基于多脑区脑电图信号分析的抑郁症检测
Depression Detection Based on Analysis of EEG Signals in Multi Brain Regions.
作者信息
Yang Jianli, Zhang Zhen, Xiong Peng, Liu Xiuling
机构信息
College of Electronic Information and Engineering, Hebei University, 071002 Baoding, Hebei, China.
Key Laboratory of Digital Medical Engineering of Hebei Province, 071002 Baoding, Hebei, China.
出版信息
J Integr Neurosci. 2023 Jul 11;22(4):93. doi: 10.31083/j.jin2204093.
BACKGROUND
As an objective method to detect the neural electrical activity of the brain, electroencephalography (EEG) has been successfully applied to detect major depressive disorder (MDD). However, the performance of the detection algorithm is directly affected by the selection of EEG channels and brain regions.
METHODS
To solve the aforementioned problems, nonlinear feature Lempel-Ziv complexity (LZC) and frequency domain feature power spectral density (PSD) were extracted to analyze the EEG signals. Additionally, effects of different brain regions and region combinations on detecting MDD were studied with eyes closed and opened in a resting state.
RESULTS
The mean LZC of patients with MDD was higher than that of the control group, and the mean PSD of patients with MDD was generally lower than that of the control group. The temporal region is the best brain region for MDD detection with a detection accuracy of 87.4%. The best multi brain regions combination had a detection accuracy of 92.4% and was made up of the frontal, temporal, and central brain regions.
CONCLUSIONS
This paper validates the effectiveness of multiple brain regions in detecting MDD. It provides new ideas for exploring the pathology of MDD and innovative methods of diagnosis and treatment.
背景
脑电图（EEG）作为一种检测大脑神经电活动的客观方法，已成功应用于重度抑郁症（MDD）的检测。然而，检测算法的性能直接受EEG通道和脑区选择的影响。
方法
为解决上述问题，提取非线性特征莱姆尔 - 齐夫复杂度（LZC）和频域特征功率谱密度（PSD）以分析EEG信号。此外，在静息状态下闭眼和睁眼时，研究了不同脑区及区域组合对MDD检测的影响。
结果
MDD患者的平均LZC高于对照组，MDD患者的平均PSD普遍低于对照组。颞区是检测MDD的最佳脑区，检测准确率为87.4%。最佳的多脑区组合由额叶、颞叶和中央脑区组成，检测准确率为92.4%。
结论
本文验证了多个脑区在检测MDD中的有效性。它为探索MDD的病理以及创新诊断和治疗方法提供了新思路。
Zotero 插件