An AI-Assisted pictorial measure of children's dental anxiety during the initial child-dentist interaction.

BACKGROUND: The potential influence of the initial child-dentist interaction on dental anxiety (DA) in preschool children remains insufficiently understood. Accurate assessment of DA using pictorial measures is essential to better characterize this relationship. However, pictorial self-report measures still have several limitations from a contemporary perspective, and evidence regarding the impact of the initial interaction on DA remains limited. This study aimed to evaluate the effect of the initial child-dentist interaction on DA and to explore potential Artificial Intelligence (AI)-based refinements in traditional pictorial measures. METHODS: An observational pre-post (within-subject repeated-measures) study included 171 children aged 3-6 years attending their first dental visit. An AI-assisted pictorial measure (AI-PM) was designed through an expert panel process addressing the limitations of existing tools. DA was assessed pre and post the initial child-dentist interaction using established pictorial measures and the new tool. Validity, reliability (intraclass correlation coefficient), and inter-measurement agreement (Bland-Altman analysis) were evaluated. Changes in anxiety levels were analyzed using the Wilcoxon signed-rank test (p < 0.05). RESULTS: Anxiety scores decreased significantly post initial child-dentist interaction (p < 0.001). The AI-PM demonstrated good-to-excellent validity and reliability, with strong correlations with established pictorial measures (Spearman r = 0.72-0.96, p < 0.001). The reliability analysis using the intraclass correlation coefficient demonstrated good-to-excellent agreement between the AI-based assessment and the traditional measures. Agreement with traditional measures was good-to-excellent, with no evidence of proportional bias in Bland-Altman analyses (p > 0.05). CONCLUSIONS: The findings suggest that the initial child-dentist interaction may play an important role in children's DA and that the AI-PM may provide a basis for further refinement of traditional pictorial measures used to assess DA in early childhood. CLINICAL TRIAL REGISTRATION: The study has been retrospectively registered in ClinicalTrials.gov (Identifier: NCT07387055).