Validation of London Atlas for forensic age estimation in Koreans by comparing with Lee's and Willems' methods

This cross-sectional study aimed to evaluate the applicability of the London Atlas for age estimation in the Korean population by comparing with Lee's and Willems' methods. Dental ages of 475 orthopantomographs from Korean individuals aged 4–15 years (mean 10.32 ± 3.31 years) were estimated using the London Atlas, Lee's and Willems' methods. Correlation between dental and chronological age was determined using Pearson's correlation coefficient, and the statistical difference between dental and chronological age was analyzed using a paired t-test. The bias (mean differences), mean absolute error, and root mean square error between dental and chronological age, stratified by age groups and estimation methods, were calculated. Differences in bias and absolute error between sexes were scrutinized using an independent-samples t-test. Age estimates of the three tested methods were combined and compared to those of each individual method. The London Atlas and Willems methods resulted in overestimations, whereas the Lee method led to an underestimation on the entire sample. The overall accuracy was observed in the order of Lee's method, the London Atlas, and Willems' method. The London Atlas demonstrated superior consistency of estimation performance across age groups and no significant differences in estimation performance between sexes. The combination of estimates from the London Atlas and Lee's method resulted in an enhancement in bias and accuracy. We conclude that the London Atlas, due to its bias and accuracy comparable to Lee's and Willems' methods, is applicable for forensic practice in the Korean population.