<![CDATA[Introduction: The high burden of childhood dental caries highlights the need for innovative approaches. Artificial Intelligence (AI) offers a breakthrough in enabling more efficient, measurable efforts to prevent and detect dental caries. Therefore, this scoping review aims to map the existing evidence on the use of AI interventions in the prevention of childhood dental caries based on Leavell and Clark’s prevention model. Review: This scoping review analyzed 49 studies from various countries that examined the application of AI in childhood caries prevention, using the Leavell and Clark framework. At the health promotion level, behavior-based chatbot applications were shown to significantly improve caregivers’ knowledge and toothbrushing habits. In the specific protection setting, predictive models based on tabular data and the salivary microbiome identified high-risk children before lesions formed, with AUC values ranging from 0.77 to 0.94. For early detection, deep learning algorithms trained on radiographic images and intraoral photographs demonstrated high accuracy in detecting caries lesions, including incipient lesions, with a sensitivity of up to 94.7%. For disability prevention, AI models successfully predicted the success of caries arrest therapy with silver diamine fluoride. The main limitations identified were a small sample size, a lack of cross-population external validation, and gaps in model interpretability at the clinical level. Conclusion: The use of AI interventions for childhood dental caries is unevenly distributed across Leavell and Clark's five preventive levels, especially for specific protection.]]>
