Objective: To map global research trends in artificial intelligence (AI) applications for oral cancer diagnosis using bibliometric analysis. Design: Publications retrieved from Scopus and PubMed (2000–2025) were analyzed using VOSviewer for keyword co-occurrence and thematic clustering. A total of 523 unique records were included. Eight clusters were identified and interpreted. Results: A total of 777 records were retrieved from Scopus and PubMed, of which 523 unique publications were included after deduplication and screening. Keyword co-occurrence analysis using VOSviewer identified 66 frequently occurring keywords organized into 8 thematic clusters, highlighting major research domains spanning image-based deep learning, early detection and screening of OPMDs, prognostic modelling and radiomics, and classical machine learning approaches. Conclusions: Bibliometric mapping highlights the interdisciplinary growth of artificial intelligence in oral cancer diagnosis. The study provides a comprehensive strategic overview for researchers and clinicians aiming to align with emerging diagnostic paradigms that will be crucial for advancing early detection strategies, enhancing diagnostic precision, and facilitating the integration of AI-driven technologies into routine clinical practice in oral oncology.
Dalmia et al. (2026) studied this question.