Artificial intelligence research in journals indexed in the web of science "infectious diseases" category: a...

Artificial intelligence (AI) is increasingly being applied in the field of infectious diseases. This study aimed to characterize publication trends, collaboration networks, and thematic patterns among AI-related original articles published between 2016 and 2025 in journals assigned to the Web of Science Core Collection (WoSCC) "Infectious Diseases" category.

A systematic bibliometric analysis was performed using the WoSCC. The query combined AI-related terms (machine learning, deep learning, neural networks, large language models, and allied concepts) restricted to the WoSCC category of "Infectious Diseases" and an English-language filter, yielding 1,252 original articles published between January 2016 and December 2025. Bibliometric computations and visualizations were conducted using the Bibliometrix R package, VOSviewer, and Scimago Graphica. Keyword co-occurrence network analysis and temporal trend mapping were employed to identify thematic clusters and emerging research foci.

The 1,252 articles were contributed by 9,160 authors from 124 countries across 2,850 institutions and published in 125 journals. Annual output grew more than 30-fold between 2016 and 2025, with 84.7% of all publications appearing in the final five years. The United States and China were the most productive countries and collectively dominated international collaboration networks. Harvard University was the leading institution. Keyword network analysis identified nine distinct thematic clusters. Recent trend analyses reveal a significant shift towards clinical applications, specifically highlighting antimicrobial resistance surveillance and early sepsis prediction as dominant research hotspots.

Scientific production on AI applications in infectious diseases has expanded exponentially over the past decade, catalyzed principally by the COVID-19 pandemic. Sepsis management, antimicrobial resistance, surveillance, and clinical prediction represent the most prominent research themes. These patterns indicate areas of growing scientific attention. This study may help guide future clinical validation and implementation research.