Use this URL to cite or link to this record in EThOS:
Title: Multifrequency methods for Electrical Impedance Tomography
Author: Malone, E. R.
ISNI:       0000 0004 5359 2329
Awarding Body: University College London (University of London)
Current Institution: University College London (University of London)
Date of Award: 2015
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Multifrequency Electrical Impedance Tomography (MFEIT) is an emerging imaging modality which exploits the dependence of tissue impedance on frequency to recover images of conductivity. Given the low cost and portability of EIT scanners, MFEIT could provide emergency diagnosis of pathologies such as acute stroke, brain injury and breast cancer. Whereas time-difference, or dynamic, EIT is an established technique for monitoring lung ventilation, MFEIT has received less attention in the literature, and the imaging methodology is at an early stage of development. MFEIT holds the unique potential to form images from static data, but high sensitivity to noise and modelling errors must be overcome. The subject of this doctoral thesis is the investigation of novel techniques for including spectral information in the image reconstruction process. The aim is to improve the ill-posedness of the inverse problem and deliver the first imaging methodology with sufficient robustness for clinical application. First, a simple linear model for the conductivity is defined and a simultaneous multifrequency method is developed. Second, the method is applied to a realistic numerical model of a human head, and the robustness to modelling errors is investigated. Third, a combined image reconstruction and classification method is developed, which allows for the simultaneous recovery of the conductivity and the spectral information by introducing a Gaussian-mixture model for the conductivity. Finally, a graph-cut image segmentation technique is integrated in the imaging method. In conclusion, this work identifies spectral information as a key resource for producing MFEIT images and points to a new direction for the development of MFEIT algorithms.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available