University of Sussex
Browse
- No file added yet -

Assesment and optimisation techniques for an electrical impedance mammography system

Download (6.87 MB)
thesis
posted on 2023-06-08, 17:37 authored by Benjamin Lingham
Breast cancer detection through the use of Electrical Impedance Tomography (EIT) has been proposed for a number of years. Typically in an EIT or when in use in breast cancer detection, Electrical Impedance Mammography (EIM) system, the patient interface is that of electrodes. Positive and negative sinusoidal signals are injected into a patient multiple times at multiple frequencies, recording the developed surface voltages that naturally develop from the impedance of the body. A 2D or 3D reconstruction and visualisation of the impedance distribution is possible through the use of the recorded voltage amplitude and phase. Higher resolution images are achieved through higher electrode density (i.e.smaller electrode distance). Furthermore, tissue has a characteristic frequency response,which can be recorded if different injection frequencies are utilised. The higher thefrequency, the deeper the signal can penetrate into tissue and deeper through the cellular structures, potentially leading to tissue characterisation through its frequency response. Our group has suggested a unique combined EIT and ultrasound multimodal imaging system to detect breast cancer. This will use an EIT system to initially scan the breast,combining this parametric data with the high-resolution images of ultrasound to give a more accurate diagnosis. The prototype, known as V2, was functional but performed poorly, especially with respect to the signal to noise ratio. Images generated with this system were unclear. My role within the group was to analyze the performance of the V2 system and research best practice methods to improve the existing design, taking into consideration the intricacies of EIT hardware design and the targeted application. The improvements I suggested were incorporated into the V3 system, which was then compared with the V2 system, and showed significant performance improvement.

History

File Version

  • Published version

Pages

104.0

Department affiliated with

  • Engineering and Design Theses

Qualification level

  • masters

Qualification name

  • mphil

Language

  • eng

Institution

University of Sussex

Full text available

  • Yes

Legacy Posted Date

2014-07-07

Usage metrics

    University of Sussex (Theses)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC