Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

Este trabalho desenvolve um modelo para análise de variáveis craniométricas que utiliza as Redes Neurais Artificiais Paraconsistentes, assentadas na Lógica Paraconsistente Anotada de dois valores. Tal lógica possui a capacidade de mensurar incerteza, inconsistência e paracompleteza. A Lógica Paraconsitente vem sendo empregada em diversas aplicações sujeitas a estas situações, constituindo nova ferramenta matemática em Inteligência Artificial. O trabalho tem como principal objetivo melhorar o diagnóstico cefalométrico. O modelo desenvolvido recebe as medidas das variáveis craniométricas de um determinado paciente e as compara com as médias das variáveis craniométricas normais de uma amostra da população brasileira. Esta amostra é composta de crianças e adolescentes de ambos os sexos, na faixa etária de 6 a 18 anos, utilizadas neste trabalho como valores de referência de normalidade. A análise cefalométrica aqui proposta consiste em quantificar discrepâncias esqueletais e dentárias sob a Lógica Paraconsistente. O uso das Redes Neurais Artificiais Paraconsistentes permite agregar ao méto do um fator de incerteza, respeitando o diagnóstico ortodôntico tradicional, e ao mesmo tempo, contextualiza diferentes regiões craniofaciais. O resultado da análise consiste dos graus de discrepância esqueletal, anteroposterior e vertical, e graus de discrepância dentárias, relativas aos incisivos inferiores e superiores. Variáveis craniométricas de 120 pacientes foram processadas pelo modelo proposto e avaliadas por três especialistas em Ortodontia. De acordo com o índice Kappa, houve desde concordância satisfatória até concordância quase perfeita entre o modelo e os especialistas, de acordo com as variáveis consideradas. As opiniões inter-especialista são substancialmente similares às comparações entre os especialistas e o modelo apresentado, o que reflete o potencial do modelo como um sistema especialista. A utilização de técnicas de Inteligência Artificial através da Lógica Paraconsistente, permitiu significante melhora na análise cefalométrica proposta. O modelo apresentado pode ser adaptado a outras amostras ou populações, com a adaptação dos valores de referência iniciais de normalidade. / This work shows the development of an unequal craniometric analysis model, which uses Paraconsistent Neural Network, based upon Paraconsistent Logic with two values. Such logical approach has the capability to handle concepts as uncertainness, inconsistency and paracompleteness. It has been used on diverse applications which present such features, constituting a new mathematical tool in Artificial Intelligence. The presented methodology had as main goal to booster diagnosis in Orthodontics. The developed model processes craniometric measures of a specific person, and compares to the expected means drawn from a Brazilian sample, comprised of children and adolescent individuals, ranging from 6 to 18 year-old, of both genders. The current cephalometric analysis, developed under the approach of Paraconsistent Logic, quantifies skeletal and dental discrepancies. The use of Paraconsistent Neural Network allows aggregating a factor of vagueness, respecting the limits of traditional orthodontic classification. At the same time, it contextualizes variables of different craniofacial regions. The results of the analysis are expressed through degrees of skeletal discrepancies, in the anteroposterior and vertical dimensions, and degrees of dental discrepancies, for the upper and lower incisors. Cephalometric va lues of a sample of 120 patients were processed by the paraconsistent model and analyzed by three specialists in Orthodontics. According to Kappa index, the agreement between the model and the specialists ranged from moderate to almost perfect, according to the variables considered. The inter-observer opinions were substantially similar to the mathematical model, which reflects the potential of the model as a specialist system. The use of Artificial Intelligence methods throughout Paraconsistent Logics, allowed significant improvement in cephalometric assessment. The presented model can be applied in different samples or populations, with adaptation of the degrees of normality as initial references.

Magister Scientiae Dentium - MSc(Dent) (Orthodontics) / Orthodontic treatment has the ability to improve the aesthetics and the function of patients. In order to create space, orthodontic treatment often requires removal of teeth. The most common teeth removed for orthodontic treatment are the premolars. It has become popular to remove second premolars in certain cases where the soft tissue profile should not be altered. The Damon self-ligating orthodontic system is renowned for not requiring dental extractions in the majority of cases. The effect of extractions on the soft tissue profile of patients, in conjunction with using the Damon system, has therefore not been researched. It is important to understand the effect that orthodontic treatment and extractions can have on the soft tissue profile of patients. This effect can accurately be determined by making use of the soft tissue cephalometric analysis, developed by Dr Reed A. Holdaway in 1983.

Este trabalho desenvolve um modelo para análise de variáveis craniométricas que utiliza as Redes Neurais Artificiais Paraconsistentes, assentadas na Lógica Paraconsistente Anotada de dois valores. Tal lógica possui a capacidade de mensurar incerteza, inconsistência e paracompleteza. A Lógica Paraconsitente vem sendo empregada em diversas aplicações sujeitas a estas situações, constituindo nova ferramenta matemática em Inteligência Artificial. O trabalho tem como principal objetivo melhorar o diagnóstico cefalométrico. O modelo desenvolvido recebe as medidas das variáveis craniométricas de um determinado paciente e as compara com as médias das variáveis craniométricas normais de uma amostra da população brasileira. Esta amostra é composta de crianças e adolescentes de ambos os sexos, na faixa etária de 6 a 18 anos, utilizadas neste trabalho como valores de referência de normalidade. A análise cefalométrica aqui proposta consiste em quantificar discrepâncias esqueletais e dentárias sob a Lógica Paraconsistente. O uso das Redes Neurais Artificiais Paraconsistentes permite agregar ao méto do um fator de incerteza, respeitando o diagnóstico ortodôntico tradicional, e ao mesmo tempo, contextualiza diferentes regiões craniofaciais. O resultado da análise consiste dos graus de discrepância esqueletal, anteroposterior e vertical, e graus de discrepância dentárias, relativas aos incisivos inferiores e superiores. Variáveis craniométricas de 120 pacientes foram processadas pelo modelo proposto e avaliadas por três especialistas em Ortodontia. De acordo com o índice Kappa, houve desde concordância satisfatória até concordância quase perfeita entre o modelo e os especialistas, de acordo com as variáveis consideradas. As opiniões inter-especialista são substancialmente similares às comparações entre os especialistas e o modelo apresentado, o que reflete o potencial do modelo como um sistema especialista. A utilização de técnicas de Inteligência Artificial através da Lógica Paraconsistente, permitiu significante melhora na análise cefalométrica proposta. O modelo apresentado pode ser adaptado a outras amostras ou populações, com a adaptação dos valores de referência iniciais de normalidade. / This work shows the development of an unequal craniometric analysis model, which uses Paraconsistent Neural Network, based upon Paraconsistent Logic with two values. Such logical approach has the capability to handle concepts as uncertainness, inconsistency and paracompleteness. It has been used on diverse applications which present such features, constituting a new mathematical tool in Artificial Intelligence. The presented methodology had as main goal to booster diagnosis in Orthodontics. The developed model processes craniometric measures of a specific person, and compares to the expected means drawn from a Brazilian sample, comprised of children and adolescent individuals, ranging from 6 to 18 year-old, of both genders. The current cephalometric analysis, developed under the approach of Paraconsistent Logic, quantifies skeletal and dental discrepancies. The use of Paraconsistent Neural Network allows aggregating a factor of vagueness, respecting the limits of traditional orthodontic classification. At the same time, it contextualizes variables of different craniofacial regions. The results of the analysis are expressed through degrees of skeletal discrepancies, in the anteroposterior and vertical dimensions, and degrees of dental discrepancies, for the upper and lower incisors. Cephalometric va lues of a sample of 120 patients were processed by the paraconsistent model and analyzed by three specialists in Orthodontics. According to Kappa index, the agreement between the model and the specialists ranged from moderate to almost perfect, according to the variables considered. The inter-observer opinions were substantially similar to the mathematical model, which reflects the potential of the model as a specialist system. The use of Artificial Intelligence methods throughout Paraconsistent Logics, allowed significant improvement in cephalometric assessment. The presented model can be applied in different samples or populations, with adaptation of the degrees of normality as initial references.

Introduction : L'objectif de cette étude est de comparer la différence entre l'analyse céphalométrique manuelle et l'analyse automatisée par l’intelligence artificielle afin de confirmer la fiabilité de cette dernière. Notre hypothèse de recherche est que la technique manuelle est la plus fiable des deux méthodes. Méthode : Un total de 99 radiographies céphalométriques latérales sont recueillies. Des tracés par technique manuelle (MT) et par localisation automatisée par intelligence artificielle (AI) sont réalisés pour toutes les radiographies. La localisation de 29 points céphalométriques couramment utilisés est comparée entre les deux groupes. L'erreur radiale moyenne (MRE) et un taux de détection réussie (SDR) de 2 mm sont utilisés pour comparer les deux groupes. Le logiciel AudaxCeph version 6.2.57.4225 est utilisé pour l'analyse manuelle et l'analyse AI. Résultats : Le MRE et SDR pour le test de fiabilité inter-examinateur sont respectivement de 0,87 ± 0,61mm et 95%. Pour la comparaison entre la technique manuelle MT et le repérage par intelligence artificielle AI, le MRE et SDR pour tous les repères sont respectivement de 1,48 ± 1,42 mm et 78 %. Lorsque les repères dentaires sont exclus, le MRE diminue à 1,33 ± 1,39 mm et le SDR augmente à 84 %. Lorsque seuls les repères des tissus durs sont inclus (excluant les points des tissus mous et dentaires), le MRE diminue encore à 1,25 ± 1,09 mm et le SDR augmente à 85 %. Lorsque seuls les points de repère des tissus mous sont inclus, le MRE augmente à 1,68 ± 1,89 mm et le SDR diminue à 78 %. Conclusion: La performance du logiciel est similaire à celles précédemment rapportée dans la littérature pour des logiciels utilisant un cadre de modélisation similaire. Nos résultats révèlent que le repérage manuel a donné lieu à une plus grande précision. Le logiciel a obtenu de très bons résultats pour les points de tissus durs, mais sa précision a diminué pour les tissus mous et dentaires. Nous concluons que cette technologie est très prometteuse pour une application en milieu clinique sous la supervision du docteur. / Introduction: The objective of this study is to compare the difference between manual cephalometric analysis and automatic analysis by artificial intelligence to confirm the reliability of the latter. Our research hypothesis is that the manual technique is the most reliable of the methods and is still considered the gold standard. Method: A total of 99 lateral cephalometric radiographs were collected in this study. Manual technique (MT) and automatic localization by artificial intelligence (AI) tracings were performed for all radiographs. The localization of 29 commonly used landmarks were compared between both groups. Mean radial error (MRE) and a successful detection rate (SDR) of 2mm were used to compare both groups. AudaxCeph software version 6.2.57.4225 (Audax d.o.o., Ljubljana, Slovenia) was used for both manual and AI analysis. Results: The MRE and SDR for the inter-examinator reliability test were 0.87 ± 0.61mm and 95% respectively. For the comparison between the manual technique MT and landmarking with artificial intelligence AI, the MRE and SDR for all landmarks were 1.48 ± 1.42mm and 78% respectively. When dental landmarks are excluded, the MRE decreases to 1.33 ± 1.39mm and the SDR increases to 84%. When only hard tissue landmarks are included (excluding soft tissue and dental points) the MRE decreases further to 1.25 ± 1.09mm and the SDR increases to 85%. When only soft tissue landmarks are included the MRE increases to 1.68 ± 1.89mm and the SDR decreases to 78%. Conclusion: The software performed similarly to what was previously reported in literature for software that use analogous modeling framework. Comparing the software’s landmarking to manual landmarking our results reveal that the manual landmarking resulted in higher accuracy. The software operated very well for hard tissue points, but its accuracy went down for soft and dental tissue. Our conclusion is this technology shows great promise for application in clinical settings under the doctor’s supervision.