基于Granger因果分析的阿尔兹海默病静息态脑电网络连接特性的研究

Study on Brain Networks Connectivity Specificity Using Resting State Electroencephalography Based on Granger Causal Analysis in Alzheimer's Disease

目的从多通道脑电(electroencephalography,EEG)的功能性连接角度,研究阿尔兹海默病患者在静息态下的脑网络连接特性。方法收集15例AD患者及15例正常对照组的16导脑电数据,对EEGs进行频域Granger因果分析,计算出因果连接矩阵,平均后得到DTFmean。基于连接矩阵构建因果网络,进一步计算网络特征参数全局效率和集群系数来定量分析脑网络的连接特性。结果在alpha,low-beta和high-beta频段,AD组的DTFmean低于正常对照组(P<0.05)。在alpha,low-beta和high-beta频段下AD组的全局效率显著低于正常对照组(P<0.01);AD组集群系数显著低于正常对照组(P<0.01)。结论阿尔兹海默病患者的静息态脑电在alpha,low-beta和high-beta频段的脑网络连接较弱,表明AD患者的脑功能网络相较于正常老年人有所受损。

Objective To analyze the resting-state brain networks connectivity specificity of Alzheimer＇s disease from functional connectivity of multi-channel EEGs. Methods Sixteen channels EEG data,which were recorded from 15 cases of AD patients and 15 cases of healthy control subjects,were analyzed by frequency domain Granger causality. Causal connection matrix was calculated and DTFmean was obtained by averaging.Causal network was built to calculate the topological parameters including the global efficiency and clustering coefficient,for quantitatively analyzing the connectivity of brain network. Results The study found that in alpha,low-beta and high-beta band,DTFmean was significantly lower in Alzheimer patients as compared with the control subjects（ P〈0. 05）. In alpha,low-beta and high-beta band,the global efficiency was significantly decreased in AD group（ P〈0. 01）,as compared with the healthy control subjects. And in alpha,low-beta and high-beta band,the clustering coefficient was also significantly decreased in AD patients as compared with the control subjects（ P〈0. 01）. Conclusion AD patients display a loss of brain functional connectivity in alpha,low-beta and high-beta band of the resting-state EEG. The results showed that the brain functional network of AD patients was impaired as compared with the normal eldly.