Detection of Masses in Digital Mammograms using K-Means and Support Vector Machine

• Autores: Leonardo de Oliveira Martins, Geraldo Braz Junior, Aristófanes Corrêa Silva, Anselmo Cardoso de Paiva, Marcelo Gattass
• Localización: ELCVIA. Electronic letters on computer vision and image analysis, ISSN-e 1577-5097, Vol. 8, Nº. 2, 2009, págs. 39-50
• Idioma: inglés
• Enlaces
  • Texto completo  1    2  
• Resumen

  • Female breast cancer is a major cause of death in occidental countries. CAD/CADx systems can aidradiologists in detection and diagnostic of lesions in mammograms. In this work, we present a methodologyto detect masses from mammograms. The K-means clustering algorithm is used to split the mammogramsin regions. Each region is then classified through a Support Vector Machine (SVM) as mass or non-massregion. SVM is a machine-learning method, based on the principle of structural risk minimization, whichperforms well when applied to data outside the training set. We use a set of textural and shape measures todetect suspicious regions, as bening and malignant masses. Each textural measure (contrast, homogeneity,inverse difference moment, entropy and energy) is computed through the co-ocurrence matrix technique.The methodology obtained an accuracy of 93.11% discriminate mass from non-mass elements.