Diffusion Tensor Imaging at High Magnetic Field

• Author / Creator

  Baron, Corey A

• Diffusion magnetic resonance imaging (dMRI) measures the diffusion (i.e. random molecular motion) of water. Since the motion of water is inhibited by cellular membranes, dMRI provides insight into the microstructural characteristics of the tissue. However, distinguishing between small anatomical subdivisions can be difficult due to dMRI methods having an inherently low resolution. Increasing the resolution necessarily reduces the signal-to-noise ratio, and to compensate for this a higher static magnetic field and larger diffusion gradients can be utilized. In addition, there are newly emerging dMRI techniques that have not been successfully implemented in the human brain due to insufficient signal and gradient strengths on clinical MRI systems. However, increasing these parameters introduces new challenges. The goal of this thesis was to address the challenges of performing dMRI with a stronger magnetic field and gradients, and then to use the extra signal for higher resolution and to enable new techniques that require more signal. To do this, a 4.7 T human MRI system with 60 mT/m gradients was utilized, which is a three times stronger field and 50\% larger gradients compared to typical clinical scanners. Approximately half of the thesis work involved solving challenges at high field or gradient strengths, while the remainder involved applications enabled by the high strength MRI system. The first challenge investigated was a previously undocumented issue for dMRI introduced by strong gradients, concomitant gradient fields. Errors introduced by these concomitant fields was found to be considerable in certain cases, and techniques to mitigate them were explored. Another challenge involved developing robust methods to perform parallel imaging, which is a technique used to prevent distortions that worsen at higher field strengths. A final challenge investigated was errors introduced by the unwanted signal that stems from cerebrospinal fluid. Traditional approaches to mitigate the error do not translate well to high field, and an alternative method was sought and characterized. A potential application of the high strength MRI system involves the probing of different tissue microstructure length scales. Typical dMRI techniques are only sensitive to length scales longer than typical microstructural dimensions because of a long "diffusion time". However, the newly developed oscillating gradient spin-echo (OGSE) technique is more sensitive to smaller length scales because it can achieve much shorter diffusion times, which may give new insight into healthy development or disease. Accordingly, OGSE was used to investigate the microstructural length scale dependence of dMRI as a function of diffusion direction for the first time in healthy subjects and in patients with stroke. The former subject group was required to better understand the healthy brain and provide a reference point for comparison with disease. The latter cohort of subjects helps to elucidate the underpinnings of why standard dMRI is sensitive to stroke, which is still not well understood even though dMRI is routinely used in its diagnosis. In addition, dMRI of stroke is traditionally performed using thick slices to maintain low scan times. Accordingly, by utilizing a high resolution dMRI sequence in stroke patients, it was found in this thesis that thinner slices yield more precise measurements of lesion volume and diffusion parameters. In summary, the thesis work shows that dMRI can be successfully translated to large magnetic field and gradient strengths. The advantages of doing so are the ability to perform novel dMRI techniques and improved performance of existing techniques. While this work was performed at a field strength not commonly utilized for human MRI (4.7 Tesla), all the work described in this thesis could be translated to 7 Tesla or 3 Tesla with modern radiofrequency coil arrays.

• Subjects / Keywords

  • Diffusion
  • DTI
  • MRI
  • High field

• Graduation date

  Fall 2014

• Type of Item

  Thesis

• Degree

  Doctor of Philosophy

• DOI

  https://doi.org/10.7939/R3R20S256

• 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.