Integration of artificial intelligence in public health dentistry: applications, challenges, and future directions -...

Artificial intelligence (AI) is transforming healthcare delivery globally, with increasing applications in public health dentistry. This scoping review maps and synthesises the current state of AI integration in public health dentistry practice, education, research, and surveillance. A comprehensive literature search was conducted across PubMed, Scopus, Web of Science, Embase, and the Cochrane Library (with Google Scholar used as a supplementary source) from database inception to December 31, 2024. Thirteen studies were included to map the range of evidence on AI applications, performance metrics, and implementation challenges using PRISMA-ScR guidelines and a structured narrative synthesis. Consistent with scoping review methodology, this study aimed to map existing evidence and identify research gaps rather than evaluate or establish the effectiveness of AI interventions. In preliminary studies conducted under varied and predominantly controlled or pilot conditions, AI applications were reported across multiple public health dentistry domains: (1) disease surveillance, with reported machine learning accuracy of 82-94% for caries prediction in individual studies; (2) community screening, with reported sensitivity of 85-92%; (3) health education, with reported 18-23% improvements in knowledge scores; (4) tele-dentistry, with reported diagnostic concordance of 81-87%; and (5) policy planning, with reported utilisation prediction accuracy of 76-82%. These figures are derived from heterogeneous, predominantly pilot studies and should not be interpreted as indicators of real-world effectiveness. Major implementation challenges included data quality issues (9/13 studies, 69%), algorithmic bias concerns (8/13, 62%), privacy and security barriers (7/13, 54%), and AI literacy gaps (10/13, 77%). Included studies suggest that AI may support several functions within public health dentistry; however, these conclusions are drawn from a small and heterogeneous evidence base dominated by pilot studies and narrative reviews, limiting generalisability to routine public health dental settings. Successful implementation will require addressing data quality, algorithmic transparency, workforce training, and ethical considerations.