A Comprehensive Review of Automatic Schizophrenia Identification Based on EEG Signals Using Dynamic Functional Connectivity Analysis and Deep Stack-Augmented Conditional Variational Autoencoder

Abstract

Schizophrenia is a severe neuropsychiatric disorder characterized by disturbances in cognition, perception, and behavior, necessitating early and accurate diagnosis for effective treatment. Electroencephalography (EEG) has emerged as a promising non-invasive modality for identifying neural abnormalities associated with schizophrenia. In recent years, advanced computational techniques, particularly deep learning and functional connectivity analysis, have significantly enhanced automated diagnostic systems. This paper presents a comprehensive review of automatic schizophrenia identification using EEG signals, focusing on dynamic functional connectivity analysis and deep stack-augmented conditional variational autoencoder (DSA-CVAE) architectures. Dynamic functional connectivity captures temporal variations in brain network interactions, providing deeper insights into neural dysfunctions compared to static approaches. Meanwhile, DSA-CVAE models enable robust feature learning, data augmentation, and improved classification performance through probabilistic latent representations. This review systematically examines recent methodologies, datasets, feature extraction techniques, and classification frameworks, highlighting strengths and limitations. Furthermore, it discusses challenges such as data variability, model interpretability, and clinical applicability. The integration of dynamic connectivity with advanced generative deep learning models demonstrates significant potential for improving diagnostic accuracy and generalization. The study concludes by outlining future research directions aimed at developing reliable, scalable, and clinically deployable EEG-based schizophrenia detection systems.