Novel Methods for T2 Estimation Using Highly Undersampled Radial MRI Data
MR parameter estimation
AdvisorAltbach, Maria I.
MetadataShow full item record
PublisherThe University of Arizona.
RightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
AbstractThe work presented in this dissertation involves the development of parametric magnetic resonance imaging (MRI) techniques that can be used in a clinical set up. In the first chapter an introduction of basic magnetic resonance physics is given. The introduction covers the source to tissue magnetization, the origin of the detectable signal, the relaxation mechanisms, and the imaging principles. In the second chapter T₂ estimation - the main parametric MRI technique addressed in this work - is introduced and the problem associated with T₂ estimation from highly undersampled fast spin-echo (FSE) data is presented. In Chapter 3, a novel model-based algorithm with linearization by principal component analysis (REPCOM) is described. Based on simulations, physical phantom and in vivo data, the proposed algorithm is shown to produce accurate and stable T₂ estimates. In Chapter 4, the concept of indirect echoes associated with the acquisition of FSE data is introduced. Indirect echo correction using the extended phase graph approach is then studied for standard sampled data. A novel reconstruction algorithm (SERENADE) is presented for the reconstruction of decay curves with indirect echoes from highly undersampled data. The technique is evaluated using simulations, physical phantom and in vivo data; decay curves with indirect echoes are shown to be accurately recovered by this technique. Chapter 5 is dedicated to correcting the partial volume effect (PVE) in T₂ estimation. For the case of small lesions within a background tissue, PVE affects T₂ estimation which in turn affects lesion classification. A novel joint fitting algorithm is proposed and compared to conventional fitting algorithms using fully sampled spin-echo (SE) images. It is shown that the proposed algorithm is more accurate, robust, and insensitive to region of interest drawing than the conventional fitting algorithms. Because the acquisition of fully sampled SE images is long, the technique is combined with a thick refocusing slice approach in order to be able to use undersampled FSE data and reduce the acquisition time to a breath hold (~ 20 s). The final chapter summarizes the results presented in the dissertations and discusses areas for future work.
Degree ProgramGraduate College