Page 58 - 磁共振成像2024年7期电子刊

磁共振成像 2024年7月第15卷第7期 Chin J Magn Reson Imaging, Jul, 2024, Vol. 15, No. 7 临床研究｜｜Clinical Articles

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

1, 2
秦姣龙 1, 2* ，李弘瑄 ，吴烨 ，倪黄晶 3*
1, 2
作者单位 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
* 通信作者 倪黄晶，E-mail: nihuangjing@njupt.edu.cn；秦姣龙，E-mail: jiaolongq@njust.edu.cn
中图分类号 R445.2；R749.1 文献标识码 A DOI 10.12015/issn.1674-8034.2024.07.009
本文引用格式 秦姣龙, 李弘瑄, 吴烨, 等 . 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[J]. 磁共振成像, 2024,
15(7): 51-57.

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

1, 2
1, 2
QIN Jiaolong 1, 2* , LI Hongxuan , WU Ye , NI Huangjing 3*
1 Key Lab of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education, School of Computer
2
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
3
and Telecommunications, Nanjing 210023, China
* Correspondence to NI H J, E-mail: nihuangjing@njupt.edu.cn; QIN J L, E-mail: jiaolongq@njust.edu.cn
Received 19 Jan 2024, Accepted 6 Jun 2024; DOI 10.12015/issn.1674-8034.2024.07.009
ACKNOWLEDGMENTS National Natural Science Foundation of China (No. 62201265, 81701346); Natural Science of Jiangsu
Province (No. BK20190736).
Cite this article as QIN J L, LI H X, WU Y, et al. Functional connectivity within the default mode network can predict the sleep
disturbance scores of the patients with depression[J]. Chin J Magn Reson Imaging, 2024, 15(7): 51-57.
Abstract 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.
Key words depression; sleep disturbance; default mode network; resting-state functional magnetic resonance imaging; magnetic
resonance imaging

0 引言 首次发现 MDD 患者有着更短的快速眼动睡眠潜伏
睡 眠 障 碍 是 抑 郁 症（major depressive disorder, 期。睡眠障碍与 MDD 的病因和发病机制密切相
MDD）最常见的临床核心症状之一 。KUPFER 等 [2] 关 ，也是 MDD 患者疾病复发或自杀的风险因素 [4-5] 。
[1]
[3]
收稿日期 2024-01-19 接受日期 2024-06-06
基金项目 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）

https://www.chinesemri.com ·51 ·