Authors
Deepika Kapoor, Deepanshu Garg, Pavithraa Jayakumar, Chetna Arora, Alok Dubey, Kailash C Dash, Prajwal Shetty, Sheetal Mujoo, Nischitha Naik
Published in
The journal of contemporary dental practice. Volume 26. Issue 12. Pages 1173-1177. Dec 01, 2025. Epub Dec 01, 2025.
Abstract
To develop and evaluate machine learning models for predicting new caries development in children using clinical and behavioral variables, and to identify the most influential predictors contributing to caries risk.
A dataset of 200 preschool children was generated, including variables such as age, sugar intake frequency, baseline dmft, fluoride exposure, mutans streptococci levels, socioeconomic status, oral hygiene, and previous caries. Logistic Regression and Random Forest models were trained and tested using an 80:20 split. Model performance was assessed using accuracy, sensitivity, specificity, precision, and a forest plot of feature importance. Descriptive statistics, group comparisons, and confusion matrices were produced.
Children who developed new caries at 12 months had higher baseline dmft (1.26 vs 0.93), higher sugar intake (2.62 vs 2.00 times/day), and a greater frequency of high mutans levels (61 vs 38%). Random Forest outperformed Logistic Regression with higher accuracy (0.847 vs 0.783) and better sensitivity and specificity. Feature-importance analysis identified Mutans-high, sugar frequency, previous caries, and baseline dmft as the top predictors. The confusion matrix for the Random Forest showed strong predictive reliability [True negatives (TN) = 50, True positives (TP) = 9, False positives (FP) = 4, False negatives (FN) = 5].
Machine learning demonstrates strong potential for early prediction of dental caries in children. The Random Forest model showed superior discriminatory performance and identified key risk factors that align with established clinical evidence.
Traditional caries risk assessment tools have limited predictive accuracy in preschool children. Artificial Intelligence-based Machine learning offers enhanced predictive capabilities and helps to identify early prediction of caries risk for caries prevention and management. How to cite this article: Kapoor D, Garg D, Jayakumar P, et al. Artificial Intelligence-based Caries Risk Prediction Model for Preschool Children. J Contemp Dent Pract 2025;26(12):1173-1177.
PMID:
42606992
Bibliographic data and abstract were imported from PubMed on 18 Aug 2026.
Read full publication at:
Please sign in
to see all details.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 5
- Comments 0