Mapping knowledge landscapes and emerging trends in AI for coronary artery disease imaging biomarkers: A bibliometric...

With the rapid advancement of artificial intelligence (AI) in medical imaging, its application to coronary artery disease (CAD) imaging biomarkers has become a key area of interdisciplinary research. Understanding the current developmental trajectory, research focus, and collaborative landscape in this field is of significant importance.

This study aimed to comprehensively assess the global research status, emerging trends, knowledge structure, and collaborative networks in the application of AI to CAD imaging biomarkers through a bibliometric and visual analysis.

Based on data from the Web of Science Core Collection (2015-2025), this study conducted a bibliometric analysis using tools such as VOSviewer, CiteSpace, and Bibliometrix. The analysis encompassed publication trends, author and institutional collaborations, national contributions, journal profiles, citation networks, keyword evolution, and interdisciplinary interactions.

A total of 1,105 publications were included, involving 5,949 authors, 1,903 institutions, and 262 journals from 67 countries. Publication output grew rapidly, with an average annual growth rate of 58.74% during 2015-2018 and 64.33% during 2019-2022, before stabilizing at 12.67% in 2023-2025. The United States and China were the leading contributors in both publication volume and citations. Research hotspots centered on "deep learning," "machine learning," and "coronary computed tomography angiography," with keyword clustering revealing eight thematic groups, including image segmentation, radiomics, and multimodal prediction. Disciplinary analysis indicated strong links between radiology and cardiovascular systems, but limited integration with engineering and computer science.

The field of AI for CAD imaging biomarkers is in a phase of steady growth, attracting widespread global participation. However, research collaboration remains fragmented, and interdisciplinary integration needs strengthening. Future directions may focus on advancing deep learning applications, enhancing multimodal data fusion, and promoting cross-disciplinary cooperation to translate AI tools into clinical practice.