Artificial intelligence-driven predictive analytics for early detection and risk stratification of cardiovascular disease

Abstract

Cardiovascular disease (CVD) remains the leading cause of morbidity and mortality worldwide, necessitating improved approaches for early detection and risk stratification. Traditional cardiovascular risk assessment models often demonstrate limited predictive accuracy because they rely on a restricted number of clinical variables and linear statistical relationships. Artificial intelligence (AI)-driven predictive analytics has emerged as a promising solution capable of processing large, complex, and multidimensional healthcare datasets to improve cardiovascular risk prediction and clinical decision-making. This systematic review aimed to evaluate the effectiveness of AI-based predictive models for the early detection and risk stratification of CVD. A comprehensive literature search was conducted across PubMed/MEDLINE, Scopus, Web of Science, Embase, CINAHL, IEEE Xplore, and the Cochrane Library for studies published between January 2015 and December 2025. A total of 130 studies involving approximately 12.8 million participants met the eligibility criteria and were included in the review. The findings demonstrated that machine learning (ML) and deep learning (DL) models consistently outperformed conventional cardiovascular risk prediction tools, achieving area under the receiver operating characteristic curve (AUC) values ranging from 0.80-0.99 across various cardiovascular conditions. The strongest evidence was observed in the prediction of coronary artery disease, heart failure, atrial fibrillation, stroke, and major adverse cardiovascular events. Electronic health records, electrocardiography, medical imaging, wearable devices, and multimodal datasets were the most commonly utilized data sources. Despite their promising performance, challenges related to external validation, interpretability, algorithmic bias, and clinical implementation remain.