AI in Oral Health Surveillance: Critical Review

Clinical Snapshot

Key Findings

• P Value: Not reported

• Effect Size: Not reported — no pooled effect sizes are calculated; the review is qualitative and conceptual in nature

• Primary Outcome: Narrative synthesis of AI applications in oral health surveillance across three domains: (1) population-level oral health trend mapping and equity analysis using traditional machine learning on structured epidemiological data; (2) remote screening of oral diseases (caries, gingivitis, oral cancer, malocclusion) from intraoral images using computer vision models; (3) multimodal data integration using large language models and multimodal LLMs for precision public health

• Nnt Or Sensitivity: No NNT, sensitivity, specificity, or AUC values are pooled or reported in the abstract; individual study performance metrics referenced within the full text are not summarised in the abstract

• Confidence Interval: Not reported

Clinical Application

Feasibility of immediate clinical or public health implementation is limited by the review's own identified barriers: lack of standardised imaging protocols for non-clinical settings, domain shift between training and deployment environments, class imbalance in surveillance datasets, and unresolved cost-effectiveness questions. LLM/MLLM-based multimodal surveillance systems remain largely aspirational and are not yet validated at scale. Traditional ML applications to structured epidemiological data are more mature but still require careful validation in new population contexts. Australia has established national oral health surveillance infrastructure through the Australian Institute of Health and Welfare (AIHW) and the Australian Research Centre for Population Oral Health (ARCPOH), which conducts the National Child Oral Health Study and the National Survey of Adult Oral Health. AI-assisted tools for remote oral screening could be particularly relevant in the Australian context given the significant oral health disparities affecting rural and remote communities, Aboriginal and Torres Strait Islander peoples, and aged care residents with limited dental access. The Chronic Disease Dental Scheme and Child Dental Benefits Schedule (CDBS) provide frameworks within which AI-assisted triage or screening could theoretically be integrated, though no TGA-approved AI oral health surveillance tools are currently listed. RACGP guidelines on preventive oral health in primary care settings could provide a pathway for integrating AI-assisted screening referral pathways. Any deployment would require TGA regulatory assessment as a Software as a Medical Device (SaMD) under the current regulatory framework. Data sovereignty considerations, particularly for Indigenous health data, would require engagement with community-controlled health organisations and compliance with the AIATSIS Code of Ethics. Public health practitioners, dental epidemiologists, health system planners, and clinical informaticists involved in population-level oral health monitoring. Potentially applicable to screening programmes targeting high-risk groups (children, elderly, socioeconomically disadvantaged populations) where remote or AI-assisted surveillance could extend reach beyond traditional clinical settings.

Abstract

Artificial intelligence (AI) holds transformative potential for advancing oral health surveillance by streamlining data collection, integration, and dissemination. This review critically synthesizes AI applications in oral health surveillance, highlighting its roles in 1) mapping population-level trends and oral health inequities using machine learning on epidemiological data; 2) enabling remote screening of oral diseases/conditions, including caries, oral hygiene, gingivitis, oral cancer, and malocclusion from intraoral images via computer vision models; and 3) integrating multimodal data through emerging large language models (LLMs) to enhance precision public health. We clarify the comparative strengths of distinct AI modeling for processing the primary data types in surveillance: structured clinical records, unstructured images, and integrated multimodal data. Traditional machine learning methods have been effectively applied to map population-level oral health disparities and identify risk factors but are constrained to structured data. Computer vision methods excel in individual-level diagnostics using intraoral photographs. To translate such capability into scalable surveillance, it is recommended to establish standardized imaging protocols for nonclinical settings, develop scalable models for fine-grained feature extraction, and implement reliable evaluation. These steps are essential to address pervasive challenges, including inconsistent image quality, domain shift, prevalence imbalance, and cost-effectiveness constraints. The future of AI-driven oral health surveillance lies in developing dental-adapted multimodal LLMs (MLLMs). Such MLLMs are uniquely capable of synthesizing disparate data streams, from structured clinical data to heterogeneous imaging modalities (e.g., intraoral photographs and radiographs), or even biomolecular data. This integration capacity facilitates a paradigm shift, moving current applications in dental consultation and clinical decision support toward a novel, tiered system for population-level monitoring. Such a system would provide actionable insights for public health policymaking via spatiotemporal analysis and causal inference. Next-generation AI-driven oral health surveillance systems can only succeed when built on a strong foundation of rigorous ethical principles and safeguards.

References

1. 1.Chen, Z., Liu, P., Han, K., Liao, P., Yang, Y., Wong, M. C. M., Yiu, C. K. Y., & Lo, E. C. M. (2026). AI in oral health surveillance: Critical review. Journal of Dental Research. Advance online publication. https://doi.org/10.1177/00220345261434568