Age 18 years is considered as the age of majority by most countries. To ascertain the age of interest, both third molar development (TMD) and eruption (TME) staging scores are beneficial without needing multiple imaging modalities. This study aimed to assess the chronological course of TMD and TME in a Malay sub-adult population and evaluate predictions when specific stage(s) of TMD and TME have been attained that are pertinent to the age group of interest (<18 years or ≥18 years). A sample of 714 digital panoramic images for subjects stratified by age between 14.1 and 23.9 years was retrospectively collected. The techniques described by Gleiser and Hunt (modified by Kohler) and Olze were employed to stage TMD and TME, respectively. A binary logistic regression was performed to predict the 18-year threshold with staging score as predictors. Stages 4-6 (TMD) and A-B (TME) for males and stages 4 (TMD) and A (TME) for females were found to discriminate the <18-year group. For both genders, stages 9-10 (TMD) and D (TME) can be used as reference stages to estimate whether a subject is likely to be ≥18 years, with 94.74-100% and 85.88-96.38% correct predictions, respectively. Stages 4 (TMD) and A (TME) can also be used to identify juveniles (<18 years) with a high degree of correct predictions, 100%. The juvenility of an individual is easily anticipated by using the specific staging scores of both third molar variables (TMD and TME) without complex calculations.
The third molar development (TMD) has been widely utilized as one of the radiographic method for dental age estimation. By using the same radiograph of the same individual, third molar eruption (TME) information can be incorporated to the TMD regression model. This study aims to evaluate the performance of dental age estimation in individual method models and the combined model (TMD and TME) based on the classic regressions of multiple linear and principal component analysis. A sample of 705 digital panoramic radiographs of Malay sub-adults aged between 14.1 and 23.8 years was collected. The techniques described by Gleiser and Hunt (modified by Kohler) and Olze were employed to stage the TMD and TME, respectively. The data was divided to develop three respective models based on the two regressions of multiple linear and principal component analysis. The trained models were then validated on the test sample and the accuracy of age prediction was compared between each model. The coefficient of determination (R²) and root mean square error (RMSE) were calculated. In both genders, adjusted R² yielded an increment in the linear regressions of combined model as compared to the individual models. The overall decrease in RMSE was detected in combined model as compared to TMD (0.03-0.06) and TME (0.2-0.8). In principal component regression, low value of adjusted R(2) and high RMSE except in male were exhibited in combined model. Dental age estimation is better predicted using combined model in multiple linear regression models.
Through numerous validation and method comparison studies on different populations, the Willems method exhibited a superior accuracy. This article aims to systematically examine how accurate the application of Willems dental age method on children of different age groups and its performance based on various populations and regions. A strategic literature search of PubMed, MEDLINE, Web of Science, EMBASE and hand searching were used to identify the studies published up to September 2014 that estimated the dental age using the Willems method (modified Demirjian), with a populations, intervention, comparisons and outcomes (PICO) search strategy using MeSH keywords, focusing on the question: How much Willems method deviates from the chronological age in estimating age in children? Standardized mean differences were calculated for difference of dental age to chronological age by using random effects model. Subgroup analyses were performed to evaluate potential heterogeneity. Of 116 titles retrieved based on the standardized search strategy, only 19 articles fulfilled the inclusion criteria for quantitative analysis. The pooled estimates were separately kept as underestimation (n=7) and overestimation (n=12) of chronological age groups for both genders according to primary studies. On absolute values, females (underestimated by 0.13; 95% CI: 0.09-0.18 and overestimated by 0.27; 95% CI: 0.17-0.36) exhibited better accuracy than males (underestimated by 0.28; 95% CI: 0.14-0.42 and overestimated by 0.33; 95% CI: 0.22-0.44). For comparison purposes, the overall pooled estimate overestimated the age by 0.10 (95% CI: -0.06 to 0.26) and 0.09 (95% CI: -0.09 to 0.19) for males and females, respectively. There was no significant difference between the young and older child in subgroup analysis using omnibus test. The mean age between different regions exhibited no statistically significant. The use of Willems method is appropriate to estimate age in children considering its accuracy among different populations, investigators and age groups.