e-space
Manchester Metropolitan University's Research Repository

    Automatic Quantification of Breast Arterial Calcification on Mammographic Images

    Mazidi, N, Roobottom, C and Masala, G ORCID logoORCID: https://orcid.org/0000-0001-6734-9424 (2019) Automatic Quantification of Breast Arterial Calcification on Mammographic Images. In: Innovation in Medincine & Heathcare, 17 June 2019 - 19 June 2019, St Julians, Malta.

    [img]
    Preview
    Accepted Version
    Download (1MB) | Preview

    Abstract

    © 2019, Springer Nature Singapore Pte Ltd. This paper describes the research and development of an automatic computer system that is used to quantify breast arterial calcifications in mammography scans. A few prior studies have attempted to establish a relationship between breast arterial calcification (BAC) and the rate of coronary artery disease (CAD) risk factors. The majority of these studies demonstrated a positive association between BAC and increasing age. Large scale cohort studies and retrospective studies have almost uniformly suggested a strong association between BAC and cardiovascular disease-related morbidity and mortality. This strong association of BAC with cardiovascular pathology suggests that BAC should also be persistently associated with radiographically determined CAD. A method of image processing, segmentation, and quantification used to highlight and recognise calcified blood vessels in the breast is proposed and described in detail. This project aims to introduce a new use for digital Mammography, which is currently solely used for diagnosing breast cancer in female patients. A method of detecting BAC is introduced at no additional cost, having an adequate degree of accuracy, around 82%, which means that this type of system could be used to assist a radiographer in diagnosing BAC by indicating whether the patient has a high or low severity of calcification.

    Impact and Reach

    Statistics

    Activity Overview
    6 month trend
    356Downloads
    6 month trend
    229Hits

    Additional statistics for this dataset are available via IRStats2.

    Altmetric

    Repository staff only

    Edit record Edit record