The role of artificial intelligence in managing hospitalized patients with mental illness: a scoping review

Abstract

Abstract Background The use of artificial intelligence (AI) in psychiatric care has the potential to revolutionize the treatment and management of hospitalized patients with mental health conditions. AI tools, such as machine learning, deep learning, and natural language processing, are being increasingly studied for their capacity to improve diagnostic precision, treatment strategies, and overall patient outcomes. Objective This scoping review aimed to investigate the role of artificial intelligence (AI) in managing hospitalized patients with mental illness. Methods Following Arksey and O’Malley’s framework, a comprehensive search of databases—including PubMed, PsycINFO, Scopus, IEEE Xplore, and the Cochrane Library—was conducted for studies published between January 15, 2015, and March 15, 2025. Data were extracted systematically, and findings synthesized narratively to identify major themes. Results Twenty-four studies were included, highlighting the various ways AI has been applied in psychiatric inpatient care. Machine learning (ML) and natural language processing (NLP) technologies were reported to enhance diagnostic processes by analyzing extensive datasets from electronic health records and clinical documentation, potentially supporting earlier detection and intervention. Predictive analytics effectively assessed patient risks, enhancing proactive care strategies. AI-supported decision systems, digital phenotyping, and chatbots facilitated tailored therapeutic interventions and continuous patient monitoring. Furthermore, AI improved hospital operations through efficient bed management, patient flow optimization, and staff scheduling. Despite these advancements, challenges were noted, such as algorithmic biases due to non-diverse training data, privacy and security risks for sensitive patient information, and clinician resistance stemming from concerns about autonomy and ethical implications of AI-supported clinical decisions. Conclusion AI demonstrates promising potential for enhancing psychiatric inpatient care through improved diagnostics, personalized treatments, and operational efficiencies. However, successful integration requires addressing ethical and practical challenges, promoting data diversity, safeguarding privacy, and ensuring clinician engagement. Future research should continue exploring these aspects to enable responsible and equitable adoption of AI in psychiatric hospitals.

Keywords

• Artificial intelligence
• Psychiatric care
• Mental illness
• Inpatient treatment