Prediction model for potential depression using sex and age-reflected quantitative EEG biomarkers

Area of Science:

• Neuroscience
• Psychiatry
• Machine Learning

Background:

• Depression is a widespread mental health issue with significant societal impact.
• Current diagnostic methods like Beck's Depression Inventory (BDI) and Hamilton Depression Rating Scale (HDRS) are subjective and time-consuming.
• There is a need for objective, reliable biomarkers for early depression detection.

Purpose of the Study:

• To develop and validate a machine learning model for early depression screening.
• To investigate the efficacy of quantitative EEG (QEEG) z-score metrics as objective biomarkers for depression.
• To assess the influence of sex and age on QEEG indicators for depression classification.

Main Methods:

• Trained machine learning models using sex and age-adjusted z-score values of quantitative EEG (QEEG) indicators.
• Utilized data from the National Standard Reference Data Center for Korean EEG.
• Included 116 potential depression subjects and 80 healthy controls in the study cohort.

Main Results:

• The classification model successfully distinguished between potential depression and healthy control groups.
• Achieved a high test accuracy of 92.31% in identifying depression.
• Demonstrated a low 10-cross-validation loss of 0.13, indicating model robustness.

Conclusions:

• Z-score QEEG metrics, incorporating sex and age, show promise as objective and reliable biomarkers for early depression screening.
• The developed machine learning model offers a potential advancement over traditional subjective diagnostic methods.
• This approach could facilitate earlier intervention and improve outcomes for individuals with potential depression.