Artificial intelligence in temporomandibular disorders: A bibliometric analysis of research trends, hotspots, and emerging frontiers

BACKGROUND: Artificial intelligence (AI) is increasingly applied to temporomandibular disorders (TMD), particularly in imaging, diagnosis, classification, and prediction. However, the global research landscape, knowledge structure, and evolving research priorities remain insufficiently characterized. This study aimed to systematically map the development and emerging trends of AI research in TMD. METHODS: A total of 311 publications from 2015 to 2026 were retrieved from the Web of Science Core Collection. Bibliometric and visualization analyses were conducted using CiteSpace, VOSviewer, and R (bibliometrix) to examine publication trends, scientific collaboration, citation networks, research hotspots, and thematic evolution. RESULTS: Publication output increased markedly after 2019, reaching its highest level in 2025. China contributed the most publications (n = 111), followed by the United States (n = 70), South Korea (n = 45), and Turkey (n = 33), with differences in international collaboration patterns. The knowledge structure reflected interdisciplinary connections among dentistry, medical imaging, computer science, and engineering. Major research hotspots included machine learning, deep learning, magnetic resonance imaging, automated diagnosis, disease classification, and prediction. Temporal analyses revealed a shift from algorithm development toward imaging-based assessment, TMD classification, and clinically oriented predictive applications. CONCLUSION: AI-related TMD research has expanded rapidly, with increasing emphasis on diagnostic and predictive applications. Future research should prioritize standardized protocols, multicenter datasets, external validation, and evaluation of clinical applicability to facilitate reliable integration into TMD management.