AI-Enabled Patient Profiling for Personalized Nursing Care and Health Support: Scoping Review

AI-enabled patient profiling has been proposed to organize heterogeneous patient data and support personalized nursing care and health support. Given the growing use of AI, machine learning, large language models (LLMs), digital phenotyping, and digital twins in health care, it is important to clarify how patient profiling has been conceptualized, applied, and evaluated in nursing-relevant contexts.

This scoping review aimed to examine and report the available evidence on AI-enabled patient profiling for personalized nursing care and health support.

This scoping review was conducted using the Joanna Briggs Institute methodological framework and the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews). Chinese and English databases were searched from inception to May 18, 2026, including China National Knowledge Infrastructure, Wanfang Data, VIP Database, Chinese Biomedical Literature Database, PubMed, CINAHL, Web of Science, and Embase. Eligible studies used AI or computational intelligence methods to construct, classify, apply, validate, or evaluate patient profiles or related patient representations relevant to personalized nursing care or health support. Two reviewers independently screened records, assessed full texts, and charted data using a predefined form. Data were synthesized descriptively and narratively using an input-process-output framework.

A total of 61 studies were included. AI-enabled patient profiling was represented through patient profiles, patient or user personas, digital phenotypes, patient segments, clinical subtypes, digital twins, and LLM-enabled personalization. Included studies used diverse data sources, including electronic health records, nursing records, patient-reported outcomes, questionnaires, mobile app data, wearable and sensor data, engagement logs, and textual or conversational data. Common AI approaches included machine learning, clustering, deep learning, natural language processing, recommender systems, knowledge graphs, digital twin modeling, and LLMs. Applications focused on personalized education, self-management support, rehabilitation, chronic disease management, remote monitoring, decision support, follow-up care, and shared decision-making. Most studies evaluated algorithmic performance, feasibility, usability, or system development, whereas fewer studies examined profile validity, actionability, equity, workflow integration, or long-term nursing-sensitive and patient-centered outcomes.

Existing evidence positions AI-enabled patient profiling as an emerging data-to-action mechanism for personalized nursing care and health support rather than as a mature clinical intervention. Future research should strengthen conceptual clarity, validate the profile, establish explicit pathways linking profiles to actions, incorporate nurse and patient involvement, assess equity and safety, and evaluate implementation in real-world nursing settings.

OSF Registry mey4u; https://osf.io/mey4u.