A novel featureless approach to mass detection in digital mammograms based on support vector machines

• Published in: Physics in medicine & biology Vol. 49; no. 6; p. 961
• Main Authors: Campanini, Renato, Dongiovanni, Danilo, Iampieri, Emiro, Lanconelli, Nico, Masotti, Matteo, Palermo, Giuseppe, Riccardi, Alessandro, Roffilli, Matteo
• Format: Journal Article
• Language: English
• Published: England 21.03.2004
• Subjects: Algorithms; Artificial Intelligence; Breast Neoplasms - classification; Breast Neoplasms - diagnostic imaging; Cluster Analysis; Computing Methodologies; Expert Systems; False Positive Reactions; Female; Humans; Information Storage and Retrieval - methods; Mammography - methods; Pattern Recognition, Automated; Radiographic Image Enhancement - methods; Radiographic Image Interpretation, Computer-Assisted - methods; Reproducibility of Results; Sensitivity and Specificity; Signal Processing, Computer-Assisted
• ISSN: 0031-9155
• DOI: 10.1088/0031-9155/49/6/007

In this work, we present a novel approach to mass detection in digital mammograms. The great variability of the appearance of masses is the main obstacle to 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 have chosen not to extract any feature, for the detection of the region of interest; in contrast, we exploit all the information available on the image. A multiresolution overcomplete wavelet representation is performed, in order to codify the image with redundancy of information. The vectors of the very-large space obtained are then provided to a first support vector machine (SVM) classifier. The detection task is considered here as a two-class pattern recognition problem: crops are classified as suspect or not, by using this SVM classifier. False candidates are eliminated with a second cascaded SVM. To further reduce the number of false positives, an ensemble of experts is applied: the final suspect regions are achieved by using a voting strategy. The sensitivity of the presented system is nearly 80% with a false-positive rate of 1.1 marks per image, estimated on images coming from the USF DDSM database.