用于同时多层MRI的选择性射频脉冲的优化设计

近年来，为提高磁共振成像（MRI）信号信噪比（SNR）、缩短成像时间，同时多层成像技术受到了极大的关注.为了实现同时多层的选择性激发，现有的多层成像序列大多使用组合射频（RF）脉冲，该脉冲可包含多个独立的幅值相同相位不同的简单脉冲，由于其采用简单的线性叠加方法，该类脉冲射频功率随脉冲数量呈现平方增长，因而应用受限.针对这一问题，基于自旋动力学和优化控制原理，本文提出了一种针对同时多层MRI的选择性射频脉冲的数值优化方法，该方法充分运用射频脉冲的调控机制，获得优化脉冲，并配合层选梯度，可实现任意层厚、层间距、层数的同时高效选择性激发.最后，通过数字模体的同时多层模拟成像实验验证了优化脉冲的有效性.

Optimization of Selective Radio Frequency Pulses for Simultaneous Multi-Slice MRI

Simultaneous multi-slice magnetic resonance imaging (MRI) has attracted much research interests recently due to its advantages in increasing signal-to-noise ratio(SNR)and decreasing acquisition time.To achieve simultaneous multi-slice excitation, the existing methods used a composite excitation pulse, yielded from the linear combination of radio frequency (RF) pulses with the same amplitude but different phases. The performance of such composite pulse, however, is limited, due to the square relationship between the peak pulse power and slice number. To solve this problem, we proposed a numerical optimization method for multi-slice excitation RF pulse based on both spin dynamics and optimal control theory. Together with slice selection gradient, the optimized pulse achieved selective excitation with arbitrary thickness, distance and number of slices. The performance of the optimized pulse in simultaneous multi-slice MRI was validated by simulation on multimodal imaging-based detailed anatomical (MIDA) digital phantom.