RT2C: Predicting Time-to-New Caries with Structured Dental Data Using RNN
X Liu1,2, K K Kookal3, R Brandon4
1McWilliams School of Biomedical Informatics, University of Texas Health Science Center in Houston, Houston, TX, USA.
Journal of Dental Research
|July 18, 2026
Summary
A new recurrent neural network model, RT2C, accurately predicts dental caries risk and time-to-event, improving upon existing methods for personalized patient care.
Area of Science:
- Oral health research
- Artificial intelligence in healthcare
- Predictive modeling in dentistry
Background:
- Dental caries is a prevalent chronic condition.
- Existing risk models often provide binary predictions and lack time-to-event insights.
- Modern caries management requires accurate risk assessment during routine dental visits.
Purpose of the Study:
- To develop RT2C, a survival model using recurrent neural networks (RNNs).
- To assess the likelihood of new caries diagnosis by the next dental visit.
- To estimate the time until caries events occur using longitudinal dental data.
Main Methods:
- Utilized structured electronic dental record (EDR) data from 466,782 patients.
- Employed a bidirectional gated recurrent unit network for temporal dependency analysis.
- Compared RT2C performance against Cox proportional hazards and random survival forest models.
Main Results:
- RT2C achieved a high concordance index (0.883 ± 0.08) on the test set.
- The model outperformed traditional survival baselines and CRA-only predictions.
- Consistent performance was observed across diverse demographic and clinical subgroups.
Conclusions:
- RT2C effectively models visit-indexed risk trajectories for caries prediction.
- The model enhances existing Caries Management by Risk Assessment (CAMBRA) tools.
- RT2C supports proactive and personalized dental caries management by integrating longitudinal EDR data.