The problem of reconstructing an image from a set of tomographic data is not new, nor is it lacking attention. However there is still a distinct gap between the mathematicians and the experimental scientists working in the computed tomography (CT) imaging community. One of the aims in this thesis is to bridge this gap with mathematical reconstruction algorithms and analysis approaches applied to practical CT problems. The thesis begins with an extensive analysis for assessing the suitability of reconstruction algorithms for a given problem. The paper presented examines the idea of extracting physical information from a reconstructed sample and comparing against the known sample characteristics to determine the accuracy of a reconstructed volume. Various test cases are studied, which are relevant to both mathematicians and experimental scientists. These include the variance in quality of reconstructed volume as the dose is reduced or the implementation of the level set evolution method, used as part of a simultaneous reconstruction and segmentation technique. The work shows that the assessment of physical attributes results in more accurate conclusions. Furthermore, this approach allows for further analysis into interesting questions in CT. This theme is continued throughout the thesis. Recent results in compressive sensing (CS) gained attention in the CT community as they indicate the possibility of obtaining an accurate reconstruction of a sparse image from severely limited or reduced amount of measured data. Literature produced so far has not shown that CS directly guarantees a successful recovery in X-ray CT, and it is still unclear under which conditions a successful sparsity regularized reconstruction can be achieved. The work presented in the thesis aims to answer this question in a practical setting, and seeks to establish a direct connection between the success of sparsity regularization methods and the sparsity level of the image, which is similar to CS. Using this connection, one can determine the sufficient amount of measurements to collect from just the sparsity of an image. A link was found in a previous study using simulated data, and the work is repeated here with experimental data, where the sparsity level of the scanned object varies. The preliminary work presented here verifies the results from simulated data, showing an "almost-linear" relationship between the sparsity of the image and the sufficient amount of data for a successful sparsity regularized reconstruction. Several unexplained artefacts are noted in the literature as the `partial volume', the 'exponential edge gradient' or the 'penumbra' effect, with no clear explanation for their cause, or established techniques to remove them. The work presented in this paper shows that these artefacts are due to a non-linearity in the measured data, which comes from either the set up of the system, the scattering of rays or the dependency of linear attenuation on wavelength in the polychromatic case. However, even in monochromatic CT systems, the non-linearity effect can be detected. The paper shows that in some cases, the non-linearity effect is too large to ignore, and the reconstruction problem should be adapted to solve a non-linear problem. We derive this non-linear problem and solve it using a numerical optimization technique for both simulatedand real, gamma-ray data. When compared to reconstructions obtained using the standard linear model, the non-linear reconstructed images show clear improvements in that the non-linear effect is largely eliminated. The thesis is finished with a highlight article in the special issue of Solid Earth, named "Pore-scale tomography & imaging - applications, techniques and recommended practice". The paper presents a major technical advancement in a dynamic 3D CT data acquisition, where the latest hardware and optimal data acquisition plan are applied and as a result, ultra fast 3D volume acquisition was made possible. The experiment comprised of fast, free-falling water-saline drops traveling through a pack of rock grains with varying porosities. The imaging work was enhanced by the use of iterative methods and physical quantification analysis performed. The data acquisition and imaging work is the first in the field to capture a free falling drop and the imaging work clearly shows the fluid interaction with speed, gravity and more importantly, the inter- and intra-grain fluid transfers.

Made available in DSpace on 2014-10-09T12:34:55Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T14:06:59Z (GMT). No. of bitstreams: 0 / Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) / Tese (Doutoramento) / IPEN/T / Instituto de Pesquisas Energeticas e Nucleares - IPEN-CNEN/SP / FAPESP:09/07417-9

Submitted by Renata Lopes (renatasil82@gmail.com) on 2015-12-02T16:29:15Z No. of bitstreams: 1 marianarochanadaes.pdf: 1380064 bytes, checksum: f78140bd2137599acdd476c8c0e433a7 (MD5) / Rejected by Adriana Oliveira (adriana.oliveira@ufjf.edu.br), reason: Coreeção letras maiúsculas título on 2015-12-03T11:30:57Z (GMT) / Submitted by Renata Lopes (renatasil82@gmail.com) on 2015-12-03T11:36:28Z No. of bitstreams: 1 marianarochanadaes.pdf: 1380064 bytes, checksum: f78140bd2137599acdd476c8c0e433a7 (MD5) / Rejected by Adriana Oliveira (adriana.oliveira@ufjf.edu.br), reason: Título: somente primeira letra da primeira palavra em maiúsculo e nomes próprios também. Palavras-chave: primeira letra de cada palavra em maiúsculo. Acrescentar uma em cada linha (não precisa colocar ponto e nem vírgula no final) on 2015-12-03T11:59:32Z (GMT) / Submitted by Renata Lopes (renatasil82@gmail.com) on 2015-12-03T13:42:04Z No. of bitstreams: 1 marianarochanadaes.pdf: 1380064 bytes, checksum: f78140bd2137599acdd476c8c0e433a7 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2015-12-03T13:59:10Z (GMT) No. of bitstreams: 1 marianarochanadaes.pdf: 1380064 bytes, checksum: f78140bd2137599acdd476c8c0e433a7 (MD5) / Made available in DSpace on 2015-12-03T13:59:10Z (GMT). No. of bitstreams: 1 marianarochanadaes.pdf: 1380064 bytes, checksum: f78140bd2137599acdd476c8c0e433a7 (MD5) Previous issue date: 2015-02-06 / FAPEMIG - Fundação de Amparo à Pesquisa do Estado de Minas Gerais / Apesar das muitas indicações da tomografia computadorizada de feixe cônico (TCFC) na Odontologia, a avaliação da qualidade óssea, por meio das Unidades Hounsfield (HU), não apresenta um consenso sobre sua precisão e confiabilidade. O objetivo neste estudo foi avaliar, por meio de uma revisão sistemática da literatura, os principais resultados de pesquisas que compararam diversos materiais em exames de tomografia computadorizada multislice (TCMS) e TCFC, utilizando HU como unidade de medida. Para isso, foi realizada uma busca na base de dados Medline (PubMed) utilizando as palavras chaves “cone beam computed tomography”, “cone beam tomography”, “cbct”, “Hounsfield unit” e “Hounsfield”. A exclusão dos artigos foi realizada, inicialmente, por meio da leitura do título e resumo e, posteriormente, após a leitura do texto completo. Os dados da amostra selecionada (n=8) foram distribuídos em tabelas, possibilitando sua comparação. Os resultados obtidos em relação à aplicação de HU em TCFC foram divergentes, com alguns autores aplicando diretamente HU nesse exame, outros desenvolvendo fatores de correção para essa comparação e outros ainda, invalidando a utilização dessa unidade em exames de TCFC. Apenas um estudo concluiu que se pode aplicar diretamente HU em TCFC, sem prejuízo para avaliação das densidades dos tecidos. Assim, pode-se concluir que uma nova abordagem é necessária para o desenvolvimento de futuros trabalhos nessa linha e que a tentativa de aplicar HU em TCFC deveria ser abandonada. / Despite the many indications of cone beam computed tomography (CBCT) in dentistry, there is no consensus regarding the accuracy and reliability of bone quality assessment, through Hounsfield Units (HU). The aim of this study was to evaluate, through a systematic review of the literature, the main results of studies comparing various tissues in multislice CT (MSCT) and CBCT scans, using the HU as the unit of measurement. To this end, a search was conducted in the Medline (PubMed) database using the keywords "cone beam computed tomography", "cone beam tomography", "CBCT", "Hounsfield unit", and "Hounsfield". Articles were initially excluded by reading the title and abstract, and were later excluded after reading the full text. The data from the selected sample (n = 8) were tabulated and compared. The results obtained regarding the application of HU in CBCT were divergent, with some authors directly applying HU in CBCT, others developing correction factors for this comparison, and still others invalidating the use of this unit in CBCT. Only one study concluded that is possible to directly apply HU in CBCT, without damaging the evaluation of tissue densities. In this way, it’s possible to affirm that a new approach is necessary for the development of future work in this line and that the attempt to apply HU in CBCT should be abandoned.

Unsicherheiten in der Reichweitevorhersage limitieren wesentlich das Ausnutzen der Vorteile von Protonentherapie gegenüber konventioneller Strahlentherapie. Die Verwendung von Zwei-Spektren-Computertomographie (DECT) zur direkten Vorhersage des Bremsvermögen (DirectSPR) ermöglicht eine relevante Verbesserung der Reichweitevorhersage gegenüber der üblicherweise verwendeten Ein-Spektren-Computertomographie (SECT). Im Rahmen dieser Dissertation wurde die Variation in der Reichweitevorhersage zwischen 17 europäischen Partikeltherapiezentren experimentell verglichen. Die Genauigkeit der Reichweitevorhersage bei Verwendung einer DirectSPR-Implementierung wurde umfassend quantifiziert und die Implementierung in die klinische Routine integriert. Dies führte zu einer Reduzierung des klinischen Sicherheitssaum um ca. 35% für die Behandlung von quasistatischen Tumoren in Kopf und Becken und damit einer Schonung des Normalgewebes sowie der das Zielgebiet umgebenden Risikoorgane. Darüber hinaus wurde die DirectSPR-Implementierung zur Bestimmung von Gewebeparametern sowie deren Variabilität für zehn Organe im Kopf und Becken in einer Patienkohorte genutzt. Die vorgestellten Ergebnisse etablieren DECT weiter als zukünftiges Standard-Bildgebungsverfahren in der Partikeltherapie.:1. Introduction 2. Proton therapy 2.1. Physical principles of proton therapy 2.2. Treatment with protons 2.3. Accuracy in proton therapy 3. CT Imaging for proton therapy 3.1. Principles of CT imaging 3.2. CT-based range prediction 3.3. Investigated phantoms and materials 3.4. DECT scan acquisition 3.5. Determination of proton stopping power for reference materials 4. Accuracy of stopping-power prediction in European proton centres 4.1. Study design 4.2. Experimental setup and analysis 4.3. Results 4.4. Discussion of determined deviations 4.5. Conclusion and outlook 4.6. Establishment of guidelines for HLUT calibration 5. Range uncertainties in DirectSPR-based treatment planning 5.1. Clinical implementation of DirectSPR 5.2. Uncertainty quantification 5.3. Resulting uncertainties in SPR prediction 5.4. Experimental validation 5.5. Dosimetric effect of range uncertainty reduction 5.6. Discussion 6. In-vivo tissue characterisation using DirectSPR 6.1. Tissue parameter determination by Woodard and White 6.2. Data preparation and analysis 6.3. Determined tissue parameters and variations 6.4. Discussion 7. The future of image-based range prediction 7.1. Particle imaging 7.2. Creation of synthetic CT images 7.3. Photon-counting computed tomography 8. Summary 9. Zusammenfassung A. Supplement A.1. Investigated materials A.2. EPTN study: Individual results A.3. DirectSPR validation results / Imaging-related range uncertainties effectively limit the full exploitation of the benefits proton therapy offers with respect to conventional photon radiotherapy. The use of dual-energy computed tomography (DECT) for direct stopping-power prediction (DirectSPR) was determined to provide relevant improvements in range prediction over commonly used singleenergy CT (SECT). Within this thesis, the variation in range prediction accuracy between 17 European particle treatment centres were experimentally quantified to determine the current status quo in the community. The overall range uncertainty when using a DirectSPR implementation in treatment planning was comprehensively quantified and the implementation integrated into the clinical workflow. This led to a reduction of clinical safety margins by about 35% for the treatment of quasi-static tumours in the head and pelvis, effectively reducing the dose to surrounding healthy tissue and organs at risk. The DirectSPR implementation was furthermore utilised to assess tissue parameters and their inter- and intra-patient variability for ten organs in the head and pelvis from a cohort of patients. The presented results further establish DirectSPR as the future standard imaging modality in particle therapy.:1. Introduction 2. Proton therapy 2.1. Physical principles of proton therapy 2.2. Treatment with protons 2.3. Accuracy in proton therapy 3. CT Imaging for proton therapy 3.1. Principles of CT imaging 3.2. CT-based range prediction 3.3. Investigated phantoms and materials 3.4. DECT scan acquisition 3.5. Determination of proton stopping power for reference materials 4. Accuracy of stopping-power prediction in European proton centres 4.1. Study design 4.2. Experimental setup and analysis 4.3. Results 4.4. Discussion of determined deviations 4.5. Conclusion and outlook 4.6. Establishment of guidelines for HLUT calibration 5. Range uncertainties in DirectSPR-based treatment planning 5.1. Clinical implementation of DirectSPR 5.2. Uncertainty quantification 5.3. Resulting uncertainties in SPR prediction 5.4. Experimental validation 5.5. Dosimetric effect of range uncertainty reduction 5.6. Discussion 6. In-vivo tissue characterisation using DirectSPR 6.1. Tissue parameter determination by Woodard and White 6.2. Data preparation and analysis 6.3. Determined tissue parameters and variations 6.4. Discussion 7. The future of image-based range prediction 7.1. Particle imaging 7.2. Creation of synthetic CT images 7.3. Photon-counting computed tomography 8. Summary 9. Zusammenfassung A. Supplement A.1. Investigated materials A.2. EPTN study: Individual results A.3. DirectSPR validation results

L'objectif de cette thèse était de quantifier le réseau lacuno-canaliculaire du tissu osseux à partir d’images 3D acquises en nano CT synchrotron de phase. Ceci a nécessité d’optimiser les processus d’acquisition et de reconstruction de phase, ainsi que de développer des méthodes efficaces de traitement d'images pour la segmentation et l’analyse 3D. Dans un premier temps, nous avons étudié et évalué différents algorithmes de reconstruction de phase. Nous avons étendu la méthode de Paganin pour plusieurs distances de propagation et l’avons évaluée et comparée à d’autres méthodes, théoriquement puis sur nos données expérimentales Nous avons développé une chaine d’analyse, incluant la segmentation des images et prenant en compte les gros volumes de données à traiter. Pour la segmentation des lacunes, nous avons choisi des méthodes telles que le filtre médian, le seuillage par hystérésis et l'analyse par composantes connexes. La segmentation des canalicules repose sur une méthode de croissance de région après rehaussement des structures tubulaires. Nous avons calculé des paramètres de porosité, des descripteurs morphologiques des lacunes ainsi que des nombres de canalicules par lacune. Par ailleurs, nous avons introduit des notions de paramètres locaux calculés dans le voisinage des lacunes. Nous avons obtenu des résultats sur des images acquises à différentes tailles de voxel (120nm, 50nm, 30nm) et avons également pu étudier l’impact de la taille de voxel sur les résultats. Finalement ces méthodes ont été utilisées pour analyser un ensemble de 27 échantillons acquis à 100 nm dans le cadre du projet ANR MULTIPS. Nous avons pu réaliser une analyse statistique pour étudier les différences liées au sexe et à l'âge. Nos travaux apportent de nouvelles données quantitatives sur le tissu osseux qui devraient contribuer à la recherche sur les mécanismes de fragilité osseuse en relation avec des maladies comme l’ostéoporose. / Osteoporosis is a bone fragility disease resulting in abnormalities in bone mass and density. In order to prevent osteoporotic fractures, it is important to have a better understanding of the processes involved in fracture at various scales. As the most abundant bone cells, osteocytes may act as orchestrators of bone remodeling which regulate the activities of both osteoclasts and osteoblasts. The osteocyte system is deeply embedded inside the bone matrix and also called lacuno-canalicular network (LCN). Although several imaging techniques have recently been proposed, the 3D observation and analysis of the LCN at high spatial resolution is still challenging. The aim of this work was to investigate and analyze the LCN in human cortical bone in three dimensions with an isotropic spatial resolution using magnified X-ray phase nano-CT. We performed image acquisition at different voxel sizes of 120 nm, 100 nm, 50 nm and 30 nm in the beamlines ID16A and ID16B of the European Synchrotron Radiation Facility (ESRF - European Synchrotron Radiation Facility - Grenoble). Our first study concerned phase retrieval, which is the first step of data processing and consists in solving a non-linear inverse problem. We proposed an extension of Paganin’s method suited to multi-distance acquisitions, which has been used to retrieve phase maps in our experiments. The method was compared theoretically and experimentally to the contrast transfer function (CTF) approach for homogeneous object. The analysis of the 3D reconstructed images requires first to segment the LCN, including both the segmentation of lacunae and of canaliculi. We developed a workflow based on median filter, hysteresis thresholding and morphology filters to segment lacunae. Concerning the segmentation of canaliculi, we made use of the vesselness enhancement to improve the visibility of line structures, the variational region growing to extract canaliculi and connected components analysis to remove residual noise. For the quantitative assessment of the LCN, we calculated morphological descriptors based on an automatic and efficient 3D analysis method developed in our group. For the lacunae, we calculated some parameters like the number of lacunae, the bone volume, the total volume of all lacunae, the lacunar volume density, the average lacunae volume, the average lacunae surface, the average length, width and depth of lacunae. For the canaliculi, we first computed the total volume of all the canaliculi and canalicular volume density. Moreover, we counted the number of canaliculi at different distances from the surface of each lacuna by an automatic method, which could be used to evaluate the ramification of canaliculi. We reported the statistical results obtained on the different groups and at different spatial resolutions, providing unique information about the organization of the LCN in human bone in three dimensions.

O propósito deste estudo foi avaliar a acurácia do exame de tomografia computadorizada multislice (TCM) e tomografia computadorizada por feixe cônico (TCFC) na identificação de lesões simuladas em mandíbula, em situações com e sem a presença de artefatos metálicos, em diversos protocolos de observação. Foram realizados exames de TCM e TCFC de mandíbulas secas, nas quais foram executadas perfurações simulando lesões. As imagens foram realizadas em dois momentos: na presença e na ausência de restaurações dentárias metálicas. Dois observadores, previamente calibrados, observaram as imagens avaliando-as quanto à presença ou ausência de lesão, número de lojas das lesões e a existência ou não de invasão medular. Os mesmos utilizaram programas de manipulação de imagens instalados em estações de trabalho independentes, para reconstruir as imagens nos seguintes protocolos de avaliação: axial, sagital + coronal, 3D, conjunto (axial+sagital+coronal+3D) e parassagital. A sensibilidade e especificidade (validade) da tomografia computadorizada multislice (64 cortes) (TCM) e da tomografia computadorizada por feixe cônico (TCFC) para diagnóstico de lesões ósseas (simuladas) em mandíbula, utilizando estação de trabalho independente foram demonstradas à medida que os valores encontrados foram superiores a 95% desde que com o protocolo de observação adequado. A influência de artefatos dentários metálicos foi pouco significativa na interpretação de lesões ósseas mandibulares, já que os valores de acurácia nas análises com e sem artefato foram bastante próximas. Os protocolos com aquisição por TCM sofreram mais influência dos artefatos do que os adquiridos por TCFC (valores sutilmente menores), exceto nas reconstruções em 3D, em que as originadas de aquisição por TCFC, apresentaram valores menores de acurácia. O melhor protocolo de pós processamento para interpretação de lesões ósseas simuladas foi o denominado RMP+3D. O protocolo que apresentou os piores resultados foi o que utilizou as reconstruções parassagitais. / The purpose of this study was to evaluate the accuracy of multislice computed tomography (MSCT) and cone-beam computed tomography (CBCT) in identification of simulated mandibular lesions in situations with and without metallic artifacts in several observation protocols. MSCT and CBCT examinations were performed in dry mandibles, in which holes were performed simulating lesions. The images were taken in two stages: in the presence and absence of metal dental restorations. Two observers, previously calibrated, observed the images by evaluating the images for the presence or absence of lesion, loci number and whether there were medullary invasion. Observers used image manipulation softwares, installed on independent workstations, to reconstruct the images in the following evaluation protocols: axial, sagittal + coronal, 3D, sets (axial+ coronal + sagittal + 3 D) and parasagittal. The sensitivity and specificity of MSCT (64 slices) and CBCT for diagnosis of simulated bone lesions in the mandible, using independent workstation were demonstrated as the values obtained were greater than 95% using the appropriate observation protocol. The influence of dental metallic artifacts was negligible in the interpretation of mandibular bone lesions, since the values of accuracy in the analysis with and without artifact were quite close. The images acquired with MCT suffered more influence of artifacts than the protocols acquired by TCFC, although the values were all high and quite close. Except for 3D reconstructions, which originated from the CBCT, showed the lowest accuracy. The best protocol for post-processing and interpretation of simulated bone lesions was called RMP +3 D. The protocol that showed the worst results was parasagital.

O sucesso da cirurgia ortognática está relacionado com diversos fatores como análise facial precisa e acurada, e exames de imagem com alta qualidade. Atualmente, diversos programas de imagens tridimensionais, comerciais e abertos, estão disponíveis no auxílio diagnóstico, elaboração do plano de tratamento e simulação do procedimento cirúrgico para correção das deformidades dentofaciais. Entretanto, estudos avaliando os diferentes métodos de mensuração linear empregados pelos programas de imagem por meio da reconstrução 3D no planejamento da osteotomia tipo Le Fort I são escassos e, maiores esclarecimentos sobre sua precisão e a acurácia são necessários. O objetivo deste estudo foi estabelecer: a) precisão e acurácia de medidas lineares tridimensionais para osteotomia tipo Le Fort I obtidas por meio da tomografia computadorizada multislice (TCMS) e feixe cônico (TCFC) e b) comparar a precisão e acurácia de medidas lineares para osteotomia tipo Le Fort I realizadas por meio de três programas de imagem utilizando a 3D-TCFC. A amostra foi constituída por onze crânios secos submetidos à TCMS 64 canais e TCFC. As reconstruções tridimensionais (3D-TC) foram geradas, e medidas lineares (n=11) baseadas em estruturas e pontos anatômicos de interesse à osteotomia tipo Le Fort I foram realizados independentemente, por dois radiologistas experientes, duas vezes cada, utilizando programa Vítrea 3.8.1 em reconstruções 3D-TCMS e 3D-TCFC e os programas OsiriX 1.2 64-bit e Dolphin Imaging versão 11.5.04.35 por meio da 3D-TCFC. Sequencialmente, um terceiro observador experiente e calibrado, que não participou da análise das imagens, realizou as medidas sobre os crânios secos utilizando paquímetro digital (padrão ouro) com quais as medidas foram comparadas. A análise intra e inter-observadores assim como as correlações individuais de cada medida foram realizadas utilizando o coeficiente de correlação intra-classe (CCI), sendo o intervalo de confiança adotado de 95%. Os resultados demonstraram na análise intra-observador, utilizando o programa Vítrea, correlação excelente para todas as medidas variando de 0,87 a 0,96 e 0,82 a 0,98 para os observadores 1 e 2 respectivamente utilizando a TCMS e de 0,84 a 0,98 e 0,80 a 0,98 utilizando a TCFC. A análise inter-observadores variou de 0,85 a 0,98 para a TCMS e de 0,80 a 0,99 para a TCFC. A análise intra-observador utilizando os diferentes programas de imagem na TCFC apresentou variações de 0,90 a 0,97 (Vítrea), 0,65 a 0,97 (OsiriX) e 0,51 a 0,94 (Dolphin). Na análise inter-obsevador obtivemos valores de 0,92 a 0,99 e 0,88 a 0,98 para os observadores 1 e 2 respectivamente utilizando o Vítrea, 0,58 a 0,90 e 0,48 a 0,85 para o programa OsiriX e de 0,80 a 0,96 e 0,57 a 0,92 com o programa Dolphin. O programa Vítrea não apresentou diferenças estatisticamente significante nas análises intra e inter-observador e medidas físicas utilizando a TCMS e TCFC. Em relação à análise dos programas, diferenças estatisticamente foram constatadas com o programa OsiriX e Dolphin em relação ao padrão ouro. / The success of orthognathic surgery depends on many factors as precise and accurate facial analysis and high quality imaging exams. Nowadays a lot of commercial and open-source three-dimensional software programs currently available to assist diagnosis, elaboration of treatment planning, and to predict outcomes related to orthognathic surgery. The aim of this study was to establish: a) the precision and accuracy of three-dimensional linear measurements for Le Fort I osteotomy, obtained from multislice computed tomography (MSCT) and cone-beam computed tomography (CBCT) scans and b) compare the precision and accuracy of linear measurements for Le Fort I osteotomy performed by three different imaging software programs and obtained from 3D-CBCT images. The study population consisted of 11 dried skulls submitted to 64-row MSCT and CBCT scans. Three-dimensional reconstructed images (3D-CT) were generated, and linear measurements (n = 11) based on anatomical structures and landmarks of interest for Le Fort I osteotomy were performed independently by 2 oral and maxillofacial radiologists, twice each, using Vítrea 3.8.1 software for 3D-MSCT and 3D-CBCT and OsiriX 1.2 64-bit and Dolphin Imaging version 11.5.04.35 from 3D-CBCT. Subsequently, a third examiner expert in anatomical analysis and calibrated, who did not evaluate the images, performed measurements on dry skulls using a digital caliper (gold standard) with which the measurements were compared. The analyses of intra- and inter-observer as well as the individual correlations of each measurement were performed using the coefficient of intra-class correlation (ICC), with a confidence interval of 95 %. The intra-observer results showed that using the Vitrea program, excellent correlation for all measurements were reached with ICC values ranging from 0.87 to 0.96 and 0.82 0.98 for observers 1 and 2 respectively using the MSCT and 0.84 to 0.98 and 0.80 to 0.98 using CBCT. Inter-observer analysis ranged from 0.85 to 0.98 for MSCT and 0.80 to 0.99 for CBCT. The intra-observer analysis using the different programs using CBCT images varied from 0.90 to 0.97 for Vitrea, 0.65 to 0.97 for OsiriX and 0.51 to 0.94 using Dolphin. Inter-observer analysis demonstrated values ranging from 0.92 to 0.99 and 0.88 0.98 for observers 1 and 2 respectively using the Vitrea; 0.58 to 0.90 and 0.48 to 0.85 for OsiriX, and 0.80 to 0.96 and 0.57 to 0.92 using Dolphin. Vítrea software showed no statistically significant differences for intra-and inter-observer analysis and physical measurements using MSCT and CBCT. Regarding the analysis of softwares, statistical differences were found with the program OsiriX and Dolphin in comparison to gold standard.

Os dentes incisivos inferiores apresentam, em sua maioria, canal radicular único, que pode ter conformações distintas. De acordo com a literatura científica, a presença de um segundo canal radicular pode ser constatada em 10 a 40% dos casos. A não-detecção do segundo canal é um importante fator para o insucesso do tratamento endodôntico. Esta pesquisa foi realizada com os seguintes objetivos: 1) comparar a eficácia na identificação de padrões anatômicos internos em incisivos inferiores permanentes extraídos com uso de imagem radiográfica periapical digital e de tomografia computadorizada de feixe cônico (TCFC); e 2) determinar a prevalência de 2 canais radiculares em incisivos inferiores em imagens de TCFC de um banco de dados de exames previamente realizados, associando com localização do dente, gênero e idade do paciente. Na primeira etapa da pesquisa, 40 dentes incisivos inferiores foram submetidos a escaneamento em microtomógrafo computadorizado da FOB-USP (Skyscan 1074) para estabelecer o padrão ouro do tipo anatômico interno de cada dente. Os dentes foram então classificados em: Tipo I (1 canal radicular regular, n=12), Tipo Ia (1 canal radicular oval, n=12) e Tipo III (2 canais radiculares, n=16). Em seguida, os dentes foram divididos em 10 grupos de 4 dentes e posicionados em alvéolos de uma mandíbula humana para exposição radiográfica periapical digital direta com duas tomadas variando o ângulo horizontal de incidência (Schick CDR) e para escaneamento em 3 tomógrafos (Kodak 9000 3D, Veraviewepocs 3De e NewTom 5G) da Universidade de Loma Linda, CA, EUA. Dois examinadores treinados classificaram o tipo anatômico de cada dente e suas respostas foram comparadas ao padrão ouro estabelecido por microtomografia. Para descrição dos resultados, utilizou-se a porcentagem (%) de respostas certas / erradas. Aplicou-se o teste do qui-quadrado (X2), com nível de significância estatística de 5%, para verificação da associação entre variáveis. As concordâncias inter e intraexaminador foram determinadas por teste de kappa. Na segunda etapa da pesquisa, foram interpretadas imagens de incisivos inferiores de 100 pacientes, cujos exames estavam disponíveis no banco de dados do tomógrafo i-CAT Classic da FOB-USP. Dois examinadores avaliaram a anatomia interna de incisivos inferiores, em reconstruções axiais, sagitais e coronais, classificando-a de acordo com o número de canais radiculares. A prevalência de 2 canais radiculares foi relacionada à localização do dente e ao gênero e idade do paciente. Para descrição dos dados coletados, utilizou-se a porcentagem (%). Para verificar a associação entre variáveis, utilizou-se o teste do qui-quadrado (X2), com nível de significância estatística de 5%. Para a concordância intra e interexaminador, utilizou-se o teste de kappa. Como resultado da primeira etapa da pesquisa, obteve-se alto índice de detecção de tipos anatômicos para todos os métodos avaliados (p <0,05). Para dentes com Tipo I (1 canal radicular regular), as imagens de TCFC foram superiores em comparação à radiografia periapical digital com dupla exposição (RP= 67% de acerto; TCFC= 98% de acerto, p <0,05). Considerando os 3 aparelhos de TCFC, não houve diferença estatística significante entre eles para identificação do Tipo I. Já para dentes com Tipo Ia (1 canal radicular oval), houve diferença estatística significante somente entre radiografia periapical digital com dupla exposição e o tomógrafo NewTom (RP= 44% de acerto; TCFC NewTom= 88% de acerto). Não houve diferença significante entre os outros 2 tomógrafos e radiografia periapical ou entre os 3 tomógrafos. Considerando dentes com Tipo III (2 canais radiculares), não houve diferença estatística significante entre nenhum dos métodos. Todos os métodos apresentaram números de respostas certas semelhantes para o Tipo III, sugerindo que a radiografia periapical com dupla angulação é suficiente para identificação de 2 canais radiculares em incisivos inferiores. A concordância intraexaminador para radiografia periapical foi regular (kappa=0,40 a 0,66) e de boa a excelente para os aparelhos de TCFC (kappa=0,62 a 0,85). A concordância interexaminador para radiografia periapical foi de ruim a regular (kappa=0,25 a 0,32) e de boa a excelente para os aparelhos de TCFC (kappa=0,62 a 0,92). Na segunda etapa da pesquisa, o total de 386 incisivos inferiores foi avaliado, sendo 192 incisivos centrais e 194 incisivos laterais. A prevalência de 2 canais radiculares constatada no total de dentes foi de 16,5%, sendo de 13% em incisivos centrais e de 20% nos incisivos laterais (p >0,05). Não houve diferença estatisticamente significante entre gêneros e idades. A concordância intra e interexaminador foi regular (kappa intra=0,60; kappa inter=0,57). Como conclusão da primeira etapa da pesquisa, observou-se que a identificação do número de canais radiculares foi efetiva em todos os métodos. No entanto, limitações foram encontradas para a diferenciação da forma do canal radicular. Na segunda etapa da pesquisa, concluiuse que a presença de 2 canais radiculares em incisivos inferiores de pacientes da região de Bauru é de aproximadamente 20% e não depende da localização do dente ou de gênero e idade do paciente. / Mandibular incisors most commonly have a single root canal, which can present with different anatomic configurations. According to the literature, the presence of a second root canal can be observed in 10 - 40% of the teeth. Difficulty in detecting the second root canal is an important factor for the endodontic treatment failure. The aim of this research was: 1) to compare the efficacy of digital periapical radiography (PA) with double exposure and cone beam computed tomography (CBCT) in the identification of internal anatomic patterns in mandibular incisors, and 2) to determine the prevalence of 2 root canals in mandibular incisors using cone beam computed tomographic images of a patient database, comparing to tooth position and patients gender and age. In the first part of this research, 40 extracted mandibular incisors underwent microcomputed tomographic (micro-CT) scanning (Skyscan 1074) in order to establish the gold standard for internal anatomic pattern. The teeth were classified according to: Type I (1 regular root canal, n=12), Type Ia (1 oval root canal, n=12), and Type III (2 root canals, n=16). Then, the teeth were divided into 10 groups of 4 teeth and placed in a preserved human mandible for direct digital periapical radiographic double exposure (Schick CDR) and CBCT scans using Kodak 9000 3D, Veraviewepocs 3De and NewTom 5G. Two blinded examiners classified the anatomic pattern of each tooth and their answers were compared to the gold standard (microtomographic images). Percentage (%) of right / wrong answers was used for the statistical analysis of the results. Chi-square test (X2) was used to verify the association between variables (p <.05). Inter and intraexaminer agreements were determined using kappa values. In the second part of this research, CBCT images of a 100 patients database from FOB-USP were examined. Two examiners assessed the internal anatomy of mandibular incisors in axial, sagittal and coronal reconstructions and classified the teeth according to the number of root canals. The prevalence of 2 root canals was related to the tooth location and patients gender and age. Percentage (%) was used to describe the collected data. Chi-square test (X2) was used to verify the association between variables (p <.05). Inter and intraexaminer agreements were determined using kappa values. The results of the first part of the research showed a high level of identification of anatomic patterns for all the methods (p <0.05). Considering Type I (1 regular root canal), CBCT images were better in comparison to PA (PA= 67%; CBCT= 98%, p <0.05). There was no significant difference between the 3 CBCT scanners. For Type Ia (1 oval root canal), there was a significant difference between PA and CBCT imaging using the NewTom unit only (PA= 44%, NewTom CBCT= 88%). No significant differences were found between the other 2 CBCT units and PA or between the 3 CBCT units. Considering Type III (2 root canals), there was no significant difference between the various methods. All the methods presented similar corrected answers index for Type III, which may suggest that PA with double exposure is sufficient for the identification of 2 root canals in mandibular incisors. The intraexaminer agreement was fair for PA (kappa=0.40 to 0.66) and good to very good for the CBCT units (kappa=0.62 to 0.85). The interexaminer agreement was poor to fair for PA (kappa=0.25 to 0.32) and good to very good for CBCT units (kappa=0.62 to 0.92). In the second part of this research, the total amount of 386 mandibular incisors was assessed in CBCT images (192 mandibular central incisors and 194 mandibular lateral incisors). The overall prevalence of 2 root canals was 16.5%, and 13% in mandibular central incisors and 20% in mandibular lateral incisors (p >0.05). There was no significant difference between gender and age. The intra and interexaminer agreement was fair (kappa intra=0.60; kappa inter=0.57). For the first part of this research, it is possible to conclude that the identification of the number of root canals was effective using all the methods. However, limitations were found in differentiating the shape of the root canal. As a conclusion of the second part of this research, the presence of 2 root canals in mandibular incisors of Bauru region patients is approximately 20% and it does not depend on tooth location or patients gender and age.

O propósito deste estudo foi avaliar a acurácia do exame de tomografia computadorizada multislice (TCM) e tomografia computadorizada por feixe cônico (TCFC) na identificação de lesões simuladas em mandíbula, em situações com e sem a presença de artefatos metálicos, em diversos protocolos de observação. Foram realizados exames de TCM e TCFC de mandíbulas secas, nas quais foram executadas perfurações simulando lesões. As imagens foram realizadas em dois momentos: na presença e na ausência de restaurações dentárias metálicas. Dois observadores, previamente calibrados, observaram as imagens avaliando-as quanto à presença ou ausência de lesão, número de lojas das lesões e a existência ou não de invasão medular. Os mesmos utilizaram programas de manipulação de imagens instalados em estações de trabalho independentes, para reconstruir as imagens nos seguintes protocolos de avaliação: axial, sagital + coronal, 3D, conjunto (axial+sagital+coronal+3D) e parassagital. A sensibilidade e especificidade (validade) da tomografia computadorizada multislice (64 cortes) (TCM) e da tomografia computadorizada por feixe cônico (TCFC) para diagnóstico de lesões ósseas (simuladas) em mandíbula, utilizando estação de trabalho independente foram demonstradas à medida que os valores encontrados foram superiores a 95% desde que com o protocolo de observação adequado. A influência de artefatos dentários metálicos foi pouco significativa na interpretação de lesões ósseas mandibulares, já que os valores de acurácia nas análises com e sem artefato foram bastante próximas. Os protocolos com aquisição por TCM sofreram mais influência dos artefatos do que os adquiridos por TCFC (valores sutilmente menores), exceto nas reconstruções em 3D, em que as originadas de aquisição por TCFC, apresentaram valores menores de acurácia. O melhor protocolo de pós processamento para interpretação de lesões ósseas simuladas foi o denominado RMP+3D. O protocolo que apresentou os piores resultados foi o que utilizou as reconstruções parassagitais. / The purpose of this study was to evaluate the accuracy of multislice computed tomography (MSCT) and cone-beam computed tomography (CBCT) in identification of simulated mandibular lesions in situations with and without metallic artifacts in several observation protocols. MSCT and CBCT examinations were performed in dry mandibles, in which holes were performed simulating lesions. The images were taken in two stages: in the presence and absence of metal dental restorations. Two observers, previously calibrated, observed the images by evaluating the images for the presence or absence of lesion, loci number and whether there were medullary invasion. Observers used image manipulation softwares, installed on independent workstations, to reconstruct the images in the following evaluation protocols: axial, sagittal + coronal, 3D, sets (axial+ coronal + sagittal + 3 D) and parasagittal. The sensitivity and specificity of MSCT (64 slices) and CBCT for diagnosis of simulated bone lesions in the mandible, using independent workstation were demonstrated as the values obtained were greater than 95% using the appropriate observation protocol. The influence of dental metallic artifacts was negligible in the interpretation of mandibular bone lesions, since the values of accuracy in the analysis with and without artifact were quite close. The images acquired with MCT suffered more influence of artifacts than the protocols acquired by TCFC, although the values were all high and quite close. Except for 3D reconstructions, which originated from the CBCT, showed the lowest accuracy. The best protocol for post-processing and interpretation of simulated bone lesions was called RMP +3 D. The protocol that showed the worst results was parasagital.

La tomographie par rayons X multi-énergétiques (MECT) permet d'obtenir plus d'information concernant la structure interne de l'objet par rapport au scanner CT classique. Un de ses intérêts dans l’imagerie médicale est d'obtenir les images de fractions d’eau et d’os. Dans l'état de l'art, les intérêts de MECT n'est pas encore découvert largement. Les approches de reconstruction existantes sont très limitées dans leurs performances. L'objectif principal de ce travail est de proposer des approches de reconstruction de haute qualité qui pourront être utilisés dans la MECT afin d’améliorer la qualité d’imagerie.Ce travail propose deux approches de reconstruction bayésiennes. La première est adaptée au système avec un détecteur discriminant en énergie. Dans cette approche, nous considérons que les polychromaticités de faisceaux sont négligeables. En utilisant le modèle linéaire de la variance et la méthode d'estimation maximum à postériori (MAP), l'approche que nous avons proposé permets de prendre en compte les différents niveaux de bruit présentés sur les mesures multi-énergétiques. Les résultats des simulations montrent que, dans l'imagerie médicale, les mesures biénergies sont suffisantes pour obtenir les fractions de l'eau et de l'os en utilisant l'approche proposée. Des mesures à la troisième énergie est nécessaire uniquement lorsque l'objet contient des matériaux lourdes. Par exemple, l’acier et l'iode. La deuxième approche est proposée pour les systèmes où les mesures multi-énergétiques sont obtenues avec des faisceaux polychromatiques. C'est effectivement la plupart des cas dans l'état actuel du practice. Cette approche est basée sur un modèle direct non-linéaire et un modèle bruit gaussien où la variance est inconnue. En utilisant l’inférence bayésienne, les fractions de matériaux de base et de la variance d'observation pourraient être estimées à l'aide de l'estimateur conjoint de MAP. Sous réserve à un modèle a priori Dirac attribué à la variance, le problème d'estimation conjointe est transformé en un problème d'optimisation avec une fonction du coût non-quadratique. Pour le résoudre, l'utilisation d'un algorithme de gradient conjugué non-linéaire avec le pas de descente quasi-optimale est proposée.La performance de l'approche proposée est analysée avec des données simulées et expérimentales. Les résultats montrent que l'approche proposée est robuste au bruit et aux matériaux. Par rapport aux approches existantes, l'approche proposée présente des avantages sur la qualité de reconstruction. / Multi-Energy Computed Tomography (MECT) makes it possible to get multiple fractions of basis materials without segmentation. In medical application, one is the soft-tissue equivalent water fraction and the other is the hard-matter equivalent bone fraction. Practical MECT measurements are usually obtained with polychromatic X-ray beams. Existing reconstruction approaches based on linear forward models without counting the beam polychromaticity fail to estimate the correct decomposition fractions and result in Beam-Hardening Artifacts (BHA). The existing BHA correction approaches either need to refer to calibration measurements or suffer from the noise amplification caused by the negative-log pre-processing and the water and bone separation problem. To overcome these problems, statistical DECT reconstruction approaches based on non-linear forward models counting the beam polychromaticity show great potential for giving accurate fraction images.This work proposes a full-spectral Bayesian reconstruction approach which allows the reconstruction of high quality fraction images from ordinary polychromatic measurements. This approach is based on a Gaussian noise model with unknown variance assigned directly to the projections without taking negative-log. Referring to Bayesian inferences, the decomposition fractions and observation variance are estimated by using the joint Maximum A Posteriori (MAP) estimation method. Subject to an adaptive prior model assigned to the variance, the joint estimation problem is then simplified into a single estimation problem. It transforms the joint MAP estimation problem into a minimization problem with a non-quadratic cost function. To solve it, the use of a monotone Conjugate Gradient (CG) algorithm with suboptimal descent steps is proposed.The performances of the proposed approach are analyzed with both simulated and experimental data. The results show that the proposed Bayesian approach is robust to noise and materials. It is also necessary to have the accurate spectrum information about the source-detector system. When dealing with experimental data, the spectrum can be predicted by a Monte Carlo simulator. For a variety of materials, less than 5% estimation errors are observed on their average decomposition fractions.The proposed approach is a statistical reconstruction approach based on a non-linear forward model counting the full beam polychromaticity and applied directly to the projections without taking negative-log. Compared to the approaches based on linear forward models and the BHA correction approaches, it has advantages in noise robustness and reconstruction accuracy.