Analysing the Curvature of the Pectoralis Muscle in Mammograms

  • Christina Olsén
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2749)


One of the most important criteria considering the assessment of the diagnostic quality in mammograms is the pectoralis muscle. In this paper a routine that automatically analysis the shape of the muscle and compute measurements needed to determine the quality concerning the muscle is presented. The method is based on a division of the m uscle in to sub images and using the Hough transform to compute the slope of each sub images line. The performance to determine the shape of the pectoralis muscle is 95.5 % correctly classified mammograms based on 155 randomly chosen images from the MIAS database. Therefore, the conclusion is that the proposed shape analysis method is a reliable method for determining the shape of the muscle.


  1. 1.
    University of south florida digital mammography. Digital Database for Screening Mammography (DDSM), August 2002.Google Scholar
  2. 2.
    S. Carlson. Personal communication with chief radiologist Stina Carlson at the University Hospital Umeå Sweden, February 2002.Google Scholar
  3. 3.
    R. Chandrasekhar. Systematic Segmentation of Mammograms. PhD thesis, University of Western Australia, Department of Electrical and Engineering, Nedlands, 1996.Google Scholar
  4. 4.
    R. Chandrasekhar, S.M. Kwok, and Y. Attikiozel. Automatic evaluation of mammographic adequacy and quality on the mediolateral oblique view. In proceedings of IWDM, pages 182–186, 2002.Google Scholar
  5. 5.
    G.W. Eklund, G. Cardenosa, and W. Parson. Assessing adequacy of mammographic image quality. Radiology., 190:297–307, 1994.Google Scholar
  6. 6.
    F. Georgsson. Algorithms and Techniques for Computer Aided Mammographic Screening. PhD thesis, UMINF-01.15 Umeå University, Department of Computing Science, 2001.Google Scholar
  7. 7.
    S. Heywang-Köbrunner, I. Schreer, and D. Dershaw. Diagnostic Breast Imaging. Thieme, Stuttgart, 1997.Google Scholar
  8. 8.
    N. Karssemeijer. Automated classification of parenchymal patterns in mammograms. Physics in Medicine and Biology., 43:365–378, 1998.CrossRefGoogle Scholar
  9. 9.
    S.M. Kwok, R. Chandrasekhar, and Y. Attikiozel. Automatic pectoral muscle segmentation on mammograms by straight line estimation and cliff detection. In proceedings of ANZIISG, pages 67–72, 2001.Google Scholar
  10. 10.
    C. Olsén. Automatic determination of mammogram adequacy. Master’s thesis, UMNAD 424/02 Umeå University, Department of Computing Science, 2002.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Christina Olsén
    • 1
  1. 1.Department of Computing ScienceUmeå UniversityUmeåSweden

Personalised recommendations