Abstract
Background
Accurate age estimation is essential in forensic and clinical contexts, particularly where birth records are unavailable. Dental development assessed through radiographic calcification is considered a reliable indicator. However, atlas-based methods have demonstrated variable accuracy across populations. This study aimed to evaluate the accuracy and agreement of the Schour and Massler chart and the London Atlas, and to develop a population-specific dental development scheme for a Pakistani cohort.Methods
A cross-sectional forensic validation study was conducted using 700 digital panoramic radiographs of individuals aged 4.0-17.9 years, stratified by age and sex. Dental age was estimated using both methods and compared with the chronological age. The analyses included paired t-tests, correlation, intraclass correlation coefficient, Bland-Altman analysis, and linear regression. The predictive performance was assessed using the mean absolute error and proportions within ± 1 and ± 2 years.Results
Both methods demonstrated strong correlations (r > 0.90) and high ICCs (> 0.85). The London Atlas showed lower mean bias (~ 0.2-0.4 years), narrower limits of agreement (± 1.5-2.0 years), and higher predictive accuracy (overall R² = 0.997; sex-stratified subgroup range: R² = 0.92-0.95) than the Schour and Massler method (overall R² = 0.982; sex-stratified subgroup range: R² = 0.85-0.90). Accuracy within ± 1 year was higher for the London Atlas (> 75%) compared with Schour and Massler (Conclusions
The London Atlas demonstrated superior agreement and predictive performance; however, population-specific variations limit its universal applicability. The newly developed Pak Atlas provides a more accurate reference for forensic age estimation in the Pakistani population, with direct relevance to medico-legal age assessment.