AI for Pre-Symptomatic Schizophrenia: Redefining Early Mental Health Diagnostics

Schizophrenia spectrum disorders (SSD) progress through distinct phases, with early detection in pre-acute stages (pre-morbid and prodromal) being critical for timely intervention and improved clinical outcomes. Traditional diagnostic approaches, such as the Structured Interview for Prodromal Syndromes (SIPS) and the Scale of Prodromal Symptoms (SOPS), often struggle with subjectivity, scalability, and the subtle nature of early symptoms, leading to significant diagnostic delays. To address these limitations, this study proposes an AI-driven framework that leverages multimodal data, including behavioral observations, structured clinical interviews, psychometric scales, and neurophysiological markers (e.g., EEG). Our approach integrates these diverse data sources using machine learning models to identify complex cognitive, emotional, and social deviations characteristic of SSD's pre-acute phases. We validate our framework using a synthetic dataset simulating early-stage symptoms, demonstrating superior sensitivity and specificity compared to traditional methods. The system also supports longitudinal tracking, enabling continuous risk assessment and early intervention. Our findings underscore the potential of AI to transform early SSD detection, reducing diagnostic latency and improving patient outcomes.