Tailored RF Pulse Design for 7T MRI

The primary concern of this thesis is in the design of RadioFrequency (RF) Pulses to be used for Ultra-High Field Magnetic Resonance Imaging. A movement towards higher magnetic field strength has noted advantages. In particular, increased Signal to Noise Ratio (SNR) with increasing field strength. However higher field strength also poses difficult problems, as a result of increasing inhomogeneity in the main magnetic field (B0) as well the Radiofrequency field (B1). RF pulses were designed for use as inversion, refocusing and excitation pulses. This was achieved by the use of Genetic Algorithms with an associated evaluation function. Such functions relate to the performance of pulses in the presence of inhomogeneities. Each pulse was described by a prescribed number of parameters N, which can then generate the required RF functions. The problem then becomes one of a directed search in N dimensional space. The Genetic Algorithms, used for this work evolved to meet the needs of the problems posed while also being informed by exploratory searches. Hill Climbing methods were also included after the Genetic Algorithm had terminated. Such methods are good at locating local maxima/minima and hence only included at the end to avoid globally weak solutions. In order to test the performance of pulses, a program which simulates