Digital twins in dementia: Early evidence and future directions for precision psychiatry

ABSTRACT Dementia remains a major global health challenge, with limited progress in early diagnosis and therapy. Emerging digital twin technologies—virtual computational replicas of patients or biological systems—offer new possibilities for precision medicine. This brief review summarizes current evidence on digital twin applications in dementia across diagnosis, care, drug discovery, and computational modelling. Twenty relevant studies published between 2015 and 2025 were identified, including empirical, conceptual, and in-silico works. The initial database search across PubMed, Scopus, Web of Science, and IEEE Xplore yielded 1,302 records, of which 20 met the inclusion criteria following PRISMA-based screening. Findings suggest that digital twins hold promise in drug discovery, where in-silico simulations accelerate compound screening and repurposing. Diagnostic models using EEG and neuroimaging twins demonstrate potential for early detection but lack clinical validation. Applications in care and rehabilitation remain conceptual, focusing on therapy planning and immersive cognitive interventions. Foundational computational studies lay theoretical groundwork for simulating neurodegenerative processes. Overall, digital twins represent an evolving frontier in dementia research, with significant opportunities, but limited clinical translation. Establishing longitudinal validation, integrating multimodal data, and adopting ethical frameworks will be critical for realizing their precision medicine potential. Ethical and governance discussions remain scarce, particularly around patient autonomy and data bias, underscoring the need for dementia-specific digital health ethics frameworks.