From Nano-Enabled Multimodal Biosensing to Health Digital Twins: A Scoping Review and Evidence-Gated Roadmap

Rapid advances in nanomaterials, wearable biosensors, multimodal acquisition, and artificial intelligence have enabled increasingly integrated health-monitoring systems, yet their progression toward health digital twins remains unclear. We conducted a protocol-driven scoping review of original studies combining nano-enabled multianalyte or multimodal sensing, AI-supported analysis, and health applications. PubMed/MEDLINE, Scopus, Web of Science Core Collection, and IEEE Xplore were searched using a publication cutoff of 10 July 2026; platform execution was completed on 13 July 2026. Two reviewers independently screened 528 unique records and assessed 20 full-text reports. A 79-item charting form was jointly verified for 12 included studies. Nine studies reported reference-method or matrix-relevant analytical validation, nine included human-sample or on-body evidence, and six acquired longitudinal or continuous data. Under the author-proposed, corpus-specific functional classification, six systems were L0, five L1, and one L2; none of the 12 met the L3 or L4 functional criteria. No included study combined dynamic individual-state assimilation with prospective prediction or simulation, and none reported external-site validation or formal predictive uncertainty quantification. Because eligibility required nano-enablement, multiple analytes or channels, AI integration, and selected clinical domains, these findings do not estimate the prevalence or maturity of health digital twins in the wider literature. Progress requires longitudinal multimodal data, validated state updating, external generalization, confidence-aware AI, and prospective evaluation of governed feedback.