基于弥散张量成像的脑结构网络参数研究

目的 利用复杂网络参数评估大脑特征,是探明大脑工作机制的新思路。方法 本研究基于弥散张量纤维束追踪和小世界网络理论构建脑结构网络。首先利用自动解剖标定（automated anatomical labeling,AAL）模板对大脑分区,并计算脑区间的纤维连接情况,对正常人的大脑结构进行网络建模。然后分析脑结构网络中节点度、簇系数和节点介数等网络参数,并通过引入小世界网络的介数和损伤性定义人脑结构网络的核心节点位置和特征。结果 脑结构网络具有小世界属性,且网络中存在少量的核心节点,具有较高的节点度和簇系数值。结论 利用弥散张量成像能够客观构建大脑结构网络,而核心节点的存在为揭示脑疾病的病理生理机制提供新的思路。

Research of brain structure network parameters based on diffusion tensor imaging

Objective To assess the brain characteristics by using complex network parameters is a new idea to ascertain the brain working mechanism.Methods In this study,we built structural network based on diffusion tensor fiber bundle tracking and small-world network theory.The fiber connections between each AAL brain region were extracted to model the normal brain structure.The location and characteristics of hub nodes were defined by using the node betweenness and vulnerability.The network node degree,clustering coefficient and node betweenness were then calculated.Results The brain structure networks exhibited efficient smallworld properties and there were a few hub nodes in the brain which had high node degree and clustering coefficient.Conclusions Brain structure networks could be objectively constructed by diffusion tensor imaging and the existence of hub nodes provided new ideas for the pathophysiology of brain disease.