Clinical applications of digital twin technology for predicting treatment response and personalizing cardiovascular interventions: a systematic review

Background and Objective: Cardiovascular digital twins (DTs) integrate multiscale data for dynamic, patient-specific in silico simulation. We systematically evaluated their diagnostic accuracy, predictive utility, and clinical application in guiding cardiovascular interventions. Methods: Following PRISMA ۲۰۲۰ guidelines, PubMed, Scopus, Web of Science, and Google Scholar were searched for human studies utilizing multidimensional DTs for pre-interventional simulation or response prediction. Methodological quality was appraised using PROBAST and QUADAS-۲. Results: Nine eligible studies were synthesized across coronary artery disease (n=۳), electrophysiology (n=۲), heart failure (n=۲), cardiac resynchronization therapy (n=۱), and structural heart disease (n=۱). DT frameworks showed high fidelity: coronary models achieved ۸۴.۳%–۹۲.۶% diagnostic accuracy for ischemia (AUC up to ۰.۹۶), while biophysical and machine learning models correctly localized >۸۰% of ventricular tachycardia reentrant isthmuses and outperformed conventional calculators in predicting post-intervention remodeling. However, most predictive frameworks exhibited high or unclear risk of bias due to small developmental sample sizes and limited prospective validation. Conclusion: Cardiovascular DTs offer high predictive fidelity for personalized intervention planning; however, clinical translation requires standardized unc

کلیدواژه ها:

Digital twin ، cardiovascular diseases ، Personalized medicine ، Treatment simulation ، Coronary intervention ، Artificial intelligence