Dental Age Estimation in East Asian Population with Least Squares Regression

The purpose of the study is to derive a machine learning method of estimating overall dental maturity or dental age to achieve higher accuracy than the former methods. We select 1697 orthopantomograms of 877 boys and 820 girls from Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine, of which 1695 (875 boys and 820 girls) are used for estimating the dental age with the Demirjian’s method, Willem’s method and our method. Analysis of Variance (ANOVA) of randomized block design is performed to search for statistically significant differences between the chronological age and the dental age of three methods. Root-mean-square error (RMSE) is also performed in order to analyze the errors of the three methods. The adapted method is validated and results in more accurate dental age estimations in this population compared with Demirjian’s method and Willem’s method.