Machine Learning in Tuberculosis Research: A Global Bibliometric Analysis of Diagnostic, Prognostic, and Drug Discovery Trends.

Abstract

Background and objectives: Tuberculosis (TB) remains a major global health challenge, driving the need for innovative approaches in diagnosis and drug development. The integration of artificial intelligence (AI), particularly machine learning (ML), has enabled significant advancements in areas such as drug resistance prediction, radiomics, prognostic modeling, and computational drug discovery. This study presents a comprehensive bibliometric analysis of global research on machine learning and tuberculosis (MLTB), highlighting trends relevant to therapeutic innovation and regulatory science.

Methods: A structured search of the Scopus database was conducted for English-language, data-driven publications on MLTB through May 1, 2024. Bibliometric indicators were analyzed using Biblioshiny and VOSviewer, focusing on publication trends, citation metrics, collaboration networks, and thematic clustering.

Results: The MLTB research field has grown rapidly, with an average annual growth rate of 22.12% between 2000 and 2024. Publications averaged 21.64 citations, and 40.11% involved international collaboration. Twelve major clusters were identified, including deep learning, drug discovery, bioinformatics, docking, random forest, and latent TB infection-highlighting the field's expanding scope in drug development and diagnostic applications.

Conclusion: MLTB research is evolving rapidly, driven by interdisciplinary collaboration and AI innovation. These findings offer insights for guiding future AI-enabled TB therapeutic strategies and aligning research efforts with regulatory and translational priorities.