Xie Songyun, Li Yabing
NPU-TUP Joint Laboratory for Neural Informatics, Northwestern Polytechnical University, Xi'an, Shaanxi Province, 710129, P. R. China.
School of Computer Science and Technology, Xi'an University of Posts and Telecommunications, Xi'an, Shaanxi Province, 710121, P. R. China.
J Integr Neurosci. 2020 Mar 30;19(1):111-118. doi: 10.31083/j.jin.2020.01.1234.
An effective network perspective focused on measuring directional interactions of electroencephalographic in different cortical regions during a sustained attentive task requiring vigilance. A novel measure referred to as dynamic partial directed coherence was used to map the cognitive state of vigilance based on graph theory. In the right parieto-occipital area, the area is significantly higher than in other regions of interest (the areas are 0.601 and 0.632 for out-degree and in-degree, respectively). A similar analysis in the right fronto-central area revealed significant differences in the different cognitive states. Across the six regions of interest, significant differences of in-degree and out-degree based alpha band are observed in the right fronto-central and the right parieto-occipital ( < 0.05). The performance was compared with those from a support vector machine using different network-based phase-locking values, partial directed coherence, and dynamic partial directed coherence. Results show that dynamic partial directed coherence can provide more information about direction (compared with phase-locking values) and accuracy (when compared with partial directed coherence). The graph theoretical analysis shows that the effective network based dynamic partial directed coherence has a small-world property for synchronizing neural activity between brain regions. Moreover, the alpha band is well correlated with the cognitive state compared to other frequency bands.
一种有效的网络视角专注于在需要警觉的持续注意力任务期间测量不同皮质区域脑电图的定向相互作用。一种称为动态偏相干性的新测量方法被用于基于图论绘制警觉的认知状态。在右侧顶枕区，该区域明显高于其他感兴趣区域（出度和入度分别为0.601和0.632）。在右侧额中央区进行的类似分析揭示了不同认知状态下的显著差异。在六个感兴趣区域中，在右侧额中央区和右侧顶枕区观察到基于α波段的入度和出度存在显著差异（<0.05）。将该性能与使用基于不同网络的锁相值、偏相干性和动态偏相干性的支持向量机的性能进行了比较。结果表明，动态偏相干性可以提供更多关于方向（与锁相值相比）和准确性（与偏相干性相比）的信息。图论分析表明，基于有效网络的动态偏相干性具有小世界特性，用于同步脑区之间的神经活动。此外，与其他频段相比，α波段与认知状态具有良好的相关性。
