The advancement of artificial intelligence in point-of-care ultrasound (POCUS): A bibliometric analysis

Abstract

Introduction: Artificial intelligence (AI) has advanced image interpretation capabilities, and AI has gained increasing prominence in medical imaging. Point-of-care ultrasound (POCUS) is a key imaging tool where AI can be applied. A bibliometric study on AI in POCUS has not yet been published in the literature. Therefore, using bibliometric and statistical methods, we aimed to analyze publications on AI in POCUS. Methods: This research is a bibliographic and descriptive analytical study. The Web of Science database was utilized to identify existing publications and conduct the analyses. A non-linear regression analysis (using an exponential model) was employed to predict the number of publications over the following years. Additionally, the study employed bibliometric network visualization, world map visualization, and publication relationship visualization. Results: Examining the distribution of publications related to AI in POCUS, 77.2% (207) are articles, 17.5% (47) are review articles, and the remaining are other types of publications. The strong relationship between lung ultrasound and AI reflects the impact of the COVID-19 pandemic on research. However, the high frequency of terms related to multiple systems, including pulmonary, cardiovascular, trauma-related, and neurologic conditions, suggests wide-ranging clinical applicability. These findings may help guide the planning and prioritization of future AI-based research in POCUS. Conclusion: This bibliometric study on artificial intelligence in POCUS examined 268 publications. The analyses in this study can be a valuable resource for researchers working on AI in POCUS or who will work on AI in POCUS in the future.