Predictability of depression severity based on posterior alpha oscillations.

Area of Science:

• Neuroscience
• Computational Psychiatry
• Biomedical Engineering

Background:

• Major Depressive Disorder (MDD) diagnosis relies on subjective clinical assessments.
• Objective biomarkers for MDD severity are needed for effective treatment monitoring.
• Magnetoencephalography (MEG) offers high temporal and spatial resolution for brain activity analysis.

Purpose of the Study:

• To integrate neural data from MEG with advanced machine learning to predict individual MDD patient severity.
• To identify specific patterns of brain activity associated with depression severity.
• To develop a quantitative and objective tool for MDD severity assessment.

Main Methods:

• Acquired resting-state MEG data from 22 MDD patients and 22 healthy controls (HC).
• Calculated individual power spectra using Fourier transform and reconstructed sources via beamforming.
• Applied Bayesian linear regression to predict depression severity based on the spatial distribution of oscillatory power.

Main Results:

• MDD patients exhibited decreased theta and alpha power in specific brain regions, and increased beta power.
• Posterior alpha power showed a significant negative correlation with depression severity.
• The machine learning model accurately predicted depression severity using alpha and beta power distributions (r=0.68 and r=0.56, respectively).

Conclusions:

• MEG data reveals distinct alterations in oscillatory brain activity in MDD patients.
• The developed model provides a quantitative and objective estimation of depression severity.
• This approach holds potential for aiding MDD diagnosis and monitoring recovery progress.