LAUSR

Objectives: The aim of the present study was to develop a machine learning model to predict the risk of molar incisor hypomineralization (MIH) and to identify factors associated with MIH in an endemic fluorosis region in central China. Methods: A cross-sectional study was conducted with 1,568 schoolchildren from selected regions. The clinical examination included an investigation of MIH based on the European Academy of Paediatric Dentistry (EAPD) criteria. In this study, supervised machine learning (e.g., logistic regression) and correlation analysis (e.g., Spearman correlation analysis) were used for classification and prediction. Results: The overall prevalence of MIH was 13.7%. The nomograph showed that non-dental fluorosis (DF) had a considerable influence on the early occurrence of MIH and that this influence became weaker as DF severity increased. We examined the association between MIH and DF and found that DF had a protective correlation with MIH; the protective effect became stronger as DF severity increased. Furthermore, children with defective enamel were more likely to experience caries, and dental caries were positively correlated with MIH (OR = 1.843; 95% CI: 1.260–2.694). However, gender, oral hygiene, and exposure to poor-quality shallow underground water did not increase the likelihood of developing MIH. Conclusions: DF should be considered a protective factor within the multifactorial etiology of MIH.