Abstract
Selective RF pulses are needed or many application in magnetic resonance imaging (MRI). The desired excitation profile is omen used as the spectrum of the applied RF pulse; the modulation waveform of the RF pulse which approximately excites the desired profile is the Fourier Transform of this spectrum. This is a first-order approximation which is most valid for small tip ogles. Known as the ''small-tip-angle approximation,'' the quality of the excitation profile resulting from such RF pulses decreases as the tip angle increases. Since large-tip-angle excitations are required in most types of imaging, a better technique is necessary. In response to this need, many workers have developed more accurate techniques for RF pulse synthesis. These method include inverse solutions to the Bloch equation as well as techniques resembling those employed in digital filter development. A fundamental shortcoming of all of these approaches is the lack of consideration for the physical characteristics of the MR scanner system itself, along with undesirable analytical or computational complexity. The no/deal system transfer function can distort the profile in ways which are not readily predicable using analytic methinks of synthesis. In this thesis, a novel technique for RF pulse refinement is presented which reacts to the physical behavior of the MR scanner, using real-time feedback techniques in lieu of a Bloch equation solution. Numerical and physical experiments are conducted to demonstrate the effectiveness of this algorithm. Finally, common applications for this technique are developed and tested to demonstrate its utility.
Lebsack, Eliot Todd (1999). An iterative technique for refinement of selective excitations for magnetic resonance imaging. Master's thesis, Texas A&M University. Available electronically from
https : / /hdl .handle .net /1969 .1 /ETD -TAMU -1999 -THESIS -L422.