ABSTRACT Background Artificial intelligence (AI) is utilized in various health‐related areas, including medical imaging, personal diagnosis, therapeutic interventions and predictive analytics through electronic health records (EHRs). The incorporation of AI is especially significant in critical care, where it has shown the ability to improve patient outcomes. Aim To analyse the research status, hotspots and development trends of AI in critical care nursing (CCN) over the past decade, providing a reference for clinical nursing practice and future studies. Study Design To explore the research on the application of AI in CCN practice, articles were retrieved from the Web of Science Core Collection (WOSCC) database from 1 January 2015 to 30 April 30 2025. Bibliometric and cluster analyses were performed using CiteSpace 6.3 R1 software, where the parameters included the number of publications annually, country/region, participating institutions, journals, authors, cited references, keywords and other related parameters. Results The research involved 1597 publications, reflecting the growth of publications each year regarding the application of AI in CCN. The top 5 most frequent keywords were ‘cardiac arrest’, ‘big data’, ‘definitions’, ‘models’ and ‘deep learning (DL)’, with DL, models and clinical decision‐making support identified as current research hotspots. Conclusions The application of AI in CCN is promising. The current research is mainly focussed on early disease detection, deep learning and risk prediction in various domains. It is suggested to further enhance the collaboration among disciplines and countries, maximize the scientific collaboration network among institutions and authors and promote the deep integration of AI technologies with CCN. This should be based on the research hotspots and trends to develop to meet the demands of the quality care of critically ill patients. Relevance to Clinical Practice This study highlights that AI‐driven deep learning and predictive models are becoming essential tools for clinical decision‐making in critical care nursing. Integrating these technologies into routine practice can facilitate earlier disease detection and risk prediction, ultimately enhancing the quality of care and safety for critically ill patients.
Wang et al. (Fri,) studied this question.