Bibliometric Analysis of Artificial Intelligence in Healthcare Diagnostics: A Decade of Research Trends and Emerging Frontiers (2015–2025)

Authors

  • Alberto Brunete Assoc. Prof., Department of Electrical, Electronic and Automation Engineering and Applied Physics, Universidad Politécnica de Madrid, Madrid, Spain Author
  • Leonor Cerdá-Alberich Assoc. Prof., Biomedical Imaging Research Group, La Fe Health Research Institute & TIC Salut Social Foundation, Barcelona, Spain Author

Keywords:

Artificial Intelligence, Healthcare Diagnostics, Bibliometric Analysis, Deep Learning, Medical Imaging, Machine Learning, Vosviewer, Publication Trends

Abstract

The integration of artificial intelligence (AI) into healthcare diagnostics has emerged as one of the most transformative developments in modern medicine. This bibliometric analysis examines the scholarly landscape of AI in healthcare diagnostics over the decade spanning 2015 to 2025. A systematic search of the Scopus database was conducted using a comprehensive search strategy to identify relevant publications. Bibliometric indicators including publication trends, geographic distribution, institutional productivity, author collaboration networks, keyword co-occurrence patterns, and citation analysis were extracted and analyzed using VOSviewer and the Bibliometrix package in R. The analysis revealed exponential growth in publications, with annual output increasing from 127 papers in 2015 to 4,587 papers in 2023. The United States and China emerged as the dominant contributors, collectively accounting for 44.4% of total publications. Radiology and medical imaging constituted the primary application domain, representing approximately 75% of AI diagnostic devices approved by regulatory bodies. Deep learning, convolutional neural networks, and machine learning were identified as the most frequently occurring keywords, while emerging themes included explainable AI, federated learning, and large language models. The findings reveal a maturing field transitioning from proof-of-concept studies toward clinical validation and regulatory approval, with significant opportunities remaining in addressing algorithmic bias, ensuring regulatory harmonization, and achieving equitable global access. This study provides a comprehensive quantitative mapping of the intellectual structure and evolution of AI in healthcare diagnostics, offering valuable insights for researchers, clinicians, policymakers, and industry stakeholders.

Downloads

Published

2026-05-15

Issue

Section

Articles

Similar Articles

You may also start an advanced similarity search for this article.