Big data in psychiatry: current applications in diagnosis, prevention and personalized treatments

Abstract

With the digitalization of healthcare systems, psychiatry has begun to integrate data science concepts to better understand the mechanisms of disorders. The complex analysis of biological, behavioral, and clinical data allows for a more complete clinical picture and improved prediction of disease progression. The aim of this paper is to explore how Big Data is used in psychiatry to improve early diagnosis, predict relapses, and personalize treatment. A systematic review of 24 publications from 2015 to 2024 was conducted using databases such as PubMed and Nature. The included studies used clinical data, brain imaging, digital monitoring, genetic information, and artificial intelligence applied in psychiatric assessment, intervention, relapse prediction, and treatment monitoring. Large-scale resources (ENIGMA, PGC, UK Biobank) and routinely collected digital footprints enabled population-level analyses in psychiatry. Methodological reviews emphasize that careful feature engineering, data harmonization, and external validation are crucial; naïve application of complex models risks overly optimistic estimates. Big Data has helped clarify the architecture of disorders and enabled scalable measurement, but its clinical utility depends on multimodal integration, robust external validation, and precise calibration. The study confirms that the integration of multidimensional data (clinical, digital, and biological) can enhance the accuracy of relapse prediction and treatment response in psychiatry. Big Data is becoming a valuable tool in the development of personalized psychiatry and preventive interventions.