Abstract
Psychiatric diagnosis continues to be grounded predominantly in symptom-based classificatory systems, most notably the Diagnostic and Statistical Manual of Mental Disorders and the International Classification of Diseases. Although these frameworks provide essential standardization and facilitate diagnostic reliability in clinical practice, they remain largely descriptive in nature and do not correspond directly to well-characterized neurobiological mechanisms. This limitation has stimulated increasing scientific interest in the identification of objective biomarkers that may enhance diagnostic precision, improve prognostic assessment, and support treatment stratification in psychiatric disorders. The present review synthesizes recent advances in the application of functional magnetic resonance imaging and electroencephalography for the purpose of psychiatric disease diagnosis. Recent progress in computational psychiatry, together with developments in machine learning methodologies and the growing availability of large-scale neuroimaging datasets, has accelerated efforts to translate these approaches into clinically relevant diagnostic instruments. In particular, studies employing functional connectivity analyses have revealed reproducible patterns of large-scale network dysregulation across several psychiatric conditions, including major depressive disorder, schizophrenia, attention-deficit hyperactivity disorder, and anxiety disorders. Complementary electrophysiological research utilizing electroencephalography has likewise demonstrated alterations in neural oscillatory activity, event-related potentials, and microstate organization, collectively suggesting disruptions in the temporal coordination of neural dynamics in psychiatric populations. Despite these advances, substantial methodological and translational challenges remain prior to the routine clinical implementation of these techniques. A considerable proportion of reported findings are derived from group-level analyses and exhibit limited generalizability at the level of individual patients. Moreover, methodological heterogeneity across studies, relatively small sample sizes, and the substantial diagnostic overlap that characterizes many psychiatric disorders continue to constrain the reliable clinical translation of neuroimaging-derived biomarkers.