Artificial Intelligence Transforms East Asian Traditional Medicine: A Bibliometric Study of Research Trends

Abstract

Background: East Asian traditional medicine (EATM) faces significant challenges in standardization. Artificial intelligence (AI) presents opportunities to overcome these hurdles through advanced pattern recognition and clinical decision support. This study provides the first comprehensive bibliometric analysis of global research at the intersection of EATM and AI to map publication trends and assess the field’s intellectual structure. Methods: We analyzed 1253 publications retrieved from the Web of Science Core Collection, covering the period from its inception to June 2025. The bibliometric analysis was conducted using VOSviewer to examine publication trends, keyword co-occurrence, co-authorship patterns, and co-citation networks. Results: Research output has grown dramatically, from a single article in 1994 to 253 articles in 2024, with a notable acceleration since 2020. Keyword analysis identified three primary research clusters: (1) computational pharmacology, (2) AI-driven diagnostics, and (3) spectroscopic quality control, showing a technological progression from classical machine learning to deep learning. China overwhelmingly dominated the field with 1108 publications (88.4%), leading to dense domestic institutional networks but limited international collaboration. Author networks were highly fragmented, with only 101 of 6428 authors meeting a five-publication threshold. Co-citation analysis revealed a heavy reliance on Chinese databases (e.g., TCMSP, SymMap, ETCM) and the direct importation of computer vision techniques. Conclusions: The EATM–AI research field is characterized by rapid expansion and increasing technical sophistication, but it is constrained by significant structural limitations. The concentration of research within China, while fostering advancement, creates barriers to global validation and clinical translation. To realize the field’s transformative potential, strategic development of distributed research infrastructure and cross-cultural validation frameworks is essential.