基于默认网络内部功能连接能预测抑郁症患者睡眠障碍因子分  

Functional connectivity within the default mode network can predict the sleep disturbance scores of the patients with depression

作　　者：秦姣龙 李弘瑄 吴烨 倪黄晶 QIN Jiaolong;LI Hongxuan;WU Ye;NI Huangjing(Key Lab of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education,School of Computer Science and Engineering,Nanjing University of Science and Technology,Nanjing 210014,China;Jiangsu Key Lab of Image and Video Understanding for Social Security,School of Computer Science and Engineering,Nanjing University of Science and Technology,Nanjing 210014,China;School of Computer Science,School of Software,School of Cyberspace Security,Nanjing University of Posts and Telecommunications,Nanjing 210023,China)

机构地区：[1]南京理工大学计算机科学与工程学院,高维信息智能感知与系统教育部重点实验室,南京210014 [2]南京理工大学计算机科学与工程学院,社会安全图像与视频理解江苏省重点实验室,南京210014 [3]南京邮电大学计算机学院、软件学院、网络空间安全学院,南京210023

出　　处：《磁共振成像》2024年第7期51-57,共7页Chinese Journal of Magnetic Resonance Imaging

基　　金：国家自然科学基金项目(编号:62201265、81701346);江苏省自然科学基金项目(编号:BK20190736)。

摘　　要：目的探究抑郁症(major depression disorder,MDD)患者大脑默认网络(default mode network,DMN)功能连接(functional connectivity,FC)能否预测其睡眠障碍因子分。材料与方法基于REST-meta-MDD公开数据集中满足本实验需求的326例MDD被试静息态功能磁共振成像数据。采用Power模板在全脑中定义了264个脑区节点,分别获取患者的DMN内部FC和DMN与其他网络间的外部FC。采用基于连接组的预测模型在发现数据集上分别基于DMN内部和DMN外部FC对MDD患者的睡眠障碍因子分进行回归预测,独立验证集上检验模型的稳定性。结果在DMN内部FC,发现数据集对MDD患者的睡眠障碍因子分具有一定的预测性(r=0.244,P<0.001),外部独立验证集也有很好的泛化预测效果(r=0.345,P=0.046)。DMN外部FC在发现数据集上对其可进行预测(r=0.238,P<0.001),而独立验证集其泛化性能不足(r=0.256,P=0.143)。结论DMN内部FC对MDD患者睡眠障碍因子分具有一定的预测性。Objective:To explore whether the functional connectivity(FC)of the default mode network(DMN)can predict the sleep disturbance scores of the patients with major depressive disorder(MDD).Materials and Methods:The resting functional magnetic resonance imaging data of 326 patients with MDD from the REST-meta-MDD project were included after undergoing rigorous selection based on the experimental criteria.The entire brain was defined into 256 regions based on the Power template,followed by separate extraction of the FC of the intra-and inter-DMN.Connectome-based predictive modeling was employed to regress individual sleep disturbance score using both types of FC feature,and the experimental findings would be subsequently validated on an external independent validation dataset.Results:The predictive model based on the intra-FC of the DMN demonstrated significant prediction capability for sleep disturbance scores in individuals with depression,not only in the discovery dataset(r=0.244,P<0.001),but also in the external validation dataset(r=0.345,P=0.046).However,models based on the inter-FC of the DMN exhibited limited prediction ability and can only predict the scores in the discovery dataset(r=0.238,P<0.001),failing to generalize to the external validation dataset(r=0.256,P=0.143).Conclusions:The intra-FC of DMN demonstrates predictive capability for the sleep disturbance scores in patients with MDD in some extent.

关 键 词：抑郁症 睡眠障碍 默认网络 静息态功能磁共振成像 磁共振成像 

分 类 号：R445.2[医药卫生—影像医学与核医学] R749.1[医药卫生—诊断学]