Depression Detection and Diagnosis Based on Electroencephalogram (EEG) Analysis: A Systematic Review

Area of Science:

• Neuroscience
• Computer Science
• Psychiatry

Background:

• Mental disorders, including depression, significantly impact global public health and suicide rates.
• Electroencephalogram (EEG) data offers insights into brain function, aiding in the diagnosis of mild depression disorder (MDD).

Purpose of the Study:

• To survey artificial intelligence (AI)-driven approaches for depression diagnosis using Electroencephalogram (EEG) data.
• To analyze methods integrating EEG with machine learning (ML) and deep learning (DL) for identifying depression biomarkers.

Main Methods:

• Systematic analysis of studies employing EEG signals with ML and DL techniques.
• Review of EEG preprocessing, feature extraction, and model development for depression detection.
• Comparison of ML and DL approaches and analysis of existing depression diagnosis datasets.

Main Results:

• AI-driven EEG analysis enhances diagnostic precision, scalability, and automation for depression detection.
• Identified limitations in current datasets and proposed data augmentation and channel selection for improved accuracy.
• Explored future directions including transfer learning, IoT integration for continuous monitoring, and distinguishing depression subtypes.

Conclusions:

• This review serves as a reference for researchers in EEG-based depression detection.
• Provides insights into future research directions for advancing AI-driven mental health diagnostics.