Esta tese é composta por três capítulos. O primeiro teve o objetivo de avaliar a acurácia e a semelhança da reconstrução facial forense computadorizada (RFFC) tridimensional realizada com softwares livres. As RFFC foram realizadas no programa Blender® a partir de tomografias do crânio, utilizando templates do MakeHuman®. A avaliação da acurácia foi feita no CloudCompare® que comparou a RFFC com a pele na tomografia, enquanto a avaliação da semelhança foi realizada no Picasa® utilizando fotografias dos participantes. Os resultados mostraram que do total de pontos que formam cada reconstrução, 63.20% a 73.67% apresentaram uma distância de -2,5 <= x <= 2,5mm entre a RFFC e a superfície da pele, enquanto a distância média variou entre -1,66 a 0,33mm. Duas das quatro reconstruções foram reconhecidas objetivamente pelo Picasa®. As RFFC realizadas utilizando esses softwares apresentam plausíveis níveis de acurácia e semelhança, portanto indicam valor para uso no campo forense. Os outros dois capítulos tiveram como objetivo avaliar a estabilidade métrica facial do mesmo indivíduo por meio da análise de fotografias tomadas em um intervalo de tempo de cinco anos. Trata-se de um estudo longitudinal realizado com fotografias frontais padronizadas de 666 indivíduos adultos divididos por faixa etária e sexo. Com o programa SAFF 2D foram marcados 32 pontos, cujas coordenadas foram utilizadas para calcular 40 medidas, sendo 20 horizontais e 20 verticais. Cada uma dessas medidas foi dividida pelo diâmetro da íris e assim foram obtidas razões iridianas. Os resultados mostraram que a maioria das razões não sofreu variação estatisticamente significante. As razões que tiveram maior variação foram aquelas da região do nariz e da boca. Quando se compara as faixas etárias entre si observa-se que a grande maioria das razões é diferente, mostrando a influência da idade nas dimensões faciais. Quando se compara a estabilidade dentro mesmo sexo observa-se que houve razões que diminuíram e outras aumentaram tanto no sexo feminino quanto no sexo masculino, enquanto outras variaram apenas em um dos sexos. Quando se compara a variação entre os sexos observa-se que a maioria das razões é diferente, mostrando o dimorfismo sexual das medidas faciais. A face passa por alterações métricas ao longo da vida em todas as faixas etárias, principalmente na região do nariz e boca, com maiores diferenças após os 60 anos. Além disso, algumas medidas faciais são mais influenciadas pelo sexo do que outras. Entretanto, a maioria das medidas levantadas se mantem relativamente estáveis dentro de um período de 5 anos tanto em relação ao sexo quanto a idade. / This thesis contains three chapters. The aim of the first chapter was to evaluate the accuracy and recognition level of three-dimensional (3D) computerized forensic craniofacial reconstruction (CCFR) performed in a blind test on open-source software using computed tomography data from live subjects. The CCFRs were completed using Blender® with 3D models obtained from the computed tomography data and templates from the MakeHuman® program. The evaluation of accuracy was carried out in CloudCompare®, by geometric comparison of the CCFR to the subject 3D face model (obtained from the CT data). A recognition level was performed using the Picasa® with a frontal standardized photography. The results were presented from all the points that form the CCFR model, with an average for each comparison between 63.20% and 73.67% with a distance -2.5 <= x <= 2.5 mm from the skin surface and the average distances were 1.66 to 0.33 mm. Two of the four CCFRs were correctly matched by the Picasa® tool. Free software programs are capable of producing 3D CCFRs with plausible levels of accuracy and recognition and therefore indicate their value for use in forensic applications. The other two chapters study the facial comparison and aimed to evaluate the facial metrical stability of an individual through photographs taken in a time interval of five years. It is a longitudinal study composed of standard frontal photographs of 666 adults divided by sex and age groups. By using the SAFF 2D® software, 32 landmarks were positioned, whose coordinates were used to calculate 40 measurements, 20 horizontal and 20 vertical. Each of these measurements was divided by iris diameter and thus iridian ratios were obtained. The results showed that most of the ratios did not suffer statistically significant variations. The ratios that had the greatest variation in the different age groups were those of the nose and mouth regions. When comparing the age groups with each other it is observed that the great majority of the reasons are different, showing the influence of age on the facial dimensions. When comparing stability with respect to sex, it was observed that there were ratios that decreased and others that increased in both sexes, while other ratios varied only in females or in males. When the sexes were compared, it was observed that the majority of the ratios were different, showing sexual dimorphism of the facial measures. The face undergoes metrical alterations throughout the life, mainly in the region of the nose and mouth, with the greatest differences seen in those who are aged 60 years and older. In addition, some facial measures are more influenced by sex than others. However, most of the measures raised have remained relatively stable within a period of five years in both sex and age groups.

L’estimation du délai post mortem est une étape fondamentale en anthropologie médico-légale. À ce jour, peu de méthodes précises et fiables existent. Les objectifs de notre travail étaient d’étudier le tissu osseux et son évolution dans les années et siècles suivant le décès en développant la Résonance Magnétique Nucléaire (RMN) des solides du carbone13C et du proton1H. Nous avons analysé une centaine d’os humains et animaux pour lesquels nous connaissions l’âge au décès, le sexe, la date de décès et les conditions de conservation. Nous avons caractérisé les os au niveau moléculaire en identifiant le collagène, les lipides et l’hydroxyapatite constitutifs du tissu osseux. Nous avons développé une méthode RMN permettant de distinguer des altérations de certains échantillons attestant de la présence d’adipocire au sein du tissu osseux, ou des dégradations sur des échantillons très anciens. L’étude de l’âge au décès et du sexe des sujets n’a pas mis en évidence une grande influence de ces facteurs sur les données RMN même si, pour des délais post mortem de 0 ou 1 an, les sujets féminins présentent quantitativement plus de lipides que les sujets masculins. L’analyse des conditions de conservation des individus montre un développement plus important d’adipocire pour les os laissés à l’air libre comparés aux os enterrés. Enfin, nous rapportons une décroissance quantitative du collagène et des lipides présents au sein du tissu osseux lorsque l’intervalle post mortem augmente. Cette décroissance est beaucoup plus rapide pour les lipides (quelques années) que pour le collagène (plusieurs millénaires) alors que l’hydroxyapatite présente une relative stabilité dans les premiers siècles suivant le décès. / The post mortem interval estimation is a fundamental step in forensic anthropology and up to now there are little accurate and reliable methods to do so. The objectives of our study were to investigate the bone composition and its evolution over years and centuries following the death by developing carbon 13C and proton 1H solid-state nuclear magnetic resonance (NMR). We analyzed about one hundred human and animal bones for which the age at death, sex, date of death and the storage conditions were known. Bones were characterized at the molecular level by identification of collagen, lipids and hydroxyapatite embedded in the bone matrix. We have designed a NMR-based method that allows determining alterations on some samples, evidencing the presence of adipocere (bone wax) within the bone, or finding bone tissue deterioration on some very old samples. Subject age at death and sex did not reveal significant changes on NMR data, except for post mortem interval ranging between 0 to 1 year, where female subjects had quantitatively more lipids in their bones than males. Storage conditions may promote a greater development of adipocere especially for bones left in the open air compared to those buried. Finally, we report a quantitative decrease of collagen and lipids present in the bone tissue when the post mortem interval increases. This decrease is much faster for lipids than for collagen where as the hydroxyapatite has a relative stability in the first centuries after the death. Decreases occur with very different time constants, ranging from years to millennia.

Esta tese é composta por três capítulos. O primeiro teve o objetivo de avaliar a acurácia e a semelhança da reconstrução facial forense computadorizada (RFFC) tridimensional realizada com softwares livres. As RFFC foram realizadas no programa Blender® a partir de tomografias do crânio, utilizando templates do MakeHuman®. A avaliação da acurácia foi feita no CloudCompare® que comparou a RFFC com a pele na tomografia, enquanto a avaliação da semelhança foi realizada no Picasa® utilizando fotografias dos participantes. Os resultados mostraram que do total de pontos que formam cada reconstrução, 63.20% a 73.67% apresentaram uma distância de -2,5 <= x <= 2,5mm entre a RFFC e a superfície da pele, enquanto a distância média variou entre -1,66 a 0,33mm. Duas das quatro reconstruções foram reconhecidas objetivamente pelo Picasa®. As RFFC realizadas utilizando esses softwares apresentam plausíveis níveis de acurácia e semelhança, portanto indicam valor para uso no campo forense. Os outros dois capítulos tiveram como objetivo avaliar a estabilidade métrica facial do mesmo indivíduo por meio da análise de fotografias tomadas em um intervalo de tempo de cinco anos. Trata-se de um estudo longitudinal realizado com fotografias frontais padronizadas de 666 indivíduos adultos divididos por faixa etária e sexo. Com o programa SAFF 2D foram marcados 32 pontos, cujas coordenadas foram utilizadas para calcular 40 medidas, sendo 20 horizontais e 20 verticais. Cada uma dessas medidas foi dividida pelo diâmetro da íris e assim foram obtidas razões iridianas. Os resultados mostraram que a maioria das razões não sofreu variação estatisticamente significante. As razões que tiveram maior variação foram aquelas da região do nariz e da boca. Quando se compara as faixas etárias entre si observa-se que a grande maioria das razões é diferente, mostrando a influência da idade nas dimensões faciais. Quando se compara a estabilidade dentro mesmo sexo observa-se que houve razões que diminuíram e outras aumentaram tanto no sexo feminino quanto no sexo masculino, enquanto outras variaram apenas em um dos sexos. Quando se compara a variação entre os sexos observa-se que a maioria das razões é diferente, mostrando o dimorfismo sexual das medidas faciais. A face passa por alterações métricas ao longo da vida em todas as faixas etárias, principalmente na região do nariz e boca, com maiores diferenças após os 60 anos. Além disso, algumas medidas faciais são mais influenciadas pelo sexo do que outras. Entretanto, a maioria das medidas levantadas se mantem relativamente estáveis dentro de um período de 5 anos tanto em relação ao sexo quanto a idade. / This thesis contains three chapters. The aim of the first chapter was to evaluate the accuracy and recognition level of three-dimensional (3D) computerized forensic craniofacial reconstruction (CCFR) performed in a blind test on open-source software using computed tomography data from live subjects. The CCFRs were completed using Blender® with 3D models obtained from the computed tomography data and templates from the MakeHuman® program. The evaluation of accuracy was carried out in CloudCompare®, by geometric comparison of the CCFR to the subject 3D face model (obtained from the CT data). A recognition level was performed using the Picasa® with a frontal standardized photography. The results were presented from all the points that form the CCFR model, with an average for each comparison between 63.20% and 73.67% with a distance -2.5 <= x <= 2.5 mm from the skin surface and the average distances were 1.66 to 0.33 mm. Two of the four CCFRs were correctly matched by the Picasa® tool. Free software programs are capable of producing 3D CCFRs with plausible levels of accuracy and recognition and therefore indicate their value for use in forensic applications. The other two chapters study the facial comparison and aimed to evaluate the facial metrical stability of an individual through photographs taken in a time interval of five years. It is a longitudinal study composed of standard frontal photographs of 666 adults divided by sex and age groups. By using the SAFF 2D® software, 32 landmarks were positioned, whose coordinates were used to calculate 40 measurements, 20 horizontal and 20 vertical. Each of these measurements was divided by iris diameter and thus iridian ratios were obtained. The results showed that most of the ratios did not suffer statistically significant variations. The ratios that had the greatest variation in the different age groups were those of the nose and mouth regions. When comparing the age groups with each other it is observed that the great majority of the reasons are different, showing the influence of age on the facial dimensions. When comparing stability with respect to sex, it was observed that there were ratios that decreased and others that increased in both sexes, while other ratios varied only in females or in males. When the sexes were compared, it was observed that the majority of the ratios were different, showing sexual dimorphism of the facial measures. The face undergoes metrical alterations throughout the life, mainly in the region of the nose and mouth, with the greatest differences seen in those who are aged 60 years and older. In addition, some facial measures are more influenced by sex than others. However, most of the measures raised have remained relatively stable within a period of five years in both sex and age groups.

Para a execução da etapa inicial da identificação médico-legal de restos humanos (antropometria e exame dos arcos dentários), faz-se necessária uma limpeza prévia da ossada, para a remoção de tecidos moles putrefeitos. Os casos não identificados por esses métodos tradicionais, poderão ser submetidos ao exame de DNA. No entanto, apesar do grande avanço da biologia molecular, utilizando a amplificação de DNA pela PCR, algumas limitações que afetam a habilidade de se obter DNA em restos humanos, permanecem. Dessa forma, o objetivo deste trabalho foi fornecer subsídios morfológicos para os analistas forenses, com ênfase na prática médico-legal, visando uma utilização mais eficiente do DNA obtido de osso compacto de restos humanos em decomposição ou já esqueletizados, sem tecidos moles aderidos. Foi realizado o estudo da microestrutura do tecido ósseo compacto femoral, de restos humanos em decomposição, ainda com tecidos moles, que foram limpos pela fervura em água (n = 7) e ossadas já esqueletizadas pela decomposição natural, que não foram fervidas (n = 8). Destes, seis ossadas foram provenientes de cemitério público regular, após 3 anos de inumação, 1 ossada proveniente da região amazônica, e 1 ossada de origem desconhecida. Estas duas ultimas, apresentado-se porosas ou quebradiças. As análises morfológicas de cortes histológicos foram coradas com hematoxilina e eosina e o DNA amplificado pela PCR para os loci CSF1PO, TPOX, TH01, F13A0, FESFPS, vWA, D16S539, D7S820, D13S317 e amelogenina. Os resultados da análise desses dois grupos foram comparados com os de cadáveres frescos (n = 5) do Serviço de Verificação de Óbitos da Capital. A fervura dos ossos, do modo como é realizada no Instituto Médico Legal de São Paulo, pode aumentar a eosinofilia da matriz óssea e, em alguns casos, pode promover a desagregação dos ósteons. Tal procedimento pode remover células, mas pode também remover possíveis inibidores da PCR, favorecendo a análise do DNA obtido destas amostras. O fator limitante para a obtenção e análise de DNA, em amostras de ossos limpos por fervura, é a quantidade exígua de células. Ossos não submetidos à fervura, após inumação por três anos ou há mais tempo em contato com a terra, podem apresentar alterações da microestrutura. No entanto, a presença de hemácias preservadas e núcleos de osteócitos nestas amostras, indica melhor preservação de células em relação às amostras de ossos fervidos. O fator limitante para a análise de DNA nestas amostras é a presença sugestiva de inibidores da reação de amplificação pela PCR. Restos humanos, sem tecidos moles, macroscópicamente não preservados (porosos e quebradiços), e não submetidos à fervura, apresentam alterações de perda de matriz mineralizada; no entanto, nestas amostras ainda é possível encontrar células preservadas. Os resultados obtidos no neste trabalho permitem traçar algumas estratégias para uma melhor utilização nos protocolos de extração e análise do DNA em osso compacto de restos humanos. / To the first essential step to forensic identification of human remains (anthropological study of race, sex, age, etc) it is necessary a previous cleaning of the bones, to remove decomposing soft tissues. Medico-legal inconclusive or non identified cases, by using these traditional methods, could be subjected to DNA analysis. However, in spite of advances in human identification techniques, specially by PCR amplified DNA, some limitations that affect the ability to obtain DNA in human remains still persist. Therefore, the aim of this study was to provide additional support from morphological analysis, to help forensic analysts personnel to utilise more efficiently the DNA, extracted from compact bones of human remains in decomposition or already skeletonized corpse, it means without soft tissues, with special emphasis in the legal-medicine practice. Femoral compact bones were obtained from: 7 human remains found on the ground, in different degree of decomposition which were cleaned by boiling to remove soft tissues; also studied were collections of bones from 8 corpses having undergone natural decomposition: 6 human remains exhumed after 3 years from a common public cemetery in São Paulo City; 1 case from amazon region and 1 case with no information, both cases remained from long time (more than 3 years) in contact with soil. All eight cases, were not boiled as no soft tissue were adhered. As a control, five cadavers 12 to 16 hours post mortem were also used. The compact bones histological sections were stained by haematoxilin and eosin and the loci CSF1PO, TPOX, TH01, F13A01,FESFPS, vWA, D16S539, D7S820, D13S317 and amelogenin were amplified by PCR.The procedure for boiling the human remains utilised in the Legal Medicine Institute of São Paulo would have increased the eosinophily of bone matrix and, in some cases, promoted the desaggregation of the osteons. In addition these procedures would have removed the cells, but in some cases would have removed possible inhibitors of the PCR, favouring in this way the analysis of DNA obtained from these samples. The limiting factor to obtain successful analysis in bones submitted to boiling seem to be the low quantity of nuclei present in these samples. For the other hand, in bones not cleaned by boiling, the presence of preserved red cells and oscteocyte nuclei inside the lacunae indicates better preservation of cells in relation to those bones cleaned by boiling. The limiting factor to obtain successful DNA analysis in bones exhumed or in contact of soil, is the suggestive presence of inhibitors of PCR. Porous and brittle bones from human remains, without soft tissues that are not processed by boiling, present alterations through loss of mineralised matrix, although it is still possible to found preserved cells in these samples. The results presented in this work clarify concerns about viability of DNA for identification analysis. They also help to establish better strategies for optimisation of DNA extraction and analysis in compact bones of human remains.

A preocupação com a identificação, que é o processo pelo qual se determina a identidade, é bastante antiga. Atualmente, as relações sociais ou exigências civis, penais, administrativas e comerciais necessitam de sua comprovação. A identificação de pessoas mortas é fundamental, não somente para aplacar as necessidades emocionais de seus amigos e familiares, como também para que providências legais relativas ao óbito possam ser tomadas. Infelizmente, amiúde ocorrem situações em que corpos chegam aos Institutos Médico-Legais em estado de putrefação ou esqueletização, e não são identificados. Em tais situações, análises antropométricas para estimar, por exemplo, idade, gênero e estatura, são de grande valia. Nestes casos, a reconstrução facial forense será muito importante, pois pode possibilitar o reconhecimento e, por conseguinte, aumentar consideravelmente as chances de identificação. A reconstrução facial forense tridimensional pode ser manual ou digital. A reconstrução facial forense digital tornou-se possível com o advento da Tecnologia da Informação, imaginologia médica e novos softwares de imagem 3D e de reconstrução. Para a realização da reconstrução facial, são necessários dados relativos à espessura dos tecidos moles da face. Não há na literatura registros de trabalhos de reconstrução facial digital realizados com dados de tecidos moles obtidos a partir de amostras constituídas por sujeitos brasileiros. Há duas tabelas de espessura de tecidos moles publicadas para a população brasileira: uma obtida a partir de medidas realizadas em cadáveres frescos (padrão cadáveres frescos), e outra a partir de medidas em exames de ressonância magnética (padrão ressonância magnética). O objetivo do presente trabalho foi realizar três diferentes reconstruções faciais forenses digitais caracterizadas (com cabelo, cílios e sobrancelha) de um sujeito brasileiro (realizadas a partir de um padrão internacional e dois padrões nacionais de espessura de tecidos moles da face), e avaliar as reconstruções faciais forenses digitais comparando-as com fotografias do próprio indivíduo e de outros nove sujeitos. Para isso, foram utilizadas imagens DICOM de uma Tomografia Computadorizada (TC) cedidas por um voluntário que, uma vez convertidas, foram utilizadas para a efetivação das reconstruções faciais digitais. Uma vez realizadas as três reconstruções, as mesmas foram comparadas com fotografias do voluntário que teve a sua face reconstruída e outros nove sujeitos. Trinta examinadores participaram desta tentativa de reconhecimento. O sujeito-alvo, que teve a sua face reconstruída, foi reconhecido por 26,67% dos examinadores na reconstrução realizada com o Padrão nacional de Ressonância Magnética, 23,33% na reconstrução realizada com o Padrão nacional de Cadáveres Frescos e 20,00% na reconstrução realizada com o Padrão Internacional, tendo sido o sujeito mais reconhecido nos dois primeiros padrões. Os reconhecimentos acertados do sujeito-alvo indicam que a reconstrução facial forense digital, realizada com parâmetros empregados neste trabalho, pode ser ferramenta útil para, havendo um ou vários sujeitos reconhecidos, chegar-se a uma identificação positiva. / The concern with the identification, that is the process by which the identity is determined, is quite old. Currently, the social relations or civil, criminal, administrative and commercial requirements need its evidence. The identification of deceased persons is essential not only to assuage the emotional needs of their friends and family, but also to allow legal actions related to death. Unfortunately, situations often occur when bodies arrive at the Medico-Legal Institutes in a state of putrefaction or skeletonization, and are not identified. In such situations, anthropometric analysis to estimate, for example, age, gender and height, are of great value. In these cases, forensic facial reconstruction is very important because it may serve to recognize and therefore increase the chances of identification. The three-dimensional forensic facial reconstruction can be manual or digital. The digital forensic facial reconstruction was made possible with the advent of Information Technology, medical imaging and new 3D image and reconstruction softwares. To perform facial reconstruction, data on the thickness of the soft tissues of the face are necessary. There is no literature records of facial reconstruction works carried out with digital data of soft tissues obtained from samples of Brazilian subjects. There are two tables of thickness of soft tissue published for the Brazilian population: one obtained from measurements performed in fresh cadavers (fresh cadavers pattern), and another from measurements on magnetic resonance imaging (magnetic resonance pattern). The aim of this study was to perform three different characterized digital forensic facial reconstructions (with hair, eyelashes and eyebrows) of a Brazilian subject (based on an international pattern and two national patterns for soft facial tissue thickness), and evaluate the digital forensic facial reconstructions comparing them to photos of the individual and other nine subjects. We used DICOM images of a computed tomography (CT) donated by a volunteer that, once converted, were used for the realization of the digital facial reconstructions. Once we\'ve performed the three reconstructions, they were compared with photographs of the volunteer who had his face reconstructed and of nine other subjects. Thirty examiners participated in this recognition attempt. The target subject, who had his face reconstructed, was recognized by 26.67% of the examiners in the reconstruction performed with the national Magnetic Resonance Pattern, 23.33% in the reconstruction performed with the national Fresh Cadavers Pattern of and 20.00 % in the reconstruction performed with the International Pattern, and the target-subject was the most recognized subject in the first two patterns. The correct recognitions of the subject indicate that the digital forensic facial reconstruction, carried out with parameters used in this study, may be a useful tool, with one or more subjects recognized to achieve a positive identification.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2018-04-26T12:16:51Z No. of bitstreams: 1 victorsylviosaggioro.pdf: 827982 bytes, checksum: 220854567786f428357cc808968543c7 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2018-04-26T13:07:07Z (GMT) No. of bitstreams: 1 victorsylviosaggioro.pdf: 827982 bytes, checksum: 220854567786f428357cc808968543c7 (MD5) / Made available in DSpace on 2018-04-26T13:07:07Z (GMT). No. of bitstreams: 1 victorsylviosaggioro.pdf: 827982 bytes, checksum: 220854567786f428357cc808968543c7 (MD5) Previous issue date: 2015-12-05 / PROQUALI (UFJF) / Atualmente, o uso de técnicas para a correta identificação de indivíduos vivos ou mortos faz-se necessária para que este, ou sua família, possa usufruir dos direitos que lhe cabem e, assim, para que se possa buscar justiça nas sociedades a que pertencem. Os direitos mais fundamentais foram negados a um grupo significativos de pessoas durante os anos do regime de exceção vividos em nosso país. A negação, sobretudo, de conhecer as verdades dos fatos alcançou as famílias que não puderam enterrar ou desconhecem o paradeiro de seus entes presos e desaparecidos políticos da ditadura militar brasileira. Neste contexto que a antropologia forense figura como importante ferramenta para identificar restos mortais humanos. A dissertação buscou, a partir de pesquisa exploratória realizar um levantamento sobre a forma praticada pela pelos agentes da ditadura para desaparecer com as evidências dos crimes cometidos, mormente os corpos, e a aplicação da antropologia forense na identificação atual destes corpos exumados de valas comuns e covas clandestinas. Relata as dificuldades encontradas pela antropologia forense brasileira e a experiência do grupo argentino de antropólogos forenses. Por fim, reitera a importância da identificação dos mortos para trazer alentos às famílias e permitir eventual responsabilização do Estado. / Nowadays, the use of techniques for the identification of alive or dead is needed for this, or your family can enjoy the rights that fit you and thus, so that they can seek justice in the societies to which they belong. The most fundamental rights were denied to a significant group of people during the years of authoritarian regime lived in our country. Denial, above all, know the truths of the events reached the families who were unable to bury or unaware of the whereabouts of their loved political prisoners and missing the Brazilian military dictatorship. In this context that the forensic anthropology appears as an important tool to identify the human remains. The dissertation sought, from exploratory research to survey the form practiced by the dictatorship of agents to disappear with the evidence of the crimes committed, especially the bodies and the application of forensic anthropology in the current identification of these bodies exhumed from mass graves and clandestine graves. Recounts the difficulties encountered by Brazilian forensic anthropology and experience of the Argentine group of forensic anthropologists. Finally, it reiterates the importance of identifying the dead to bring breaths families and allow potential liability of the state.

Les méthodes osseuses de diagnose sexuelle donnent d’excellents résultats, avec des taux de bonnes prédictions du sexe supérieurs à 90%. Néanmoins, les éléments osseux ne sont pas toujours disponibles et, lorsqu’ils le sont, selon les causes de la mort et/ou les conditions de conservation du corps, ils peuvent avoir été lourdement dégradés. Pour parvenir à déterminer le sexe d’un individu dans ces situations là, des techniques dentaires de diagnose sexuelle sont nécessaires.Contrairement à la détermination du sexe, la problématique de la détermination de l’âge a fait l’objet de nombreuses études et d’autant de propositions de techniques différentes, signe d’une insuffisance de l’ensemble de ces méthodes, que la littérature n’a pas manqué de mettre en exergue.Le premier objectif de ce travail est de proposer une technique dentaire de détermination du sexe simple et précise, de façon à pouvoir répondre à cette problématique anthropologique de façon fiable en l’absence d’autres éléments du squelette.Le second est de développer une méthode d’estimation de l’âge de mise en œuvre facile, rapide et ne nécessitant ni un plateau technique lourd, ni la détérioration du matériel, pour une application aussi bien anthropologique que médico-légale. L’échantillon d’apprentissage était composé de 210 scanners d’individus présentant chacun 4 canines saines. L’échantillon d’application était composé de 55 scanners d’individus, présentant chacun au moins 1 canine saine. Les volumes pulpaire et total de chaque dent ont été modélisés et calculés à l’aide du logiciel Mimics&#63720;. Une régression logistique binaire a été utilisée pour déterminer sept modèles de prédiction du sexe, en fonction du nombre et du type de canines exploitables. La comparaison des aires sous la courbe ROC a mis en évidence une plus grande performance du modèle utilisant les volumes des 4 canines. La méthode des moindres carrés pondérés a été utilisée pour déterminer les équations d’estimation de l’âge pour les sept mêmes modèles. Une plus grande performance du modèle utilisant les volumes des 2 canines maxillaires a été mise en évidence. Toutes ces régressions ont été testées sur l’échantillon d’application pour procéder à leur validation externe.Il s’avère que dans un contexte scientifique historique où les canines mandibulaires constituaient l’outil dentaire de référence dans la diagnose sexuelle, la modélisation tridimensionnelle souligne l’intérêt potentiel de leurs homologues maxillaires. Par ailleurs, il semblerait que l’évaluation de la réduction pulpaire physiologique, imputable à l’apposition naturelle de dentine secondaire sur les canines uniquement, soit un critère performant dans la détermination de l’âge.A défaut d’être parfaits, les modèles proposés à l’issue cette étude permettent donc toutefois de disposer d’éléments fiables, qui ont toute leur place dans l’indispensable faisceau de preuves, nécessaire à la détermination de l’âge et du sexe dans le cadre d’une identification estimative. / The bone methods of sexual diagnosis give excellent results, with rates of good predictions over 90%. Nonetheless, bones are not always available and, when they are, according to the causes of death and/or storage conditions of the body, they may have been severely degraded. To be able to determine the sex of an individual in these situations, dental techniques of sexual diagnosis are needed.In contrast to sex determination, the problem of age estimation has been the topic of many studies and so many different techniques, evidence of failure of all these methods, which the literature did not fail to highlight.The first aim of this work was to propose a simple and accurate dental technique for determining sex, in order to resolve this anthropological issue, reliably in the absence of other elements of the skeleton.The second was to develop a method for estimating the age of easy and fast implémentation, neither requiring a heavy technical support nor damage to material, for an application as well in forensics as in anthropology.The training sample consisted of 210 subjects’ CT-scans, each with four canines healthy. The validation sample was composed of 55 CT-scans, each with at least one healthy canine. Pulp and total volumes of each tooth were modeled and calculated using the software Mimics &#63720;. A binary logistic regression was used to determine seven prediction models of sex, according to the number and type of canine available in practice. The comparison of the areas under the ROC curve showed a greater performance of the model using the volumes of the four canines. The weighted least squares method was used to determine the equations for estimating the age for the same seven models. Greater performance of the model using the volumes of maxillary canines has been demonstrated. All these regressions were tested on the validation sample to perform their external validation.It turns out that in a scientific context where historical mandibular canines were the reference tool in the sex diagnosis, three-dimensional modeling emphasizes the potential value of maxillary counterparts. Moreover, it appears that the evaluation of the physiological pulp reduction, due to the natural apposition of secondary dentine on canines only, is a performance criterion in determining age.The models proposed in this study can therefore have reliable evidence, which all have all their place in the necessary body of evidence for determining age and sex, in estimating identification.

Sex estimation is a challenging problem in both forensic anthropology and bioarchaeology. Sexual dimorphism is most noticeably displayed by the pelvis; however in instances when it is not preserved, sex is estimated by skull. There is a multitude of approaches that use the skull, however, their population specificity and variable sexual dimorphism oscillation reduces their effectiveness (Bruzek and Murail, 2006). We base our contribution on the study by Abdel Fatah et al. (2014) that estimates sex based on exocranial and endocranial surfaces with a high success rate of 97%. Our approach uses anonymized CT scans of skulls from recent french population, from which the exocranial surface was segmented. On these surfaces, CPD-DCA (Dupej et al., 2014) was performed. We analyzed both form and shape (form after size normalization) of these surfaces in 104 skulls (53 males, 51 females) aged 18 to 92 years. The mean age was 58 years in females and 52.46 years in males. Classification was performed using support vector machines (SVM) with a radial kernel. Leave-one-out crossvalidation was also applied. The highest success rate (87.5 %) was achieved with the first 27 principal components of form. Classification of shape was less accurate by only 2 %. Even though our success rate was lower than that of Abdel...

This master's work was focused on the assessment of sexual dimorphism of external and internal morphology of frontal bone. Sexual diagnosis based on external surface of frontal bones and volume and surface of frontal sinuses was carried out using geometric morphometrics methods. Our approach used anonimized CT scans of 103 skulls from recent Czech population, from which 3D models of frontal bones and frontal sinuses were created. At first the analysis was aimed at studying form and shape of external surface of frontal bone. Our methodology is based on the work of Musilová et al. (2016) that estimates sex using exocranial surface of the entire skull. We analyzed variability and tested differences in males and females frontal bones. The main target of this work was sex classification. The highest success rate acquired 86,41 % with the first 20 main components of form. The following part of this study was focused on sex estimation using the methodology developed by Bulut et al. (2016), which was applied in Turkish population. This method classified Czech males and females with accuracy 70,87 %. In order to complete sex determination based on the frontal bone we also used classification of the radius of the frontal bone curvature with success rate of 77,66 %. The attention was also paid to studying...

No description available.