Artificial intelligence (AI) is rapidly emerging as a transformative technology in healthcare, particularly in the demanding field of critical illness medicine. Critical illness presents unique challenges due to high data complexity, significant individual patient differences, and rapid changes in conditions, which traditional methods often struggle to manage effectively. AI, especially through machine learning and deep learning, offers novel solutions by enabling earlier disease identification, predicting disease progression, and enhancing clinical decision-making through the analysis of vast amounts of patient data. Given the increasing volume of research in this area, a systematic bibliometric review is crucial to understand current trends and future directions.
This study, titled “A bibliometric analysis of artificial intelligence research in critical illness: a quantitative approach and visualization study,” systematically analyzed 900 articles published between 2005 and 2024 from the Web of Science database. Utilizing bibliometric methods and visualization tools like R-bibliometrix, VOSviewer, and CiteSpace, the research mapped the landscape of AI application in critical illness.
The findings highlight a significant surge in publications since 2020, indicating a rapidly growing interest in this field, with a rapid growth from 73 articles in 2018 to 230 in 2023. The United States is identified as the leading contributor, with Harvard University having the highest betweenness centrality among institutions, closely followed by China in publication volume and collaborative research efforts. Noseworthy PA is recognized as a core author due to his high publication and citation counts, demonstrating significant influence in the field. Key journals include Frontiers in Cardiovascular Medicine and Diagnostics, which lead in publication output.
Crucially, the study identifies heart failure and sepsis as key research hotspots where AI demonstrates tremendous potential for identification and management. Recent keyword trends underscore the growing importance of AI in electrocardiogram analysis and cardiology, indicating a shift towards high-precision medical diagnosis and treatment planning.
Despite its vast potential in enhancing diagnostic accuracy, personalized treatment, and clinical decision support, the widespread clinical application of AI in critical illness faces notable challenges. These include critical concerns such as data privacy, the interpretability of AI models (the “black box” nature), and complex ethical issues. Future research is urged to prioritize the transparency, interpretability, and rigorous clinical validation of AI models to ensure their effectiveness and safety in clinical settings. Strengthening interdisciplinary cooperation and promoting data sharing are also vital for advancing the field.
Reference for the Article:
Luo, Z., Lv, J., & Zou, K. (2025). A bibliometric analysis of artificial intelligence research in critical illness: a quantitative approach and visualization study. Frontiers in Medicine, 12, 1553970. https://doi.org/10.3389/fmed.2025.1553970
