Integrating Digital Twins and Predictive Analytics for Personalized Patient Monitoring and Care Optimization

Abstract

The utilisation of digital twins in conjunction with predictive analytics has seen a marked rise in its promotion for applications in personalised patient monitoring, treatment planning and service optimisation. Nevertheless, the extent to which these claims are substantiated by real-world healthcare evidence remains uncertain. To map empirical applications of these technologies and summarise their impacts and implementation constraints, a scoping review was conducted using a Population–Concept–Context framework. The review searched MEDLINE/ PubMed and Web of Science for peer-reviewed studies published in English between 2015 and 2025. Nineteen studies met the inclusion criteria, with most published after 2022. Most were clinical twins spanning cardiovascular, hepatology, diabetes, oncology, and intensive care; one addressed emergency communication operations. Imaging-anchored and hybrid physiological twins achieved the strongest patient-specific validation, while continuous-data twins improved glycaemic control, risk prediction, and therapy optimisation. Acute-care and operational twins showed feasibility and safety gains, but inconsistent effects on efficiency. Barriers clustered around fragmented data standards, missing or noisy inputs, workflow fit, clinician trust, and limited multicentre prospective trials. Predictive digital twins are transitioning from proof-of-concept to early utility. However, scalable impact will depend on interoperable data ecosystems, rigorous staged validation, and human-centred deployment with governance for safety and equity.