University of Sussex
Browse
(2014.03.26)_Beqo,_Nevis.pdf (33.04 MB)

An investigation into combining electrical impedance mammography with 3D ultrasound for breast cancer detection

Download (33.04 MB)
thesis
posted on 2023-06-08, 15:37 authored by Nevis Beqo
Worldwide, breast cancer accounts for 22% of all cancer incidences in women, causing 458,503 deaths per annum [WHO 2008]. The most common screening method is X-ray mammography, an ionising method, which has many technical and age group limitations and has come under scrutiny for accuracy and safety on many occasions. Electrical Impedence Mammography (EIM) is a novel, non-invasive, non-ionising imaging modality based on bioimpedence. Initial test results show it as very promising; however the image resolution is quite low. Ultrasound imaging is widely used for high-resolution medical imaging and clinical diagnostics. Ultrasound is non-invasive and is effective in imaging soft body tissue, including subcutaneous structures and organs, but fails to distinguish tissue type. Merging the information of the above modalities and integrating them in an automated device offers a safe, non-invasive, fast, higher accuracy, breast cancer detection method for all age groups. To explore the proposed system, the work was divided into cumulative integrative stages: investigation of the technical challenges of a real world breast scanner device for each modality and the combination of both, including engineering, repeatability, safety and ergonomics, to adhere to medical device standards; data acquisition systems design, signal processing and calibration; image geometry correction; 3D image reconstruction; volumetric merger and visualisation; validation with dual modality scans on phantom and in-vivo; DICOM image porting. These modalities were successfully combined in a unified automated breast scanner that can accommodate and scan 95% of women (breast volume up to 1200cc) in a safe and comfortable way. Data acquisition and scan time achieved is five minutes per breast. The image results achieved from this research complement each other by integrating the boundary information of Ultrasound with the impedence data and tissue discrimination of EIM, therefore potentially providing a more complete and accurate cancer diagnostic method. The images were successfully ported into the DICOM radiology imaging standard therefore becoing platform independent. This work concluded with over twenty academic publications and two filed patents on the technology of a breast scanner and on the methodology of its imaging.

History

File Version

  • Published version

Pages

161.0

Department affiliated with

  • Engineering and Design Theses

Qualification level

  • doctoral

Qualification name

  • phd

Language

  • eng

Institution

University of Sussex

Full text available

  • Yes

Legacy Posted Date

2013-09-27

Usage metrics

    University of Sussex (Theses)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC