Model-based detection of spiculated lesions in mammograms

Reyer Zwiggelaar, Timothy C. Parr, James E. Schumm, Ian W. Hutt, Christopher J. Taylor, Susan M. Astley, Caroline R M Boggis

    Research output: Contribution to journalArticlepeer-review

    Abstract

    Computer-aided mammographic prompting systems require the reliable detection of a variety of signs of cancer. In this paper we concentrate on the detection of spiculatcd lesions in mammograms. A spiculatcd lesion is typically characterized by an abnormal pattern of linear structures and a central mass. Statistical models have been developed to describe and detect both these aspects of spiculated lesions. We describe a generic method of representing patterns of linear structures, which relics on the use of factor analysis to separate the systematic and random aspects of a class of patterns. We model the appearance of central masses using local scale-orientation signatures based on recursive median filtering, approximated using principal-component analysis. For lesions of 16 mm and larger the pattern detection technique results in a sensitivity of 80% at 0.014 false positives per image, whilst the mass detection approach results in a sensitivity 80% at 0.23 false positives per image. Simple combination techniques result in an improved sensitivity and specificity close to that required to improve the performance of a radiologist in a prompting environment.
    Original languageEnglish
    Pages (from-to)39-62
    Number of pages23
    JournalMedical Image Analysis
    Volume3
    Issue number1
    Publication statusPublished - 1999

    Keywords

    • Central mass detection
    • Mammogram
    • Oriented line patterns
    • Spiculated lesions

    Fingerprint

    Dive into the research topics of 'Model-based detection of spiculated lesions in mammograms'. Together they form a unique fingerprint.

    Cite this