Artificial Intelligence in Tuberculosis: Global Research Trends and Bibliometric Insights (2000–2025) (Preprint)

Abstract

<sec> <title>BACKGROUND</title> Tuberculosis (TB) remains a major global health challenge despite prevention efforts. Artificial intelligence (AI) shows promise for TB management through imaging analysis and decision support, but systematic evaluation of research trends and implementation challenges is needed. </sec> <sec> <title>OBJECTIVE</title> To analyze development patterns, collaboration networks, and knowledge structure of AI applications in TB management through bibliometric analysis of Web of Science publications, identifying research frontiers and translational gaps. </sec> <sec> <title>METHODS</title> This study conducted bibliometric analysis of Web of Science publications (2000-2025) using VOSviewer, CiteSpace, and Bibliometrix R, examining publication trends, collaboration networks, and knowledge domains through scientific knowledge mapping. </sec> <sec> <title>RESULTS</title> Analysis shows substantial growth in AI applications for TB management during 2000-2025, with post-2018 publication growth of 29. 7% and citation growth of 54. 2% annually. Research concentrates on imaging analysis (53. 3% of highly-cited papers), drug resistance prediction (20%), and drug discovery (26. 7%). The US leads with 346 publications, 12,143 citations, and collaboration strength of 409, while China ranks second with 304 publications but lower collaboration (141) and citations (4,251). High-income nations focus on technology development; high-burden countries emphasize clinical translation. Technological journals focus on algorithm optimization, while clinical journals prioritize validation. "Machine learning" dominates the keyword landscape (1,459 occurrences), while clinical terminology such as "intensive care unit" (243) appears frequently but at substantially lower frequencies. Of highly-cited studies, 60% focus on technical validation and only 26. 7% address clinical applications. Main research gaps include insufficient cross-regional data sharing, limitations in atypical TB image recognition, and poor integration of socioeconomic variables. </sec> <sec> <title>CONCLUSIONS</title> AI-tuberculosis research shows rapid development focused on technical innovation, but clinical translation remains limited. Future advances require strengthening interdisciplinary collaboration and implementation science to transition from technology-driven to needs-based approaches, particularly addressing deployment challenges in high-burden settings. </sec>

Ähnliche Arbeiten