Enabling clinical Hyperpolarised 13C-MR cancer imaging through phantom development, pulse sequence optimisation and quantitative image processing - PhDData

Access database of worldwide thesis




Enabling clinical Hyperpolarised 13C-MR cancer imaging through phantom development, pulse sequence optimisation and quantitative image processing

The thesis was published by Chowdhury, Rafat, in November 2023, UCL (University College London).

Abstract:

Hyperpolarised 13C-Magnetic Resonance Imaging (13C-MR), via dissolution Dynamic Nuclear Polarisation (d-DNP), is an emerging technique which uses a non-ionising contrast agent to quantify metabolic processes in vivo. Reactions such as the conversion of carbon-13 labelled pyruvate into lactate, in a process analogous to the Warburg effect, can now be observed in real time.

Hyperpolarised 13C-MR has previously been used to demonstrate significant differences in metabolism between healthy tissue and prostate cancer. Clinical hyperpolarised 13C-MR studies are on-going with several imaging techniques presented in recent years.

This thesis aims to develop physical and in silico platforms on which to optimise and test pulse sequences, whilst also optimising an existing pulse sequence and applying this in clinical hyperpolarised 13C-MR studies.

Firstly, a test object was developed which successfully reduced the variance of analysis outcomes in hyperpolarised 13C-MR studies by incorporating an automated injection system for the hyperpolarised agent (coefficient of variation: 11-23% (n = 8)). Secondly, an existing numerical simulator was empirically validated (r: >0.95) and then used to identify the optimal spectral parameters for an ME-bSSFP sequence, for use in prostate cancer imaging during hyperpolarised 13C-MR studies.

Building on this work the optimised ME-bSSFP sequence was used to image a small cohort (n = 5) of subjects with biopsy confirmed prostate cancer tumours in a hyperpolarised 13C-MR study. Significant differences between tumourous and healthy tissues were found (p



Read the last PhD tips