基于格兰杰因果的效应性连接分析方法综述  

Review of the Analysis Methods of Effective Connectivity Based on Granger Causality

作　　者：尚志刚[1,2] 沈晓阳 李蒙蒙[1,2] 万红 SHANG Zhigang;SHEN Xiaoyang;LI Mengmeng;WAN Hong(School of Electrical Engineering, Zhengzhou University, Zhengzhou 450001, China;Henan Key Laboratory of Brain Science and Brain-Computer Interface Technology, Zhengzhou University, Zhengzhou 450001, China)

机构地区：[1]郑州大学电气工程学院,河南郑州450001 [2]郑州大学河南省脑科学与脑机接口技术重点实验室,河南郑州450001

出　　处：《郑州大学学报（工学版）》2020年第3期1-7,13,共8页Journal of Zhengzhou University（Engineering Science）

基　　金：国家自然科学基金资助项目(61673353,U1304602)。

摘　　要：基于格兰杰因果关系的效应性连接分析方法目前在多脑区神经信号分析中得到了越来越多的应用。首先系统介绍了此类方法中常用代表性算法的计算原理与功能特点,然后对实际应用该类方法时应注意的要点进行了总结,最后以广义偏定向相干的改进算法为例,展示了在实际脑电数据集上的应用效果。At present,the effective connectivity analysis methods based on Granger causality was widely used in neural signals analysis of multiple brain regions.First of all,the calculation principle and functional characteristics of representative algorithms commonly used in this kind of method were systematically introduced.Then the key points that should be paid attention to in practical application of this kind of methods were summarized.Finally,the improved algorithm for Generalized Partial Directed Coherence was taken as examples to show the application effect on the actual electroencephalogram data set.

关 键 词：格兰杰因果 效应性连接 神经信号 信息流 

分 类 号：R318[医药卫生—生物医学工程]