1.Research Questions:
(A)How can multi-modal, longitudinal, and cross-organ data be integrated to accurately model individual cardiac structure and function using cardiac digital twins?
(B)What genetic, biochemical, lifestyle, and environmental factors contribute to early, subclinical cardiac changes?
(C)Can personalised cardiac digital twins models offer new features to aid predictions for disease onset and progression?
2.Objectives:
(A)Develop high-precision cardiac digital twin models by combining UK Biobank imaging, biochemical, genomic/proteomic, and lifestyle/environmental data.
(B)Identify early biomarkers and phenotypic signatures associated with increased cardiovascular risk.
(C)Model dynamic interactions between cardiac and systemic health indicators to predict individual disease trajectories.
(D)Evaluate the clinical utility of simulation-based predictions for prevention and intervention strategies.
3.Scientific Rationale:
Cardiovascular disease often progresses silently until irreversible damage occurs. AI-driven digital twin technology offers a novel means to detect early changes, forecast progression, and personalise interventions. The UK Biobank’s large, standardised, multi-modal dataset-spanning imaging, omics, biochemical, lifestyle, and follow-up health outcomes-provides an unparalleled foundation for robust model development and validation. Leveraging these resources will enable building a digital twinning framework that is reproducible, generalisable, and thus offers clinically relevant findings, contributing to more effective strategies for reducing the global CVD burden.
