Evaluating the accuracy of Kvaal’s method for age estimation in a selected Iranian population

Abstract

Objectives: This study aimed to evaluate the efficacy of Kvaal et al.’s method (1995) for dental age estimation of adult individuals in a sample of Iranian population. For this purpose, the accuracy of the regression formulas and M and W-L variables were studied.Materials and methods: Pulp and tooth lengths and widths were measured on 100 digital panoramic radiographs. Then Kvaal’s M and W-L variables were computed and substituted in Kvaal’s formulas to evaluate their accuracies. The efficacy of these two variables in our population was assessed using a k-fold cross-validation technique for regression analysis. Principle component analysis was also performed to develop population-specific dental variables.Results: Applying Kvaal’s regression formulas on multiple teeth in different jaws resulted in highly insensible estimations. In contrast, developing regression formulas based on Kvaal’s M and W-L variables yielded reasonable and sensible estimates especially for younger individuals; standard error of estimate (SEE) values ranged from 6.36-6.80 years. The models based on multiple teeth in different jaws performed similar.Conclusions: For young adults, the M and W-L variables anticipated accuracy rates lower than those of Kvaal’s reference study which were within an acceptable threshold for forensic application (SEE<10 years). Therefore, the two variables and not formulas are proper measurements for forensic age estimation in Iranian young adults. However, underestimation was predominant for the middle age and old age participants.