基于自回归模型和相位斜率指数的动态脑连接分析

doi:10.3969/j.issn.0258-8021. 2016.01.001

摘要
图/表
参考文献
相关文章 (0)

全文: PDF (5089 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要随着脑网络理论的发展，人们越来越关注不同脑区域之间的功能性和效应性连接。在常见的脑活动无创检测方法中，脑电图（EEG）具有较高的时间分辨率，适于进行效应性连接分析。提出一种估计不同通道的EEG信号间的效应性连接的方法，即自回归相位斜率指数（AR-PSI）。该方法结合多元自回归（MVAR）模型对短时数据进行谱估计频率分辨率高的特点和相位斜率指数(PSI)对源信号混叠不敏感的特点。与传统的格兰杰因果模型相比，它可以有效地摒除由于容积导体效应造成的信号混叠所带来的干扰；与传统的PSI相比，它在短时数据上能够更准确地估计不同通道间EEG信号的效应性连接。首先，分别生成具有强效应性连接和相互独立的混叠噪声这两组信号进行模拟实验，结果表明AR-PSI方法确实能够更有效地检测出信号中存在的效应连接，排除信号线性混叠可能引起的误检。然后，应用此方法并结合滑动窗技术，对Stroop实验记录到的EEG数据进行网络连接的动态分析，发现250～500 ms和550～800 ms时间段内脑网络连接密度在两种条件下存在显著性差异。分析结果，显示语义和颜色不一致条件的刺激能够引起脑网络连接密度更迅速地增加，且连接跨度更大。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	黄亮吴超华高小榕^#*

关键词 ：效应性连接, MVAR模型, 相位斜率指数(PSI), 动态分析

Abstract：Functional and effective connectivity are important branches in brain network research. Electroencephalogram (EEG) has sufficient temporal resolution to catch the fast brain dynamic changes, so it is suitable for effective connectivity analysis. We proposed a new method to estimate the effective connectivity base on EEG, namely autoregressive phase slope index (AR-PSI). Spectral estimation based on MVAR has high frequency resolution even on short time data. PSI is insensitive to linear mixture of noninteracting sources. AR-PSI combined advantages of the two methods. Compared with conventional Granger causality model, AR-PSI could eliminate the interference caused by volume conduction. Compared with conventional PSI, AR-PSI could get more accurate estimation of effective connectivity with short time data. Experimental data indicated that AR-PSI could exactly detect the effective connectivity between two signals and exclude the false detection. AR-PSI was also applied to dynamic brain connectivity analysis based on EEG recorded in Stroop paradigm. We found that brain connectivity density had significant difference under two conditions in 250～500 ms and 550～800 ms. The results indicated that the incongruent stimulus could make the density of brain network increase more quickly and expand the span of connectivity.

Key words： effective connectivity MVAR model phase slope index( PSI) dynamic analysis

基金资助:国家自然科学基金重点项目（61431007, 91320202, 91220301）

引用本文:

黄亮吴超华高小榕^#*. 基于自回归模型和相位斜率指数的动态脑连接分析[J]. 中国生物医学工程学报, 2016, 35(1): 1-9.
Huang Liang Wu Chaohua Gao Xiaorong^#*. Dynamic Brain Connectivity Analysis Based on Autoregressive-Model and Phase Slope Index. journal1, 2016, 35(1): 1-9.

链接本文:

http://cjbme.csbme.org/CN/10.3969/j.issn.0258-8021. 2016.01.001 或 http://cjbme.csbme.org/CN/Y2016/V35/I1/1

［1］Sporns O, Tononi G, Ktter R. The human connectome: a structural description of the human brain［J］. PLoS Computational Biology, 2005, 1(4): e42.
［2］蒋田仔, 刘勇, 李永辉. 脑网络：从脑结构到脑功能［J］. 生命科学, 2009（2）:181-188.
［3］梁夏, 王金辉, 贺永. 人脑连接组研究：脑结构网络和脑功能网络［J］. 科学通报, 2010(16): 1565-1583. 
［4］孙俊峰, 洪祥飞, 童善保. 复杂脑网络研究进展——结构、功能、计算与应用［J］. 复杂系统与复杂性科学, 2011, 7(4): 74-90. 
［5］Bullmore E, Sporns O. The economy of brain network organization［J］. Nature Reviews Neuroscience, 2012, 13(5): 336-349.
［6］Van Essen DC, Smith SM, Barch DM, et al. The WUMinn human connectome project: an overview［J］. Neuroimage, 2013, 80: 62-79.
［7］Sporns O. Making sense of brain network data［J］. Nature Methods, 2013, 10(6): 491-493.
［8］Uehara T, Yamasaki T, Okamoto T, et al. Efficiency of a “smallworld” brain network depends on consciousness level: a restingstate FMRI study［J］. Cerebral Cortex, 2014, 24(6): 1529-1539.
［9］Koch MA, Norris DG, HundGeorgiadis M. An investigation of functional and anatomical connectivity using magnetic resonance imaging［J］. Neuroimage, 2002, 16(1): 241-250.［10］Fingelkurts AA, Fingelkurts AA, Khknen S. Functional connectivity in the brain—is it an elusive concept?［J］. Neuroscience & Biobehavioral Reviews, 2005, 28(8): 827-836.［11］Horwitz B. The elusive concept of brain connectivity［J］. Neuroimage, 2003, 19(2): 466-470.
［12］Handbook of Time Series Analysis: Recent Theoretical Developments and Applications［M］.
New York: John Wiley & Sons, 2006.
［13］Nolte G, Ziehe A, Nikulin VV, et al. Robustly estimating the flow direction of information in complex physical systems［J］. Physical Review Letters, 2008, 100(23): 234101.
［14］Haufe S, Nikulin VV, Müller KR, et al. A critical assessment of connectivity measures for EEG data: a simulation study［J］. Neuroimage, 2013, 64(1):120-133.
［15］Lütkepohl H. New Introduction to Multiple Time Series Analysis［M］. Berlin: Springer Science & Business Media, 2007.
［16］Schneider T, Neumaier A. Algorithm 808: ARfit—A Matlab package for the estimation of parameters and eigenmodes of multivariate autoregressive models［J］. ACM Transactions on Mathematical Software (TOMS), 2001, 27(1): 58-65.
［17］Morf M, Vieira A, Lee DTL, et al. Recursive multichannel maximum entropy spectral estimation［J］. IEEE Trans
Geosci Electron, 1978, 16(2): 85-94.
［18］ Marple Jr SL. Digital Spectral Analysis with Applications［M］. Englewood Cliffs: PrenticeHall, 1987.
［19］Schlgl A. The Electroencephalogram and the Adaptive Autoregressive Model: Theory and Applications ［M］. Aachen: Shaker, 2000.
［20］Schlgl A. A comparison of multivariate autoregressive estimators［J］. Signal Processing, 2006, 86(9): 2426-2429.
［21］Akaike H. A new look at the statistical model identification［J］. IEEE Transactions on Automatic Control, 1974, 19(6): 716-723.
［22］Hosking JRM. The multivariate portmanteau statistic［J］. Journal of the American Statistical Association, 1980,
75(371): 602-608.
［23］Ding M, Bressler SL, Yang W, et al. Shortwindow spectral analysis of cortical eventrelated potentials by adaptive multivariate autoregressive modeling: data preprocessing, model validation, and variability assessment［J］. Biological Cybernetics, 2000, 83(1): 35-45.
［24］Nolte G, Bai O, Wheaton L, et al. Identifying true brain interaction from EEG data using the imaginary part of coherency［J］. Clinical Neurophysiology, 2004, 115(0): 2292-2307.［25］高文静, 陈小刚, 高小榕. 基于 Hermite 滤波器的稳态视觉诱发电位背景标记 Stroop 效应包络分析［J］. 中国生物医学工程学报, 2014, 31(1): 1-7.
［26］Qiu J, Luo Y, Wang Q, et al. Brain mechanism of Stroop interference effect in Chinese characters［J］. Brain Research, 2006, 1072(1): 186-193.
［27］Liotti M, Woldorff MG, Perez R, et al. An ERP study of the temporal course of the Stroop colorword interference effect［J］. Neuropsychologia, 2000, 38(5): 701-711.
［28］Zysset S, Müller K, Lohmann G, et al. Colorword matching Stroop task: separating interference and response conflict［J］. Neuroimage, 2001, 13(1): 29-36.
［29］Schack B, Chen ACN, Mescha S, et al. Instantaneous EEG coherence analysis during the Stroop task［J］. Clinical Neurophysiology, 1999, 110(8): 1410-1426.
［30］Harrison BJ, Shaw M, Yücel M, et al. Functional connectivity during Stroop task performance.［J］. NeuroImage, 2005, 24(1):181-191.
［31］Hanslmayr S, Pasttter B, Buml KH, et al. The electrophysiological dynamics of interference during the stroop
task［J］. Journal of Cognitive Neuroscience, 2008, 20(2):215-225.
［32］MacLeod CM. Half a century of research on the Stroop effect: an integrative review［J］. Psychological Bulletin, 1991, 109(2): 163-203. 
［33］Marzetti L, Nolte G, Perrucci MG, et al. The use of standardized infinity reference in EEG coherency studies［J］. Neuroimage, 2007, 36(1):48-63.
［34］Yun Q, Peng X, Yao D. A comparative study of different references for EEG default mode network: the use of the infinity reference.［J］. Clinical Neurophysiology, 2010, 121(12):1981-1991.