Artificial Intelligence in Coronary Computed Tomography Angiography: A Bibliometric Analysis of Trends and Themes

Abstract

Background

Artificial intelligence (AI) in coronary computed tomography angiography (CCTA) is increasingly recognized, but comprehensive bibliometric analyses on its integration are scarce.

Hypothesis

This study aims to bridge this gap by delineating research trajectories, influential entities, and evolving themes within this field.

Methods

A bibliometric analysis was conducted on literature from the Web of Science Core Collection, spanning publications from January 2007 to October 2024. Analysis tools included VOSviewer, CiteSpace, and R-bibliometrix, which were used to map and visualize the publication trends, collaborative networks, and thematic evolution. A manual relevance audit and a sensitivity analysis using a narrower query were performed to validate the search strategy.

Results

A total of 499 articles were analyzed. The number of publications increased annually in the past decade, with a peak of 113 in 2022. China led with 193 articles, and the USA served as a central node in the cooperation network. The Medical University of South Carolina and Yonsei University tied for first with 54 articles each. Schoepf U. Joseph was the most influential author. European Radiology led in h-index and total publications. The keyword analysis revealed “coronary artery disease” as the most frequent, followed by “deep learning” and“machine learning.” Notable burst keywords included “fractional flow reserve,” “diagnostic accuracy,” and “SCCT guidelines.”

Conclusion

The field of AI in CCTA is rapidly evolving, with increasing international collaboration and a focus on technological advancements. Machine learning prediction has emerged as a bibliometrically prominent theme with intense citation activity, warranting further validation and standardized integration into clinical practice.