中文摘要：

目的 探讨抑郁症患者与正常人脑结构网络节点连接强度、聚类系数和局部效率属性的异同,分析抑郁症患者大脑各区域的局部连接模式和局部信息传递效率改变及其与疾病严重程度的相关性.方法 对24例抑郁症患者、25名正常人进行弥散张量成像扫描（Diffusion Tensor Imaging,DTI）,利用解剖学自动标记模板（AAL）将整个大脑划分为90个区域,采用FACT（fiber assignment by continuous tracking）方法进行纤维跟踪,结合基于图论的复杂网络方法构建脑结构网络,并对所得抑郁症患者与正常人脑结构网络的节点局部属性值进行双样本t检验.结果 （1）抑郁症组局部效率属性值（左侧额中回眶部：0.64±0.30、左侧海马：0.57±0.07、右侧海马旁回：0.50±0.15）较对照组（0.88±0.10、0.64±0.06、0.66±0.13,P=0.00098、0.00039、0.00017,通过FDR校正）显著下降;（2）抑郁症组聚类系数属性值（左侧额中回眶部：0.14±0.07）较对照组（0.22±0.06,P=0.000030,通过FDR校正）显著下降;（3）抑郁症组左侧额中回眶部局部效率属性值与HAMD-17总分呈显著负相关（r=-0.48,P=0.02）.结论 抑郁症患者大脑额叶相关脑区连接的集团化程度及局部信息传递效率均发生下降,海马结构的局部信息传递效率也发生下降;其中,额叶相关脑区的局部信息传递效率下降导致疾病严重程度增加.

英文摘要：

Objective To explore the differences of the connectivity strength,the clustering coefficient and the local efficiency of the nodes in the brain structural networks in the depression and healthy subjects and then analyze the mode of the local connections of the brain regions and their local efficiency of the transmission of information and their relationship with the severity of the disease in the depression.Methods The Diffusion Tensor Imaging data were obtained from 24 depression patients and 25 healthy controls.The whole cerebral cortex was parcellated into 90 regions by the anatomical label map.Tractography was performed in the whole cerebral cortex of each subject to reconstruct white matter tracts of the brain using the fiber assignment by continuous tracking （FACT） algorithm.And then the brain structural networks were constructed using the complex network theory.The local topological properties of the brain structural networks of the depression and healthy were examined by two sample t-test.Results The local efficiency of the nodes of the networks in depression decreased significantly （the left middle frontal gyrus （orbital part）：0.64±0.30,the left hippocampus：0.57±0.07,the right parahippocampal gyrus：0.50±0.15） compared with the healthy （0.88±0.10,0.64±0.06,0.66±0.13 respectively,P=0.00098,0.00039,0.00017,survived critical FDR threshold for multiple comparisons） ; and the clustering coefficient of the nodes of the networks in depression （the left middle frontal gyrus （orbital part）：0.14±0.07） decreased significantly when compared with the healthy （0.22±0.06,P=0.000030,survived critical FDR threshold for multiple comparisons）.Significant negative correlation was found between the local efficiency of the left middle frontal gyrus （orbital part） and the total scores of HAMD-17 in the depression （r=-0.48,P=0.02）.Conclusion The degree of the localization and the local efficiency of the information transmission of the frontal lobe are decreased.The local effici