Integrating artificial intelligence (AI) into cardiovascular clinical trials is emerging as a key factor in streamlining patient selection, data collection, endpoint monitoring, and outcome analysis. On the one hand, machine learning and deep learning algorithms facilitate the management and review of ever-increasing volumes of clinical, imaging, and telemonitoring data, identifying predictive patterns and automating repetitive tasks. On the other hand, the high cost and long duration of traditional trials, coupled with the need for adequate population diversity, underscore the urgency of re-engineering trial design. AI can contribute to more adaptive study protocols, minimize interobserver variability, and improve endpoint accuracy. However, technical and ethical challenges remain, including algorithmic bias, privacy, model interpretability, and legal accountability for errors. Looking ahead, the introduction of digital biomarkers, synthetic control arms, and increasingly decentralized trials may redefine experimental paradigms and make cardiovascular trials faster, more inclusive, and more targeted. The aim of this review is to describe the use of AI in randomized controlled trials in cardiology.
Basile et al. (Fri,) studied this question.