified by means of an SVM classifier. Training, validation and test are accomplished on images coming from USF DDSM database. The sensitivity of the presented system is 84% with a false-positive rate of 3.1 marks per image.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
