A systematic review of EEG based automated schizophrenia classification through machine learning and deep learning

Area of Science:

• Neuroscience
• Computer Science
• Psychiatry

Background:

• Electroencephalogram (EEG) is crucial for understanding brain activity, especially in mental health research.
• Schizophrenia (SCZ) classification remains a challenge, necessitating advanced diagnostic tools.
• Artificial intelligence (AI) offers novel approaches for analyzing complex neurological data.

Approach:

• This review synthesizes literature on machine learning (ML) and deep learning (DL) applications for SCZ detection via EEG.
• It examines various ML models (e.g., Support Vector Machines, Decision Trees) and DL architectures (e.g., CNNs, LSTMs).
• Challenges and methodologies in AI-driven EEG analysis for SCZ are critically evaluated.

Key Points:

• ML models offer interpretability and efficiency with limited data.
• DL techniques excel at capturing intricate EEG patterns but demand substantial data and computational resources.
• Data quality, interpretability, and ethical considerations are significant hurdles in AI-based SCZ classification.

Conclusions:

• AI, particularly ML and DL, holds significant potential for enhancing schizophrenia diagnosis through EEG analysis.
• Integrating diverse AI techniques can improve classification accuracy and diagnostic reliability.
• Collaborative and ethically sound approaches are essential for the responsible development of automated SCZ classification systems.