| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 83.28 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Este projecto teve como objectivo validar, numa população portuguesa, a Queiloscopia como método forense de identificação humana na fase reconstrutiva, na estimativa médico-legal do sexo do indivíduo. A amostra populacional foi composta por 202 indivíduos vivos, 109 indivíduos do sexo feminino e 93 do sexo masculino. Classificou-se o padrão labial para cada indivíduo, recorrendo à divisão labial por dois métodos (1-por quadrantes e 2-por porção central e laterais), tendo por base a Classificação de Suzuki e Tsuchihashi. Procedeu-se à análise estatística dos dados obtidos recorrendo-se a tabelas de contingências e o seu coeficiente e regressão logística binária. O nível de significância foi definido em 5%. Os resultados mostraram que, estatisticamente, não se verificaram diferenças significativas no padrão labial entre géneros. A regressão logística binária também não apresentou valores significativos, demonstrando que não estamos na presença de um método a utilizar. Quando comparados os métodos 1 e 2, verificou-se um coeficiente de determinação de Nagelkerke superior para a metodologia por quadrantes, tendo essa um maior poder discriminativo. Contudo, a utilização conjunta dos dois métodos na mesma regressão logística apresenta um resultado mais discriminativo. Conclui-se que, na população analisada, embora haja dimorfismo sexual entre os sulcos labiais, eles não são significativamente diferentes, propondo que o género não afecta a classificação. Pela regressão logística há diferenças estatisticamente significativas entre o parâmetro de identificação sexo e o sexo estimado pelo sistema de classificação de Kazuo Suzuki e Yasuo Tsuchihashi. No futuro, sugere-se a continuação de estudos na área da Queiloscopia com amostras maiores para verificar o contributo do padrão labial no dimorfismo sexual, de forma a ampliar a evidência científica deste método qualitativo e aprimorar a informação disponível para as várias populações.
The aim of this project was to validate, on a Portuguese population, the Cheiloscopy as a forensic method for human identification on the reconstructive phase, for the medico-legal estimation of the individual’s sex. 202 living individuals, 109 women and 93 males composed the test sample. To classify the lip pattern for each individual was applied lip division using two methods (1- quadrants and 2.by central part and lateral portions), based on Suzuki and Tsuchihashi classification. The statistical analysis was done using tables of contingency and the contingent coefficient and binary logistic regression The level of significance used was 5%. The results showed that, statistically, there were no significant differences between lip pattern and sex. The binary logistic regression didn’t show significant values as well, showing that it’s not a method to use. If we compare the method 1 and the method 2, there is a Nagelkerke R-square higher for the method 1, having a bigger discriminative action. However, using both methods together in the same logistic regression has a better result. We conclude that, in the evaluated sample, despite the sexual dimorphism on the labial grooves, they are not significantly different, proposing that sex doesn’t affect the classification. By the logistic regression, there are statistical differences between the sex identification parameter and the estimated sex by Kazuo Suzuki and Yasuo Tsuchihashi classification. In the future, we suggest to continue with the cheiloscopy studies with bigger test samples to verify the contribution of lip pattern on the sexual dimorfism. This way, we can expand the cientific evidence of this qualitative method and improve the information available to the various populations.
The aim of this project was to validate, on a Portuguese population, the Cheiloscopy as a forensic method for human identification on the reconstructive phase, for the medico-legal estimation of the individual’s sex. 202 living individuals, 109 women and 93 males composed the test sample. To classify the lip pattern for each individual was applied lip division using two methods (1- quadrants and 2.by central part and lateral portions), based on Suzuki and Tsuchihashi classification. The statistical analysis was done using tables of contingency and the contingent coefficient and binary logistic regression The level of significance used was 5%. The results showed that, statistically, there were no significant differences between lip pattern and sex. The binary logistic regression didn’t show significant values as well, showing that it’s not a method to use. If we compare the method 1 and the method 2, there is a Nagelkerke R-square higher for the method 1, having a bigger discriminative action. However, using both methods together in the same logistic regression has a better result. We conclude that, in the evaluated sample, despite the sexual dimorphism on the labial grooves, they are not significantly different, proposing that sex doesn’t affect the classification. By the logistic regression, there are statistical differences between the sex identification parameter and the estimated sex by Kazuo Suzuki and Yasuo Tsuchihashi classification. In the future, we suggest to continue with the cheiloscopy studies with bigger test samples to verify the contribution of lip pattern on the sexual dimorfism. This way, we can expand the cientific evidence of this qualitative method and improve the information available to the various populations.
Description
Tese de mestrado, Medicina dentária, Universidade de Lisboa, Faculdade de Medicina Dentária, 2015
Keywords
Antropologia forense Identificação forense Características sexuais Medicina legal Teses de mestrado - 2015