edoc

Exploring configuration-based relaxometry and imaging with balanced steady-state free precession

Nguyen, Damien. Exploring configuration-based relaxometry and imaging with balanced steady-state free precession. 2018, Doctoral Thesis, University of Basel, Faculty of Science.

[img]
Preview
PDF
Available under License CC BY-NC-ND (Attribution-NonCommercial-NoDerivatives).

6Mb

Official URL: http://edoc.unibas.ch/diss/DissB_13076

Downloads: Statistics Overview

Abstract

Over the years, magnetic resonance (MR) imaging has become a fundamental part of the diagnostic process in hospitals worldwide. While the underlying physics dates back more than 60 years with the development of nuclear magnetic resonance, methods that aim to accurately measure the multitude of parameters governing the signal formation are still a topic of active research and developments today. The main aim of this work is to explore the possibilities of developing quantification techniques based on a particular type of MR acquisitions: balanced Steady-State Free Precession (bSSFP). The first chapter briefly introduces the most relevant basic concepts of MR physics that will serve as foundation for the development of the methods presented thereafter. In the second chapter, a new method using multiple bSSFP scans is presented that aim to achieve motion insensitive three-dimensional quantification of relaxation times and thereby improve a recently published technique based on unbalanced gradient echo acquisitions. The method is then evaluated both in phantoms and in vivo studies at
3 T and the results discussed. These include an interesting bias of the method that might provide useful insights into the underlying tissue micro-structure.
One the major challenges with bSSFP imaging is the presence of dark regions inside the images, which are due to inhomogeneities in the main magnetic field. In the third chapter, a new approach to address those issues is proposed, termed trueCISS, which combines fast imaging using sparse sampling with compressed sensing reconstructions and multi-parametric fitting, which ultimately allows the synthesis of artefact-free images. Evaluation of the new method is done at 3 T for the human brain.
Finally, an extension of the trueCISS technique is presented in chapter four, where the process used to model the data to the signal equation is replaced by a novel algorithm based on configuration theory, which is essentially a representation of the signal formation processes in the Fourier domain. The improved trueCISS imaging method is then successfully evaluated with measurements at ultra high field strengths such as 7 T and 9.4 T which demonstrate the advantages of the new approach.
Advisors:Bieri, Oliver and Kozerke, Sebastian
Faculties and Departments:03 Faculty of Medicine > Bereich Querschnittsfächer (Klinik) > Radiologie USB > Radiologische Physik (Bieri)
03 Faculty of Medicine > Departement Klinische Forschung > Bereich Querschnittsfächer (Klinik) > Radiologie USB > Radiologische Physik (Bieri)
05 Faculty of Science
UniBasel Contributors:Bieri, Oliver
Item Type:Thesis
Thesis Subtype:Doctoral Thesis
Thesis no:13076
Thesis status:Complete
Number of Pages:1 Online-Ressource (nicht paginiert)
Language:English
Identification Number:
edoc DOI:
Last Modified:08 Feb 2020 15:09
Deposited On:13 Jun 2019 08:44

Repository Staff Only: item control page