Advances in Electroencephalography for Post-Traumatic Stress Disorder Identification: A Scoping Review.

Area of Science:

• Neuroscience
• Computational Psychiatry
• Biomedical Engineering

Background:

• Post-traumatic stress disorder (PTSD) is a debilitating condition often diagnosed using subjective methods.
• Objective diagnostic and therapeutic tools for PTSD are critically needed.

Purpose of the Study:

• To conduct a scoping review of computational methods utilizing electroencephalography (EEG) signal processing for PTSD diagnosis, differentiation, and therapy.
• To provide a comprehensive overview of the EEG analysis pipeline for PTSD research.

Main Methods:

• A systematic literature search adhering to the PRISMA-ScR protocol was conducted across Scopus, Web of Science, and PubMed (2013-2024).
• Analysis included 73 studies focusing on EEG-based PTSD diagnosis (52), differentiation (8), and therapy (15).
• Dominant techniques analyzed were EEG bands and Event-Related Potentials (ERPs), alongside various statistical and machine learning models.

Main Results:

• The Alpha band and LPP ERP showed significant utility in PTSD diagnosis and therapy.
• Supervised Support Vector Machine (SVM) models demonstrated high accuracy (e.g., 0.997 for diagnosis).
• Multimodal Random Forest models integrating EEG with other biosignals achieved high accuracy (e.g., 0.993).

Conclusions:

• EEG-based computational methods offer a promising avenue for objective PTSD assessment and treatment monitoring.
• Limitations include underutilization of ERPs, sleep data, and full-band EEG analysis, alongside a lack of diverse datasets.
• Future research should explore deep learning, adaptive signal decomposition, and multimodal integration for enhanced PTSD research.