| The recent trend in magnetic resonance (MR) imaging has been to deploy higher field systems to exploit the larger magnetic resonance signal and signal-to-noise ratio (SNR) predicted by the underlying physics. Three Tesla systems have been approved for clinical use by the United States Food and Drug Administration, and fields above 7 T are now routinely used for animal imaging. With stronger magnetic fields also come new technical challenges. The shorter T*2 and the accompanying signal dropout, as well as increased geometric distortions, are the primary obstacles that must be overcome when dealing with magnetic field perturbations at high main field strengths.;One solution to this problem is to acquire the MR signal at a higher bandwidth, which requires stronger and faster magnetic field gradients and higher bandwidth receivers for acquisition. These improvements entail upgrading gradient coils, gradient amplifiers, and analog to digital (A/D) hardware, which generally require extensive hardware modifications to current MR systems.;Another method is to reduce susceptibility effects by improving magnetic field uniformity. Achieving magnetic field uniformities better than one part per million is possible using a variety of active and passive shimming techniques already published in the literature. There are also several pulse sequences and protocols that can be used to compensate for through-slice susceptibility effects. This work focuses on relatively minor hardware enhancements that can be easily implemented on almost any MR system, and the accompanying pulse sequences and processing software that can help compensate for magnetic field inhomogeneities.;The initial chapters of this thesis detail enhancements to a technique developed by Jesmanowicz to compensate for high order magnetic field perturbations in the human brain. First, the limits of the technique are explored via simulations. From these simulations, parameters for mapping the magnetic field perturbations in a human brain are determined. The simulations also define the optimal distribution of magnetic material on the surface of a cylinder surrounding the head that is needed to correct the measured perturbations. The magnetic dipole model will be used to approximate the effect of the material on the main magnetic field.;Several dipole distributions have been created and used to correct the magnetic field perturbations within the brain of a subject.;Later chapters in this work describe a new echo-planar type of pulse sequence in which magnetic field maps can be collected and computed in real time. (Abstract shortened by UMI.). |
