• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.23.2021.tde-26102021-114604
Document
Author
Full name
Caroline Paula Alves
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2021
Supervisor
Committee
Biazevic, Maria Gabriela Haye (President)
Azevedo, Alana de Cassia Silva
Deitos, Alexandre Raphael
Silva, Ricardo Henrique Alves da
Title in Portuguese
Utilização do seio frontal para avaliar dimorfismo sexual em amostra brasileira
Keywords in Portuguese
Antropologia forense
Odontologia forense
Seio frontal
Abstract in Portuguese
Os seios frontais são comumente usados na estimativa sexual por serem únicos para cada indivíduo, devido à variação considerável em tamanho, forma e número. Estudos anteriores apresentaram média acurácia mensurando área, altura e profundidade do seio frontal, porém, autores têm associado tais medidas com o índice do seio frontal, obtendo melhores resultados. Sendo assim, o objetivo do presente estudo foi avaliar dimorfismo sexual do seio frontal em adultos brasileiros. A amostra foi composta por 255 telerradiografias laterais, de indivíduos entre 20 e 40 anos, sendo 132 do sexo feminino e 123 do sexo masculino. Pautado na metodologia de Luo et al., 2018, foram mensuradas a área (S), a altura e a profundidade máximas do seio frontal (AB e EF, respectivamente) por meio de programa de desenho assistido por computador; e a razão entre AB e EF foi tida como índice do seio frontal (R). Aplicou-se, então, a função discriminante desenvolvida pelos autores para avaliar o dimorfismo sexual na população brasileira. Foram realizados os testes de Shapiro-Wilk para avaliar a normalidade das variáveis e os testes t de Student e Mann-Whitney para verificar se houve diferença entre os sexos. A confiabilidade do método foi calculada através do coeficiente de correlação intraclasse (ICC) e o coeficiente de variação (CV), além do diagrama de Bland-Altman. Uma nova fórmula discriminante foi calculada com os dados do estudo e técnicas de machine learning foram utilizadas para melhorar a predição do sexo. O ICC foi acima de 0,90 para todas as variáveis e o CV abaixo de 5%. Na análise de Bland-Altman as medidas estavam dentro dos valores de confiança. Adesão à curva de normalidade para o sexo masculino foi observado apenas na variável EF. As variáveis apresentaram diferenças significantes em relação ao sexo e, com exceção de R, onde a média masculina foi 2,00 e a feminina, 2,40, todas as médias foram maiores para os homens. A fórmula original do estudo teve baixa acurácia, com nível de acerto de apenas 8,33% para o sexo feminino. No entanto, a fórmula calculada para brasileiros apresentou uma acurácia de 70,20%; das técnicas de machine learning, apenas a rede neural apresentou valor superior ao já obtido, de 73,30%. Concluindo, a nova fórmula apresentou uma acurácia de 70,20%, podendo ser aplicada como método auxiliar na avaliação do dimorfismo sexual do seio frontal em adultos brasileiros.
Title in English
Use of the frontal sinus to evaluate sexual dimorphism in a Brazilian sample
Keywords in English
Forensic anthopology
Forensic odontology
Frontal sinus
Abstract in English
The frontal sinuses are commonly used in sexual estimation because they are unique for each individual, due to the considerable variation in size, shape and number. Previous studies have shown average accuracy measuring frontal sinus area, height and width; however, authors have associated such measurements with the frontal sinus index, obtaining better results. Therefore, the aim of the present study was to evaluate sexual dimorphism of the frontal sinus in Brazilian adults. The sample consisted of 255 lateral cephalometric radiographs of subjects between 20 and 40 years of age, 132 females and 123 males. Based on the methodology of Luo et al., 2018, the area (S) and the maximum frontal sinus height and width (AB and EF, respectively) were measured using computer-aided design software; and the ratio between AB and EF was taken as the frontal sinus index (R). The discriminant function developed by the authors was then applied to evaluate sexual dimorphism in the Brazilian population. The Shapiro-Wilk test was performed to assess the normality of the variables, and the Students t-test and the Mann-Whitney test were used to check for gender diferences. The reliability of the method was calculated using the intraclass correlation coefficient (ICC) and the coefficient of the variation (CV), and the Bland-Altman diagram. A new discriminant formula was calculated with the study data and machine learning techniques were used to improve the prediction of sex. The ICC was above 0.90 for all variables and the CV below 5%. In the Bland-Altman analysis the measurements were within confidence values. Adherence to the normality curve for males was observed only in the EF variable. The variables showed significant differences in relation to gender, and with the exception of R, where the male mean was 2.00 and the female mean was 2.40, all the means were higher for males. The original formula of the study had low accuracy, with a level of accuracy of only 8.33% for females. However, the formula calculated for Brazilians presented an accuracy of 70.20%; of the machine learning techniques, only the neural network presented a higher values than the one already obtained, of 73.30%. In conclusion, the new formula showed an accuracy of 70.20% and can be applied as an auxiliary method in the assessment of frontal sinus sexual dimorphism in Brazilian adults.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2022-01-17
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.