ORIGINAL PAPER Discovering Mammography-based Machine Learning Classifiers for Breast Cancer Diagnosis Raúl Ramos-Pollán & Miguel Angel Guevara-López & Cesar Suárez-Ortega & Guillermo Díaz-Herrero & Jose Miguel Franco-Valiente & Manuel Rubio-del-Solar & Naimy González-de-Posada & Mario Augusto Pires Vaz & Joana Loureiro & Isabel Ramos Received: 16 December 2010 / Accepted: 28 March 2011 # Springer Science+Business Media, LLC 2011 Abstract This work explores the design of mammography- based machine learning classifiers (MLC) and proposes a new method to build MLC for breast cancer diagnosis. We massively evaluated MLC configurations to classify features vectors extracted from segmented regions (pathological lesion or normal tissue) on craniocaudal (CC) and/or mediolateral oblique (MLO) mammography image views, providing BI-RADS diagnosis. Previously, appropriate combina- tions of image processing and normalization techniques were applied to reduce image artifacts and increase mammograms details. The method can be used under different data acquisition circumstances and exploits computer clusters to select well performing MLC configu- rations. We evaluated 286 cases extracted from the repository owned by HSJ-FMUP, where specialized radiologists segmented regions on CC and/or MLO images (biopsies provided the golden standard). Around 20,000 MLC configurations were evaluated, obtaining classifiers achieving an area under the ROC curve of 0.996 when combining features vectors extracted from CC and MLO views of the same case. Keywords Breast cancer CAD . Machine learning classifiers . Mammography classifiers Introduction Breast cancer is a major concern and the second-most common and leading cause of cancer deaths among women [1]. According to published statistics, breast cancer has become a major health problem in both developed and R. Ramos-Pollán (*) : C. Suárez-Ortega : G. Díaz-Herrero : J. M. Franco-Valiente : M. Rubio-del-Solar CETA-CIEMAT Center of Extremadura for Advanced Technologies, Calle Sola 1, 10200 Trujillo, Spain e-mail: raul.ramos@ciemat.es C. Suárez-Ortega e-mail: cesar.suarez@ciemat.es G. Díaz-Herrero e-mail: guillermo.diaz@ciemat.es J. M. Franco-Valiente e-mail: josemiguel.franco@ciemat.es M. Rubio-del-Solar e-mail: manuel.rubio@ciemat.es M. A. Guevara-López : N. González-de-Posada : M. A. P. Vaz INEGI-FEUP Institute of Mechanical Engineering and Industrial Management, Faculty of Engineering, University of Porto, Rua Dr. Roberto Frias 400, 4200–465 Porto, Portugal M. A. Guevara-López e-mail: mguevaral@inegi.up.pt N. González-de-Posada e-mail: nposada@inegi.up.pt M. A. P. Vaz e-mail: gmavaz@inegi.up.pt J. Loureiro : I. Ramos HSJ-FMUP Hospital de São João - Faculty of Medicine, University of Porto, Al. Prof. Hernani Monteiro, 4200–319 Porto, Portugal J. Loureiro e-mail: joanaploureiro@gmail.com I. Ramos e-mail: radiologia.hsj@mail.telepac.pt J Med Syst DOI 10.1007/s10916-011-9693-2