磁共振脉冲设计优化算法研究进展  

Progress on the algorithms of optimal pulse design in MR

作　　者：蒋瑞瑞[1,2,3] 常严[3] 徐雅洁[3] 张广才[3] 王骁冠[1,2,3] 杨晓冬[3] 

机构地区：[1]中国科学院长春光学精密机械与物理研究所,吉林长春130033 [2]中国科学院大学,北京100049 [3]中国科学院苏州生物医学工程技术研究所医学影像室,江苏苏州215163

出　　处：《中国医学影像技术》2013年第12期2041-2045,共5页Chinese Journal of Medical Imaging Technology

基　　金：国家自然科学基金项目(11105096);中国科学院科研装备研制项目(YZ201253);苏州科技项目(SH201207)

摘　　要：脉冲优化设计是进行磁共振波谱分析、磁共振成像等的关键技术,目前常用的优化工具是最优控制理论。本文首先简要介绍MR最优控制理论,在此基础上对梯度上升算法、Krotov算法、伪谱法等三种常用的脉冲优化算法进行介绍和对比,并针对实际应用中存在的带宽、射频场不均匀性、弛豫等影响因素给出相应的算法扩展。It is essential to develop and optimize the pulse for MR spectroscopy and MR imaging, and optimal control method is a frequently used tool for pulse design. In this article, a brief introduction for the theoretical model of the optimal control was made, and 3 extensively studied optimal control algorithms, including gradient ascent pulse engineering （GRAPE） algorithm, Krotov algorithm and Pseudospectral method were summarized and compared. For practical applications, it is often desirable to achieve the optimal performance for a range of parameters, such as chemical shift, radio-frequency inhomogeneity and relaxation. The corresponding extensions of the algorithms were given respectively in this article.

关 键 词：脉冲优化 最优控制理论 GRAPE算法 Krotov算法 伪谱法 

分 类 号：R445.2[医药卫生—影像医学与核医学]