Hiring in life sciences? Share your open positions with our professional community. Read more Close

Advertisement

Digital Twin-Guided Personalized Nutrition for Prediabetes in Primary Care: A Simulation-Based Technical Feasibility Study and Pragmatic Trial Protocol.

Created on 21 Nov 2026

Authors

Wulfran Fendzi Mbasso, Ambe Harrison, Zhe Liu, Raman Kumar, V Senthil Murugan, Zokir Mamadiyarov

Published in

Journal of primary care & community health. Volume 17. Pages 21501319261508571. Epub Nov 20, 2026.

Abstract

Background/ObjectivesPersonalized nutrition may support a healthy lifestyle and improve prediabetes management, but most digital tools remain open loop. We developed a clinician-governed digital-twin framework for adaptive nutrition decision support and a prospective clinical evaluation pathway.MethodsThe framework integrates continuous glucose monitoring (CGM), heart-rate variability (HRV), activity, sleep, and dietary data within an individualized state-space model. Technical/computational feasibility was assessed using one deterministic 42-day reference simulation and six additional prespecified 42-day phenotype/failure stress scenarios (seven scenarios in total). Synthetic generation rules, random seeds, estimator settings, action constraints, and executable reference code are specified for reproduction. A future 24-week, three-arm multicentre pragmatic randomized trial with an embedded four-week micro-randomized component is specified using SPIRIT 2025 and SPIRIT-AI principles. No human participant was enrolled in the simulation component.ResultsIn the 42-day reference run, 126 nominal decision opportunities were generated and 97 (77.0%) met the prespecified availability rule. Thirty-five adaptive recommendations were delivered, 52 available decisions resulted in no adaptive change, 29 decisions were withheld for insufficient inputs, and 10 candidate actions were blocked by hard constraints and routed to fallback; no hard-constraint violation was delivered. CGM missingness was 5.65%. A programmed 10-minute CGM lag was recovered as 10 minutes; alignment reduced CGM-to-latent-state RMSE from 3.53 to 3.01 mg/dL, and the UKF/RTS glucose-state RMSE was 1.17 mg/dL (1.26 mg/dL within missing-CGM windows). All seven scenarios (reference plus six additional stress conditions) completed without numerical failure or delivered hard-constraint violations.ConclusionsThe results support end-to-end software-integration feasibility under explicitly programmed synthetic conditions and provide a reproducible basis for prospective validation. They do not establish physiological validity, clinical effectiveness, or safety. The future trial has not commenced or enrolled participants; prospective registration, ethics approval, final statistical review, and informed consent will precede enrollment.

PMID:
43064522
Bibliographic data and abstract were imported from PubMed on 21 Nov 2026.

Read full publication at:
Please sign in to see all details.

Advertisement

Stats

  • Community rating n/a 0 votes
  • Reviewers' rating n/a 0 votes
  • Your rating

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

  • Recommendations n/a n/a positive of 0 vote(s)
  • Views 6
  • Comments 0

Recommended by

  • No recommendations yet.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.

Advertisement