基于图论分析对自闭症谱系障碍儿童脑功能网络异常拓扑学属性的研究

The abnormal topological properties of brain functional network in children with autism spectrum disorders based on graph theory analysis

目的 探讨自闭症谱系障碍(autism spectrum disorder,ASD)儿童与正常发育(typically developing,TD)儿童的脑功能网络拓扑结构差异。材料与方法 从国际自闭症脑成像交换数据库的4个站点中提取符合纳入标准的65例ASD儿童(ASD组)及65例TD儿童(TD组)的脑部静息态功能磁共振成像及T1加权结构像数据。应用图论分析研究大脑网络的拓扑结构,包括全局和局部拓扑参数。结果 在阈值范围内,ASD在聚类系数(clustering coefficient,Cp)和特征路径长度(characteristic path length,Lp)低于TD组(P<0.05);两组在全局效率(global efficiency,Eglob)、局部效率(local efficiency,Eloc)等指标的差异无统计学意义。ASD组和TD组的脑功能连接网络满足标准化的聚类系数(γ)>1和标准化的特征路径长度(λ)≈1,具有小世界网络属性,但两组间的γ和λ值差异无统计学意义。节点水平分析结果显示ASD组在双侧前扣带回、双侧尾状核、双侧海马、右侧顶下小叶等脑区的节点效率较TD组减低(P<0.05)。结论 与TD组相比,ASD组大脑拓扑学改变仍符合小世界网络属性,但有向随机网络转变的趋势,其功能整合能力增强,功能分化能力降低,某些脑区的节点效率降低,为探讨ASD的发病机制以及解释ASD临床症状提供了一个新的角度。

Objective: To investigate the topological reorganization of brain functional networks in autism spectrum disorder(ASD) and typically developing(TD) children. Materials and Methods: In this study, resting-state functional magnetic resonance imaging from the Autism Brain Imaging Data Exchange data(65 patients with ASD, 65 TD as controls) was applied to investigate the topological architecture of the brain network including global and local parameters using the graph theoretical analysis. Results: Complex network analysis based on graph theory showed that children with ASD demonstrated shortened characteristic path length(Lp) and clustering coefficient(Cp) compared with TD group. Small world network properties were demonstrated in both ASD and TD groups with γ>1, λ≈1.No significant differences were found in neither γ nor λ. Significantly lower nodal efficiency was observed in bilateral anterior cingulate cortex, caudate nucleus, hippocampus and right inferior parietal lobule in ASD group compared with TD group. Conclusions: Small-world architecture in brain functional networks was identified for both ASD and TD groups with a tendency to turn into random network.whatever, the ASD groups exhibit increased functional integration and decreased functional segregation of the brain functional networks.And the node efficiency of some brain regions was reduced. Our study provides a new perspective for exploring the pathogenesis and the clinical symptoms of ASD.