Advertisement

Subjective and computer-based characterization of mammographic patterns

  • Kostas Marias
  • Styliani Petroudi
  • Ruth English
  • Rosie Adams
  • Michael Brady

Abstract

We investigate the subjective characterisation of breast patterns and the possibility of automatic classification using computed measurements. Several clinical studies used mammogram pattern classification in order to investigate their relation to breast cancer. However, the establishment of an objective ground truth for pattern classification remains a problem that diminishes the potential of such studies since there is significant variability in the interpretation and coding of the patterns. For this reason, we investigate the observer variability in classifying mammographie patterns as well as the variability between film-screen and computer screen reading with respect to pattern classification. We also investigate the role of quantitative measures in characterising the breast parenchyma, discuss their agreement with the observer and argue that a normalised mammographie representation can improve the correspondence between measures and the perceived mammogram patterns.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Wolfe, J. N.: Breast pattern as an index for developing breast cancer. American Journal of Roentgenology. 126 (1976) 1130–1139PubMedGoogle Scholar
  2. 2.
    ACR.: Illustrated breast imaging reporting and data system (BI-RADS™). American College of Radiology (1998)Google Scholar
  3. 3.
    Sala, E., Warren, R., McCann. J., Duffy, S., Lüben, R., Day, N.: High-risk mammographie parenchymal patterns, hormone replacement therapy and other risk factors: a case-control study. Int J Epidemiol. 29 (4) (2000) 629–362PubMedCrossRefGoogle Scholar
  4. 4.
    Miller, P., Astley, S.: Classification of breast tissue by texture analysis. Image and Vision Computing. 10 (5) (1992) 277–282CrossRefGoogle Scholar
  5. 5.
    Highnam, R.P., Brady, J.M.: Mammographie image processing. Kluwer Academic Press (1999)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Kostas Marias
    • 1
  • Styliani Petroudi
    • 1
  • Ruth English
    • 2
  • Rosie Adams
    • 2
  • Michael Brady
    • 1
  1. 1.Robotics Research Group, Engineering ScienceOxford UniversitySummertown, OxfordUK
  2. 2.Breast Care UnitChurchill HospitalOxfordUK

Personalised recommendations