Files in this item



application/pdfChao_Ma.pdf (3MB)
(no description provided)PDF


Title:Design of multidimensional radiofrequency pulses in magnetic resonance imaging
Author(s):Ma, Chao
Director of Research:Liang, Zhi-Pei
Doctoral Committee Chair(s):Liang, Zhi-Pei
Doctoral Committee Member(s):Bresler, Yoram; Georgiadis, John G.; Sutton, Bradley P.
Department / Program:Electrical & Computer Eng
Discipline:Electrical & Computer Engr
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):magnetic resonance imaging
radiofrequency pulses
multidimensional radiofrequency pulses
Shinnar-Le Roux radiofrequency pulse design
parallel excitation
spoke trajectory
nonlinear gradient
Abstract:Multidimensional radiofrequency (RF) pulses have wide applications in magnetic resonance imaging and spectroscopy experiments. These applications include reduced field-of-view (FOV) imaging of region-of-interest (ROI), B1 inhomogeneity correction, and simultaneous spatial-spectral selective excitation. While an elegant method known as Shinnar-Le Roux (SLR) method has been developed in one-dimensional RF pulse design, multidimensional RF pulse design remains an open question. The primary challenges are: a) the nonlinearity of the governing Bloch equation, inaccurate treatment of which can cause excitation pattern distortions in the large-tip-angle regime; and b) the long duration of multidimensional RF pulses, which can cause signal loss and excitation pattern distortions. The research in this thesis has developed new RF pulse design methods to address these two challenges. To address the first challenge, a novel approach is proposed to generalize the conventional SLR method to the multidimensional case. In the proposed method, the multidimensional RF pulse design problem is converted to a series of 1D polynomial design problem, each of which is efficiently solved as a convex optimization problem. The proposed method is able to accurately treat the nonlinearity of the Bloch equation, generate excitation patterns with equiripple errors, make explicit tradeoff of design parameters and allow fast computation. The proposed method is further generalized to correct B0 inhomogeneity effects and to design spatial-spectral RF pulses. The effectiveness of the proposed method is demonstrated through representative design examples using simulation and experimental results. To address the second challenge, two methods are proposed. In the first method, we propose a novel joint design of gradient and RF waveforms to shorten spoke trajectory based RF pulses in parallel excitation. The joint design problem is formulated as an optimal spoke selection problem and is solved using a greedy algorithm. Simulation and experimental results demonstrate that the proposed method is able to achieve superior excitation accuracy with high computational efficiency compared with conventional methods. In the second method, we propose a novel reduced FOV excitation pulse using second-order gradients and a spatial-spectral RF pulse. By leveraging the unique multidimensional spatial dependence of second-order gradients, the proposed method is able to achieve 3D spatial selectivity, i.e., a circular ROI in a thin slice, using a 2D spatial-spectral RF pulse. Simulation and experimental results demonstrate that the proposed method can significantly improve excitation profiles compared with conventional methods. In addition, we have performed a perturbation analysis to investigate the effects of B1 mapping errors on excitation accuracies in parallel excitation. A closed-form solution of the perturbations on the excitation patterns caused by B1 mapping errors is derived by locally linearizing the Bloch equation. Through the perturbation analysis, we show that the excitation errors are increasingly sensitive to B1 mapping errors as the tip-angle and reduction factor of an RF pulse increase.
Issue Date:2013-08-22
Rights Information:Copyright 2013 Chao Ma
Date Available in IDEALS:2013-08-22
Date Deposited:2013-08

This item appears in the following Collection(s)

Item Statistics