Related Experiment Video
Updated: Jun 16, 2025
Precision of In Vivo Quantitative Tooth Wear Measurement Using Intra-Oral Scans
Published on: July 12, 2022
The Brazilian (FRANCO) method for dental age estimation: Willems' model revisited
Ademir Franco1,2, Débora Duarte Moreira3, Rafael Cidade3
1Division of Forensic Dentistry, Faculdade São Leopoldo Mandic, Rua Dr José Rocha Junqueira 13, Swift Campinas, Block E Office 3, São Paulo, 045-755, Brazil. ademir.junior@slmandic.edu.br.
Objectives:
This study aimed (I) to test the Willems' dental age estimation method in different geographic samples of the Brazilian population, and (II) to propose a new model combining the geographic samples in a single reference table of Brazilian maturity scores.
Materials And Methods:
The sample consisted of 5017 panoramic radiographs of Brazilian males (n = 2443) and females (n = 2574) between 6 and 15.99 years (mean age = 10.99 ± 2.76 years). The radiographs were collected from the Southeastern (SE) (n = 2920), Central-Western (CW) (n = 1176), and Southern (SO) (n = 921) geographic regions. Demirjian's technique was applied followed by Willems' method and the proposed Brazilian model.
Results:
Willems' method led to mean absolute errors (MAE) of 0.79 and 0.81 years for males and females, respectively. Root mean squared errors (RMSE) were 1.01 and 1.03 years, respectively. The Brazilian model led to MAE of 0.72 and 0.74 years for males and females, respectively, and RMSE of 0.93 years for both sexes. The MAE was reduced in 70% of the age categories. Differences between regions were statistically (p < 0.05) but not clinically significant.
Conclusion:
The new model based on a combined population had an enhanced performance compared to Willems' model and led to reference outcomes for Brazilians.
Clinical Relevance:
Assessing patients' biological development by means of dental analysis is relevant to plan orthopedic treatments and follow up. Having a combined-region statistic model for dental age estimation of Brazilian children contributes to optimal age estimation practices.
