Scholarlink

<p>The use of scalp electroencephalography (EEG) provides a cost-effective utility in identifying features and treatment outcomes for Major Depressive Disorder (MDD). We utilised a publicly available dataset to explore spectral features, complexity, and large-scale dynamics in the eyes closed resting state of MDD compared to controls. Relative band power revealed significantly higher beta power (13–30&#xa0;Hz) in MDD compared to controls, further, a significantly reduced aperiodic exponent was observed. Upon observing multiscale entropy and Higuchi fractal dimension we observed significantly higher values in MDD with both metrics. To explore this further, we observed the temporal dynamics of brain states through microstate analysis and the transmission of information through network topology. We found reduced stability for the temporal components of brain microstates in MDD compared to controls. Further, small worldness index values were significantly lower in MDD through the phase locking value (PLV) as well, indicating a greater deviation towards the topology of random networks. Through rank ordering the features extracted with the area under the receiver operating characteristic curve (ROC), and found relative beta power, HFD, aperiodic exponent, and short-scale entropy to be the best predictors for MDD. From the data presented, it is clear that EEG activity is not only unpredictable at the level of the channel but also in the domain of communication between regions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

EEG Unpredictability in the Resting State of Major Depressive Disorder: A Multidomain EEG Analysis

  • 期刊论文

  • 发表日期：

    2026年3月27日

查看全文
  • Kassra Ghassemkhani,
  • Blake T. Dotta

摘要

The use of scalp electroencephalography (EEG) provides a cost-effective utility in identifying features and treatment outcomes for Major Depressive Disorder (MDD). We utilised a publicly available dataset to explore spectral features, complexity, and large-scale dynamics in the eyes closed resting state of MDD compared to controls. Relative band power revealed significantly higher beta power (13–30 Hz) in MDD compared to controls, further, a significantly reduced aperiodic exponent was observed. Upon observing multiscale entropy and Higuchi fractal dimension we observed significantly higher values in MDD with both metrics. To explore this further, we observed the temporal dynamics of brain states through microstate analysis and the transmission of information through network topology. We found reduced stability for the temporal components of brain microstates in MDD compared to controls. Further, small worldness index values were significantly lower in MDD through the phase locking value (PLV) as well, indicating a greater deviation towards the topology of random networks. Through rank ordering the features extracted with the area under the receiver operating characteristic curve (ROC), and found relative beta power, HFD, aperiodic exponent, and short-scale entropy to be the best predictors for MDD. From the data presented, it is clear that EEG activity is not only unpredictable at the level of the channel but also in the domain of communication between regions.