Annals of Human Biology (May 2022)
Dental age estimation of Malaysian Indian children and adolescents: applicability of Chaillet and Demirjian’s modified method using artificial neural network
Abstract
Background Recognising the importance of dental age (DA) estimation in forensic investigations, a variety of methods abound in the literature due to population-specific attributes. A reference eight-tooth method developed by Chaillet and Demirjian estimated the DA of children and adolescents. Aim This study aims to investigate the applicability of Chaillet and Demirjian’s method among Malaysian Indians aged 5.00–17.99 years. Subjects and methods Dental panoramic tomographs of Malaysian Indians aged 5.00–17.99 years were statistically analysed using paired t-test and artificial neural networks multilayer perceptron (ANN-MLP). Results A total of 1015 dental panoramic tomographs were analysed. Paired t-test analysis against the reference dental maturity scores revealed underestimation of DA in boys of 1.68 years and girls of 2.56 years indicating inaccurate age estimation. A population-specific prediction model with a new set of dental maturity scores was established on Chaillet and Demirjian’s scores using ANN-MLP. The new dental maturity scores showed accurate estimation of DA with differences between CA and DA being 12 and 25 days for boys and girls, respectively. Furthermore, a new DA prediction formula was developed using regression analysis following the establishment of new dental scores based on ANN-MLP. Conclusion A novel Malaysian Indian-specific prediction model that demonstrated accurate DA estimation was established.
Keywords