Abstract
Objectives
Accurate dental age (DA) estimation is essential in pediatric dentistry, orthodontic treatment planning, and forensic practice; however, the applicability of international reference standards varies across populations. This study aimed to evaluate the accuracy of a population-adapted Nolla method for dental age estimation in children from northwestern Romania and to compare its performance with the original approach.Materials and methods
A retrospective analysis was performed on 860 digital panoramic radiographs of children aged 3-15 years. Dental development was assessed using the Nolla staging system, and a third-degree polynomial regression model was developed to estimate chronological age (CA) based on the total Nolla score. Model performance was evaluated using coefficient of determination (R²), five-fold cross-validation, and non-parametric statistical tests.Results
The original Nolla method underestimated CA by approximately 1.5 years. The population-specific model showed strong predictive performance (R² = 0.846; cross-validated R² = 0.84) and close agreement between estimated DA and CA (mean difference 0.000 ± 0.976 years; p = 0.983). Accuracy was stable across most age groups, although larger deviations were observed after 12 years. Girls showed slightly more advanced dental maturation than boys, but differences were clinically small.Conclusions
The proposed population-calibrated equation eliminates systematic bias and provides a reliable tool for dental age estimation in children from northwestern Romania, although external validation is recommended.