A Predictive Biomarker Model Using Quantitative Electroencephalography in Adolescent Major Depressive Disorder

Area of Science:

• Neuroscience
• Psychiatry
• Biomarker Research

Background:

• Growing need for objective biomarkers in child and adolescent psychiatric disorders.
• Major Depressive Disorder (MDD) in adolescents lacks precise diagnostic tools.
• Quantitative Electroencephalography (qEEG) shows potential for neurological assessments.

Purpose of the Study:

• To develop a predictive model for adolescent Major Depressive Disorder (MDD) using only qEEG data.
• To differentiate adolescents with MDD from healthy controls (HCs) using qEEG biomarkers.
• To evaluate the efficacy of a singular logistic regression model based on qEEG.

Main Methods:

• Collected qEEG data (alpha, beta, theta, delta bands) and psychometric measures from adolescents (14-17 years) with MDD (n=35) and HCs (n=14).
• Analyzed qEEG data for coherence, cross-correlation, and power.
• Developed a two-stage logistic regression model, validated using receiver-operating characteristic (ROC) curve analysis.

Main Results:

• Identified four key qEEG predictors: F3-C3 alpha coherence, P3-O1 theta coherence, CZ-PZ beta coherence, and P8-O2 theta power.
• The final logistic regression model achieved a robust ROC area of 0.8226.
• Confirmed previous findings of qEEG differences between adolescents with MDD and HCs.

Conclusions:

• A single-value predictive model for adolescent MDD was successfully developed using qEEG.
• The model demonstrates qEEG's potential as a low-cost, effective intermediate biomarker for youth MDD.
• Observed differences implicate brain areas related to behavioral disinhibition and default mode networks.