Analysis of a phase synchronized functional network based on the rhythm of brain activities

Rhythm of brain activities represents oscillations of postsynaptic potentials in neocortex, therefore it can serve as an indicator of the brain activity state. In order to check the connectivity of brain rhythm, this paper develops a new method of constructing functional network based on phase synchronization. Electroencephalogram (EEG) data were collected while subjects looking at a green cross in two states, performing an attention task and relaxing with eyes-open. The EEG from these two states was filtered by three band-pass filters to obtain signals of theta (4--7 Hz), alpha (8--13 Hz) and beta (14--30 Hz) bands. Mean resultant length was used to estimate strength of phase synchronization in three bands to construct networks of both states, and mean degree K and cluster coefficient C of networks were calculated as a function of threshold. The result shows higher cluster coefficient in the attention state than in the eyes-open state in all three bands, suggesting that cluster coefficient reflects brain state. In addition, an obvious fronto-parietal network is found in the attention state, which is a well-known attention network. These results indicate that attention modulates the fronto-parietal connectivity in different modes as compared with the eyes-open state. Taken together this method is an objective and important tool to study the properties of neural networks of brain rhythm.     

基于因子分析方法的相位同步脑电源的时-空动力学分析

头皮脑电时间序列的相关性是大脑皮层源的相位同步性的一种体现,因此对相位同步源进行定位,同时找到源对应的时间序列在脑成像研究领域具有重要意义.基于Rössler 模型提出仿真相位同步偶极子源的时间序列的方法,利用时间序列进行同心四层球头模型正演,获得仿真头皮脑电数据.提出了基于最大似然因子分析的相位同步脑电源的时-空动力学分析方法,对仿真和真实头皮脑电数据进行了验证,并与主成分分析法进行对比.仿真实验结果表明:最大似然因子分析法估计的时间序列与仿真源的时间序列具有更高的相关系数,同时估计源与仿真源 关键词： 脑电图 相位同步 因子分析 主成分分析