Out-of-hospital cardiac arrest (OHCA) remains a major global health challenge with low survival rates. Artificial intelligence (AI) has emerged as a promising tool to enhance early detection, response, and management of OHCA cases. This study explores the current use of AI in OHCA, identifying challenges and opportunities related to its implementation. This scoping review followed the PRISMA-ScR guidelines, utilizing a systematic search of international databases to identify relevant literature. A total of 10 studies were included, comprising cohort studies, observational studies, randomized controlled trials (RCTs), and pilot projects from 10 different countries. AI implementation in OHCA management demonstrated several opportunities, including improved early detection (increasing sensitivity by 5.5–15% and reducing EMS response time by up to 26 seconds), enhanced decision support for termination of resuscitation (with specificity up to 99.0%), and increased bystander engagement through real-time CPR guidance. However, challenges remain, such as data privacy, ethical concerns (especially with visual surveillance and GDPR compliance), infrastructure limitations, and variability in local protocol. The paradox between faster detection and improved CPR quality was also noted. AI has significant potential to improve OHCA outcomes by optimizing detection, response, and clinical decision-making. Successful implementation requires multidisciplinary collaboration, robust external validation, and ethical considerations to address privacy and local adaptation. Integrating AI into emergency systems and public training can enhance survival rates, but further large-scale studies are needed to ensure effectiveness and equity.
Dalle et al. (Fri,) studied this question.