Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.798870
Title: Model-based reconstruction of accelerated quantitative magnetic resonance imaging (MRI)
Author: Bano, Wajiha
Awarding Body: University of Edinburgh
Current Institution: University of Edinburgh
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:
Quantitative MRI refers to the determination of quantitative parameters (T1,T2,diffusion, perfusion etc.) in magnetic resonance imaging (MRI). The 'parameter maps' are estimated from a set of acquired MR images using a parameter model, i.e. a set of mathematical equations that describes the MR images as a function of the parameter(s). A precise and accurate highresolution estimation of the parameters is needed in order to detect small changes and/or to visualize small structures. Particularly in clinical diagnostics, the method provides important information about tissue structures and respective pathologic alterations. Unfortunately, it also requires comparatively long measurement times which preclude widespread practical applications. To overcome such limitations, approaches like Parallel Imaging (PI) and Compressed Sensing (CS) along with the model-based reconstruction concept has been proposed. These methods allow for the estimation of quantitative maps from only a fraction of the usually required data. The present work deals with the model-based reconstruction methods that are applicable for the most widely available Cartesian (rectilinear) acquisition scheme. The initial implementation was based on accelerating the T*2 mapping using Maximum Likelihood estimation and Parallel Imaging (PI). The method was tested on a Multiecho Gradient Echo (MEGE) T*2 mapping experiment in a phantom and a human brain with retrospective undersampling. Since T*2 is very sensitive to phase perturbations as a result of magnetic field inhomogeneity further work was done to address this. The importance of coherent phase information in improving the accuracy of the accelerated T*2 mapping fitting was investigated. Using alternating minimization, the method extends the MLE approach based on complex exponential model fitting which avoids loss of phase information in recovering T*2 relaxation times. The implementation of this method was tested on prospective(real time) undersampling in addition to retrospective. Compared with fully sampled reference scans, the use of phase information reduced the error of the accelerated T*2 maps by up to 20% as compared to baseline magnitude-only method. The total scan time for the four times accelerated 3D T*2 mapping was 7 minutes which is clinically acceptable. The second main part of this thesis focuses on the development of a model-based super-resolution framework for the T2 mapping. 2D multi-echo spin-echo (MESE) acquisitions suffer from low spatial resolution in the slice dimension. To overcome this limitation while keeping acceptable scan times, we combined a classical super-resolution method with an iterative model-based reconstruction to reconstruct T2 maps from highly undersampled MESE data. Based on an optimal protocol determined from simulations, we were able to reconstruct 1mm3 isotropic T2 maps of both phantom and healthy volunteer data. Comparison of T2 values obtained with the proposed method with fully sampled reference MESE results showed good agreement. In summary, this thesis has introduced new approaches to employ signal models in different applications, with the aim of either accelerating an acquisition, or improving the accuracy of an existing method. These approaches may help to take the next step away from qualitative towards a fully quantitative MR imaging modality, facilitating precision medicine and personalized treatment.
Supervisor: Davies, Michael ; Marshall, Ian Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.798870  DOI:
Keywords: quantitative MRI ; model-based reconstruction ; accelerated MRI
Share: