Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.779413
Title: Investigating advanced magnetic resonance imaging for improved diagnosis and prediction of treatment response in Wilms' tumour patients
Author: Rogers, Harriet J.
ISNI:       0000 0004 7965 1091
Awarding Body: UCL (University College London)
Current Institution: University College London (University of London)
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Abstract:
Wilms' tumour is the most common paediatric renal tumour. In Europe treatment involves pre-operative chemotherapy followed by surgery. Many patients receive MRI scans which include diffusion weighted imaging (DWI) throughout their diagnosis and treatment. This thesis retrospectively acquired Wilms' tumour MRI data and prospectively acquired renal DWI data in healthy volunteers. Four models of diffusion were used throughout this thesis; mono-exponential, IVIM (intravoxel incoherent motion), stretched exponential, and kurtosis. In healthy volunteers, models were compared based on the reproducibility of the parameters, when calculated based on different b values and magnetic fields strengths. It was shown that ADC, D (IVIM), f (IVIM), Dk (kurtosis), and α (stretched exponential) had high levels of reproducibility whereas reproducibility was poorer in D* (IVIM), K (kurtosis) and DDC (stretched exponential). Model fits were compared in Wilms' tumour and contralateral normal kidney data using the Akaike Information Criterion. It was shown that all raw DWI data favoured non- Gaussian models as opposed to a mono-exponential model. DWI data acquired in Wilms' tumour favoured the stretched exponential model, and DWI data acquired in normal kidneys favoured the IVIM model. The volume of necrotic tissue post-chemotherapy is an important marker of treatment response. However, currently identification of necrosis relies on gadolinium contrast enhancement. It was shown that a combination of T1weighted imaging and ADC could provide an alternative method to visualising and quantifying necrosis, allowing future studies to estimate the volume fraction of necrosis in Wilms' tumour without gadolinium. Finally, it was shown that certain Wilms' tumour subtypes could be distinguished in vivo using DWI, whereas currently this relies on histological tissue analysis post- surgery. The parameters D* (IVIM) and K (kurtosis) provided the best stratification between subtypes, however, the earlier study demonstrated that the reproducibility of these parameters was poor, which may limit their clinical utility.
Supervisor: Clark, C. ; Hales, P. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.779413  DOI: Not available
Share: