DefinePK

Abstract

Background: Myocardial infarction (MI) remains a leading global cause of morbidity and mortality, with early detection being critical for improving clinical outcomes. Conventional diagnostic methods often rely on patient presentation to healthcare settings, leading to delayed intervention. Wearable devices integrated with artificial intelligence (AI) offer the potential for real-time, non-invasive, and accessible MI detection. However, there is limited consolidated evidence evaluating the diagnostic accuracy and clinical utility of these technologies.
Objective: This systematic review aims to evaluate the effectiveness and diagnostic performance of AI-driven wearable devices in the early detection of myocardial infarction.
Methods: A systematic review was conducted in accordance with PRISMA guidelines. Literature was searched across PubMed, Scopus, Web of Science, and IEEE Xplore for studies published between 2019 and 2024. Included studies evaluated adult human subjects using AI-integrated wearable or portable ECG devices for MI detection. Study designs encompassed randomized trials, cohort studies, and validation models. Risk of bias was assessed using the Cochrane Risk of Bias Tool and the Newcastle-Ottawa Scale. Data were synthesized narratively due to heterogeneity in study design and outcomes.
Results: Eight studies were included in the final review. AI algorithms demonstrated high diagnostic accuracy (sensitivity and specificity >90%) across multiple device platforms. Notably, one study reported an AUC of 0.9954 and another achieved an F-score of 88.10%. AI models were successfully integrated into real-time or embedded systems, and performance was comparable or superior to clinician-based interpretations. Study quality was moderate to high, although variations in design and small sample sizes were noted.
Conclusion: AI-enhanced wearable technologies show strong promise for the early detection of myocardial infarction, offering significant clinical benefits in timely diagnosis and intervention. While the current evidence supports their diagnostic value, further large-scale prospective trials are needed to validate performance and guide clinical implementation.