EUVIMEDEUVIMED entdecken
Crossref

Artificial Intelligence in Precision Cardiovascular Medicine: Risk Prediction, Digital Twins, and Explainability

Integrative Biomedical Research · 2026

DOI: 10.25163/biomedical.10110915 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Integrative Biomedical Research, 2026.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Cardiovascular disease remains the leading cause of global mortality despite decades of guideline-driven prevention, and population-averaged risk calculators such as the Framingham Risk Score, QRISK3, SCORE2, and the Pooled Cohort Equations were not designed to capture individual patient variability. This narrative review examines whether artificial intelligence (AI) and machine learning (ML) can address this limitation and identifies barriers to clinical implementation. A structured narrative synthesis of peer-reviewed literature (2008–2026) was conducted via manual reference-list screening, with eligible sources covering AI/ML-based cardiovascular risk prediction, digital twin architectures, explainable AI, and cardiac rehabilitation trials. Evidence was organized across four thematic domains and summarized in four evidence tables spanning 12 randomized controlled rehabilitation trials, 11 hypertension-focused ML/deep-learning models, 10 general cardiovascular AI models, and one autonomic-phenotyping cohort study. Multimodal AI architectures consistently outperformed conventional statistical models, achieving AUCs of 0.98–0.99 in select cohorts; a random-forest model combining clinical variables with photoplethysmography-derived heart-rate-variability metrics reached an AUC of 0.9988 for distinguishing older adults with cardiovascular disease from at-risk controls. Digitally delivered and hybrid cardiac rehabilitation matched or exceeded center-based outcomes while improving access for underserved groups. Still, fewer than 15% of validated predictive models reach routine clinical use, driven by algorithmic bias, thin prospective evidence, black-box opacity, and workflow friction. Personalized, AI-enabled risk prediction is technically mature but organizationally immature, requiring harmonized datasets, prospective multicenter trials, explainability tooling, and clinician-supervised frameworks like the Team-Implementation Multidisciplinary Approach.

Zitieren

APA

(2026). Artificial Intelligence in Precision Cardiovascular Medicine: Risk Prediction, Digital Twins, and Explainability. Integrative Biomedical Research. https://doi.org/10.25163/biomedical.10110915

Vancouver

Artificial Intelligence in Precision Cardiovascular Medicine: Risk Prediction, Digital Twins, and Explainability. Integrative Biomedical Research. 2026. doi:10.25163/biomedical.10110915

BibTeX
@article{euvimed2026,
  title = {Artificial Intelligence in Precision Cardiovascular Medicine: Risk Prediction, Digital Twins, and Explainability},
  journal = {Integrative Biomedical Research},
  year = {2026},
  doi = {10.25163/biomedical.10110915}
}
Kostenlos starten →