Benedicto EN. Validation of three methods for age estimation of Brazilian children and adolescents [thesis]. São Paulo: Universidade de São Paulo, Faculdade de Odontologia; 2017. Versão Corrigida. Age estimation is an important resource in Forensic Dentistry. Among several ways to obtain the age estimation, researchers may do it by analyzing the mineralization stage of permanent teeth. This type of analysis allows that a simple panoramic radiograph to provide relevant data to identify a deceased individual, for example. This study proposes the validation, accuracy analysis and construction of formulas based on the methods Liliequist and Lundberg with the adaptation of Hägg and Matsson (LLH), Haavikko (HKK) and Mornstad et al. (MSW) for the estimation of age by the stage of mineralization of permanent teeth in Brazilian children and adolescents. To achieve the results, panoramic radiographs of Brazilians were selected in a sample containing 1009 radiographs (387 males and 622 females) aged 8-15.99 years. Of these radiographs, it was selected only those that fulfilled all the requirements of the inclusion and exclusion criteria. After applying the methods, the results were inserted and worked in the Excel Program, Statistical Package STATA 13.0. (StataCorp LP, College Station, Texas, USA) and MedCalc (Medcalc® Software, Mariakerke, Belgium) according to the needs of the study. The samples were divided without splitting sex, with splitting sex and with splitting sex and age. The comparison of methods was performed by the mean differences between dental age and chronological age (DA-CA). Accuracy was analyzed by Absolute Difference (AD). The Skewness e Kurtosis normality test was employed and the Student t test, or the Wilcoxon test, for paired samples was submitted depending on the results. The results of comparisons informed, without any distinction as to sex, the best accuracy was obtained by LLH (AD = 0.97 and DA-CA =-0.58), HKK (AD = 1.42 and DA-CA =-1.35) and MSW (AD = 2.48 and DA-CA = 0.78). With distinction to sex the values for male were: LLH (AD = 0.91 and DA-CA =-0.45), HKK (AD = 1.80 and DA-CA =-1.75) and MSW (AD = 2.74 and DA-CA = 1.17); and for female: LLH (AD = 1.01 and DA-CA =-0.67), HKK (AD = 1.17 and DA-CA =-1.09) and MSW (AD = 2.31 and DA-CA = 0.53). In the comparison with distinction between sex and age, there was a predominance of the LLH technique, followed by HKK and MSW, respectively. The best multiple regression equations (LLH) applied in the same sample presented mean AD = 0.72 and DA-CA =-0.01. At the end, it was possible to conclude that it is possible to validate the methods presented in the Brazilian population and the method that most approached the Brazilian population was the Liliequist and Lundberg method with the adaptation of Hägg and Matsson.