Paper Details

Name: Anu Arora, Nipun Chhabra
Country: India

Year: 2026
Volume: Volume No: 13, January, Year: 2026 (Special Issue)
Page Number: 946-957
DOI: https://doi.org/10.5281/zenodo.20624156

Schizophrenia (SCZ), a chronic and crippling psychiatric disorder, is marked by the onset of psychotic episodes, typically in late adolescence or early adulthood. Prior to these episodes, individuals often experience subtle changes in cognition, behavior, and emotion, collectively known as the pre-episodic or prodromal stage. By intervening promptly, early identification of psychosis during this stage may enhance results. This paper explores the development of personalized predictive models using machine learning to forecast the onset of psychosis in individuals at high risk. By combining neuro-imaging biomarkers, clinical data, and behavioral metrics, we aim to enhance predictive accuracy and personalize risk assessments for schizophrenia. We propose a framework that integrates multiple data sources, leveraging advanced ML algorithms to identify individuals most at risk of transitioning to psychosis. Through this approach, we seek to provide the foundation for more proactive and individualized therapeutic strategies.