Kernel Granger Causality Mapping Effective Connectivity on fMRI Data

被引：49

作者：

Liao, Wei ^{[1
]}

Marinazzo, Daniele ^{[2
]}

Pan, Zhengyong ^{[3
]}

Gong, Qiyong ^{[4
]}

Chen, Huafu ^{[1
]}

机构：

[1] Univ Elect Sci & Technol China, Key Lab NeuroInformat, Minist Educ, Sch Life Sci & Technol, Chengdu 610054, Peoples R China

[2] Univ Paris 05, Lab Neurophys & Neurophysiol, CNRS, UMR 8119, Paris 06, France

[3] Univ Elect Sci & Technol China, Sch Appl Math, Chengdu 610054, Peoples R China

[4] Sichuan Univ, HMRRC, Dept Radiol, Med Imaging Ctr,W China Hosp, Chengdu 610041, Peoples R China

来源：

IEEE TRANSACTIONS ON MEDICAL IMAGING | 2009年 / 28卷 / 11期

关键词：

Effective connectivity; functional magnetic resonance imaging (fMRI); kernel Granger causality (KGC); CORTICAL MOTOR SYSTEM; FUNCTIONAL CONNECTIVITY; LINEAR-DEPENDENCE; BRAIN; MODULATION; NETWORKS; FEEDBACK;

D O I：

10.1109/TMI.2009.2025126

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

Although it is accepted that linear Granger causality can reveal effective connectivity in functional magnetic resonance imaging (fMRI), the issue of detecting nonlinear connectivity has hitherto not been considered. In this paper, we address kernel Granger causality (KGC) to describe effective connectivity in simulation studies and real fMRI data of a motor imagery task. Based on the theory of reproducing kernel Hilbert spaces, KGC performs linear Granger causality in the feature space of suitable kernel functions, assuming an arbitrary degree of nonlinearity. Our results demonstrate that KGC captures effective couplings not revealed by the linear case. In addition, effective connectivity networks between the supplementary motor area (SMA) as the seed and other brain areas are obtained from KGC.

引用

页码：1825 / 1835

页数：11

共 46 条

[1]

[Anonymous], 1994, An introduction to the bootstrap: CRC press

[2] Partial directed coherence:: a new concept in neural structure determination [J].