Quantitative Susceptibility Mapping in Human Brain: Short Echo Timing, Thin Slab, and Non-harmonized Multisite Applications

• Author / Creator

  Naji, Nashwan A

• Quantifying tissue magnetic susceptibility provides a non-invasive way to monitor iron and myelin abnormalities associated with several neurodegenerative diseases. Quantitative susceptibility mapping (QSM) is a post-processing technique that retrieves the susceptibility distribution from the phase of T2*-sensitive magnetic resonance (MR) images, providing localized and discriminative contrast of the underlying sources. QSM is typically obtained from a three-dimensional (3D) multi-echo gradient-echo sequence, though it can also be produced from single-echo sequences such as echo-planar imaging with some sacrifice on resolution and/or quality of the resultant map.

  QSM methodology has improved notably over the last decade addressing several technical challenges. However, it is still not commonly included in clinical protocols, mostly due to the burden of its lengthy acquisition time. Moreover, even focal acquisitions require extended 3D coverage beyond the region of interest, that otherwise might be imaged in a few slices. This acquisition burden becomes much heavier when exploring higher spatial resolution. Furthermore, QSM multisite reproducibility has been studied mainly under the assumption that the parameters of the imaging sequence are fixed and replicable, which does not consider possible sequence variations typically found in large-scale multisite studies. Using in-vivo and simulated MR data of human brain, this thesis investigates the above-mentioned limitations and introduces new approaches to further facilitate QSM integration into clinical applications. To minimize the acquisition burden, producing QSM from 3D magnetization-prepared rapid gradient-echo (MPRAGE) phase data was proposed and possible applications were explored. In addition, a new method was introduced to allow QSM from thin slabs aided by a rapid low-resolution scan. Furthermore, QSM multisite reproducibility was investigated in the presence of sequence variations.

  The MPRAGE sequence is widely used in clinical studies to segment gray and white matter regions. Only MPRAGE magnitude is utilized for segmentation and its phase is abandoned. In Chapter 2, producing QSM from the MPRAGE phase was introduced and investigated, and possible applications were explored. Despite the limited contrast at 3T and very-short echo time (TE), MPRAGE-QSM was found useful for improving segmentation of iron-rich regions and roughly quantifying their susceptibility. Thus, producing MPRAGE-QSM adds value to volumetric studies at no additional cost. Also, the quality of MPRAGE-QSM can be improved notably if extending TE to 4.4 ms at 3T is tolerable.

  Another important application of MPRAGE-QSM is assessing the load of cerebral microbleeds (CMBs) without adding an extra sequence (dedicated to QSM) to the imaging protocol. Being mainly hemosiderin deposits, CMBs introduce strong contrast on QSM within a few milliseconds of echo time. Utilizing MPRAGE-QSM for quantifying microbleeds was explored in Chapter 3 and the results showed that it is promising at field strength of 3T and above.

  In Chapter 4, a new method is proposed to accelerate QSM data acquisition by allowing QSM from thin slabs. To achieve this, a rapid low-resolution scan with wider coverage is used to roughly inform the reconstruction algorithm of the expected susceptibility distribution outside the imaged slab. Compared to the standard approach, simulation and in-vivo results showed that applying the new method improved QSM measurements from as few as 8 slices with aid from low-resolution data of at most 4-times larger voxel dimensions, potentially allowing up to 7-fold reduction in acquisition time.

  Studying QSM reproducibility with non-harmonized sequence parameters is an important step toward clinical integration as matching sequence specification in each site is not always possible due to hardware and/or software limitations. In Chapter 5, QSM and R2* reproducibility were studied using 24 subjects who travelled between three sites and were scanned using 3T scanners from two vendors. Each site optimized the sequence parameters independently. Cross-site and within-site measurements for QSM and R2* were found to be reproducible and highly correlated. Certain post-processing choices for QSM helped in reducing cross-site variability, such as excluding less reliable regions, matching spatial resolution and echo-timings, and minimizing streaking artifacts.

  To conclude, this thesis proposed new methods to minimize the time burden of including QSM in clinical protocols by utilizing phase from a common MPRAGE volumetric acquisition to examine iron-rich regions and microbleeds, or by speeding up focal QSM acquisitions. In addition, QSM multisite reproducibility was examined when the acquisition sequence is not harmonized, and insights were provided on reducing the effects of cross-site variations via post-processing techniques.

• Subjects / Keywords

  • QSM
  • Quantitative susceptibility mapping
  • reproducibility
  • MPRAGE-QSM
  • thin slab QSM
  • 3T
  • Human brain
  • MRI

• Graduation date

  Spring 2023

• Type of Item

  Thesis

• Degree

  Doctor of Philosophy

• DOI

  https://doi.org/10.7939/r3-2het-ef03

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.