Reliable information about the micro-processes during filtration and dewatering of filter cakes allows more accurate statements about process development and design in any industrial application with solid-liquid separation units. Distributed particle properties such as shape, size, and material influence the formation of the porous network structure, which can show considerable local fluctuations in vertical and horizontal alignment in the cake forming apparatus. The present work relates to a wide range of particle sizes and particle shapes and presents their effects on integral, but preferably local, structural parameters of cake filtration. Current models for the relationship between particle properties and resulting porous structure remain inaccurate. Therefore, the central question focus on the model-based correlation between obtained tomographic 3D information and characteristic cake and process parameters. In combination with X-ray computed tomography and microscopy (ZEISS Xradia 510), data acquisition of the structural build-up of filter cakes is possible on a small scale (filter area 0.2 cm²) and a conventional laboratory scale (filter area 20 cm², VDI 2762 pressure nutsch). Thereby, the work focuses on structural parameters at the local level before, during, and after cake dewatering, such as porosity, coordination number, three-phase contact angle, characteristics of pores and isolated liquid regions, the liquid load of individual particles, tortuosity, and capillary length, and the corresponding spatial distributions. Seven different particle systems in the range of 20 and 500 µm, suspended in aqueous solutions with additives for contrast enhancement, served as raw materials for the filter cake formation. Image data processing from 16-bit greyscale images with a resolution of 2 to 4 µm/voxel edge length includes various operations with a two-stage segmentation to identify air, solid particles, and liquid phase, resulting in a machine learning-based automated approach. Subsequent modeling and correlation of measured parameters rely on experimentally verified quantities from mercury porosimetry, laser diffraction, dynamic image analysis, static and dynamic droplet contour analysis, as well as filtration and capillary pressure tests according to VDI guidelines. The tomography measurements provide microscopic information about the porous system, quantified using characteristic key parameters and distribution functions. By studying the cake structure concerning the local distribution of particle size and shape and the resulting porosity, segregation effects can be avoided by increasing the feed concentration of particles, whereby swarm inhibition of particles in the initial suspension strongly hinders or completely suppresses layer formation in the cake according to distributed particle properties (Publication A). In the subsequent dewatering of the filter cake to the irreducible saturation, the measurement of the local coordination number as well as the remaining liquid volumes at the particle contacts allows the determination of a discrete liquid load distribution by correlation with the respective particle volume (Publication B). The determination of the capillary length - shortest capillary for single-phase pore flow and capillary of least resistance for multiphase pore flow - provides modeling approaches for the cake formation from publication A as well as the dewatering process from publication B (Publication C). The parameter sets obtained also help to transfer and extend existing, theoretical models of multiphase pore flow to the application example of filter cake dewatering (Publication D). At the microscopic level, the measurement of the three-phase contact angle at isolated liquid volumes within the porous matrix provides a deeper understanding of the macroscopic models from publications C and D (Publication E).:List of Figures List of Tables Notation 1 Introduction 2 Multiphase Processes in Porous Media 2.1 Cake Filtration and Single Phase Porous Media Flow 2.2 Cake Dewatering 2.2.1 Particle Surface Wettability 2.2.2 Capillarity in Porous Media 2.2.3 Static Capillary Pressure 2.2.4 Dynamic Capillary Pressure 3 Acquisition of 3D Information of Porous Media 3.1 Absorption and Scattering of X-rays 3.2 X-ray Microscopy 3.2.1 Image Acquisition 3.2.2 Image Reconstruction 3.2.3 Image Quality and Artifacts 3.3 Image Post-Processing 3.3.1 Image Enhancement 3.3.2 Segmentation and Thresholding 3.3.3 Processing Binary Images 3.4 Image Measurement 4 Materials and Methods 4.1 The Solid Phase 4.2 The Liquid Phase 4.3 Suspension Stability 4.4 Experimental Design and Down-Scale for Tomography Measurements 4.5 Experimental Characterization of Filtration and Dewatering Properties 4.5.1 Cake Filtration 4.5.2 Cake Dewatering (Capillary Pressure Measurements) 5 Conclusion and Outlook Literature Publications A to E Appendix

Asphalt joints are potentially weakest parts of every pavement. Despite of their importance, reliable tools for measuring their mechanical properties for design and performance assessments are still scarce. This is particularly true for cold joints when attaching a new hot pavement to a cold existing one as in case of large patches for pavement repair. In this study, three static fracture testing methods, i.e. indirect tensile test (IDT), direct tension test (DTT) and 4 point bending (4PB), were adapted and used for evaluating different laboratory made joints. The results suggested that joints with inclined interfaces and also the ones with combined interface treatments (preheated and sealed) seemed to show more promising behaviors than the vertical and untreated joints. It was also confirmed that compacting from the hot side towards the joint improved the joint properties due to imposing a different flow pattern as compared to the frequent compaction methods. The latter finding highlighted the importance of asphalt particle rearrangements and flow during the compaction phase as a very little known subject in asphalt industry. Studies on compaction are of special practical importance since they may also contribute to reducing the possibility of over-compaction and aggregate crushing. Therefore, in this study, a new test method, i.e. Flow Test (FT), was developed to simulate the material flow during compaction. Initially, asphalt materials were substituted by geometrically simple model materials to lower the level of complexity for checking the feasibility of the test method as well as modeling purposes. X-ray radiography images were also used for capturing the flow patterns during the test. Results of the FT on model materials showed the capability of the test method to clearly distinguish between specimens with different characteristics. In addition, a simple discrete element model was applied for a better understanding of the test results as a basis for further improvements when studying real mixtures. Then, real mixtures were prepared and tested under the same FT configuration and the results were found to support the findings from the feasibility tests. The test method also showed its potential for capturing flow pattern differences among different mixtures even without using the X-ray. Therefore, the FT was improved as an attempt towards developing a systematic workability test method focusing on the flow of particles at early stages of compaction and was called the Compaction Flow Test (CFT). The CFT was used for testing mixtures with different characteristics to identify the parameters with highest impact on the asphalt particle movements under compaction forces. X-ray investigations during the CFT underlined the reliability of the CFT results. In addition, simple discrete element models were successfully generated to justify some of the CFT results. / <p>QC 20151104</p>

This project is in collaboration with the Hydro-Quebec Research Institute (IREQ) and the Canadian Centre for Electron Microscopy (CCEM) on the mechanical test performance of miniature-scale micro-tensile specimens. The objective of the thesis project is to create an efficient and reliable fabrication route for producing micro-tensile specimens and to validate the accuracy of a newly custom-built micro-tensile bench at IREQ. The fabrication techniques developed and outlined in this thesis use the underlying fundamental physical mechanisms of secondary electron microscopy (SEM), focused-ion beam (FIB), and the femtosecond (fs)-laser machining for producing optimal quality micro-tensile specimens. The mechanical testing of the specimens is geared towards studying the localized deformation occurring in the microstructure when the size of the specimen only limits a number of grains and grain boundaries in order to target the specific detailed measurement of the mechanical behaviour of individual grains and interfaces. The goal for creating an optimal fabrication route for micro-tensile specimens is to carry out micro-mechanical testing of the primary turbine steels of 415 martensitic stainless steel used in the manufacture of Francis turbine components at Hydro-Quebec. The mechanical testing of single phase and interphase interface 415 steel micro-tensile specimens are considered building blocks to developing digital twin models of the steel microstructure. The experimental data from the mechanical tests would be fed into the crystal plasticity finite element models (CPFEM) that are currently being developed by researchers at IREQ. With the development of digital twin models, engineers at IREQ would be able to predict crack initiation at the microstructure level (prior to crack propagation into macro-scale cracks) by observing the evolution of the grain’s crystallographic orientation and morphology, as well as deformation mechanisms such as martensite formation and twinning produced from localized induced strains in the microstructure. In addition, self-organized dislocation processes such as dislocation nucleation and dislocation escape through the free surface can also be studied using the CPFEM models for size-limited mechanical deformation behaviour of miniature-scale mechanical test specimens. The fabrication routes studied in this thesis project use the combination of the fs-laser and plasma focused ion beam (PFIB) to machine the micro-tensile specimens. (100) single crystal copper was the ideal material chosen to validate the accuracy of the micro-tensile bench and quality of the fs-laser-machined tensile specimens, due to its ductile nature and well-characterized properties studied in literature. A mechanical size effect was studied for single crystal copper specimens with different gauge thicknesses. It was observed from the micro-tension testing that the strength of the specimens increased with decreasing gauge thickness occurring in the size-limited tensile gauges. In addition, it was determined there was negligible differences in the size effect seen between the PFIB-machined copper micro-tensile specimens and the fs-laser-machined micro-tensile specimens, demonstrating that the fs-laser is a reliable machining route for the micro-tensile specimens. X-ray computed tomography was used to validate the correct geometry of the machined gauge section produced from an innovative gauge thinning method adopted from IREQ’s research collaborator, Dr. Robert Wheeler. As well, finite-element analysis (FEA) was performed to determine the deformation behaviour under both linear-elastic and non-linear elastoplastic conditions of (100) copper and 415 steel models simulated in pure tension, prior to the fabrication of the micro-tensile specimens, respectively. Furthermore, significant progress has been made towards targeting martensite grains in the 415-steel microstructure using electron backscattered diffraction (EBSD) analysis to produce single crystal and interphase interface micro-tensile specimens. A workflow towards grain targeting using EBSD analysis has been developed, as well as for the relocation of grains using reference fiducial marks for future fabrication of the single crystal and interphase interface 415 micro-tensile specimens. / Thesis / Master of Applied Science (MASc) / Hydro-Quebec is an energy utilities company that operates the design of Francis hydro-turbines to supply hydroelectric power across the province of Quebec. The hydro-turbines have an expected service life of 70 years. Unfortunately, the turbines can get replaced by new ones prior to reaching half of its service life, due to the development of severe fatigue crack growth in the primary components of the turbines. A solution proposed by the researchers at the Hydro-Quebec Research Institute (IREQ) is to determine a linkage between the turbine’s steel’s microstructure and the mechanical behaviour of the turbine steels. Deformation of the material starts at the microstructure level, where dislocations glide through the material lattice, causing both reversible (elastic) and irreversible (plastic) deformation. Therefore, a solution was proposed by the researchers at IREQ to create computational models of the steel microstructure to predict the deformation of the steel’s microstructure. Being able to predict the deformation mechanisms through the simulation models of the microstructures allows for engineers at Hydro-Quebec to schedule regular maintenance of the turbines more efficiently and provide metallurgists the knowledge on what is occurring at the microstructure level and what can be done to improve the chemical and physical composition of the steel. To develop the digital twin models, experimental data must be collected through mechanical testing of miniature mechanical test specimens of the turbine steels. The mechanical properties of the single phases and interphase interface specimens are fed into the models as building blocks to building a microstructure map of the turbine steels. Micro-tension testing of micro-tensile specimen provides direct information about the material’s mechanical properties. In this work, a reliable and efficient fabrication route for micro-tensile specimens was developed for the purpose of extracting mechanical properties of single phase and interphase interface turbine steel specimens using focused ion beam (FIB) and femtosecond laser machining.

In this contribution, a methodology for particle tracing based on computed tomography and digital image processing is presented. It enables the tracing of particles in opaque structures using laboratory X-ray microcomputed tomography (μCT) systems that are not capable of time-resolved particle tracking. Through the development, it becomes apparent that an X-ray source with a cone beam geometry and the ability to perform fast, dynamic scans is a prerequisite for generating parabolic motion artefacts. Moreover, experimental tests are used to acquire data from simple particle sedimentations as well as from self-developed filter structures based on deep bed filtration. These experiments confirm that the particle position is located at the apex of the motion artefacts. Following the data assessment, multiple options for the particle coordinate extraction are discussed, and strategies thoroughly examined. A combination of random sample consensus (RANSAC) and the least squares method proves to be the most useful for determining the particle position. Besides, the developed methodology is validated using artificially generated data in which the motion artefact parameters of size, spatial orientation, and curvature, as well as noise, are varied. Supplementary, data is analysed manually in order to draw a comparison. In addition, to the presentation and discussion of the application of the methodology, a comparison with an artificial neural network (ANN) and the advantages and disadvantages of both methods are discussed. Finally, a first comparison of an extracted particle trace with a flow simulation through the complex structure is carried out, which shows that the particle trace follows the flow.:Table of Contents List of Figures ............................................................................................................................. i List of Tables ............................................................................................................................. vi List of Formula Symbols ......................................................................................................... vii List of Abbreviations ................................................................................................................. x 1 Introduction ....................................................................................................................... 1 2 Fundamentals .................................................................................................................... 5 2.1 Methods for Particle Tracking and Tracing .............................................................. 5 2.2 Computed Tomography .......................................................................................... 10 2.2.1 Tomography Design and Functional Principle ................................................ 10 2.2.2 Data Reconstruction ......................................................................................... 15 2.3 Digital Image Processing .......................................................................................... 18 3 Material............................................................................................................................. 30 3.1 Laboratory X-ray Tomography System TomoTU ................................................... 30 3.2 Experimental Setup .................................................................................................. 33 3.3 Choice of Particles and Medium ............................................................................. 34 4 Method development ...................................................................................................... 36 4.1 Characterisation of the Motion Artefacts ............................................................... 38 4.2 Method Consideration ............................................................................................. 45 4.3 Pre-processing .......................................................................................................... 46 4.4 Combination of Random Sample Consensus and Least Squares Method.......... 48 4.5 Multiple Particle Tracing .......................................................................................... 51 4.6 Coordinate Processing ............................................................................................. 53 4.7 Method Validation .................................................................................................... 53 5 Results and Discussion .................................................................................................... 59 5.1 Evaluation experimental data ................................................................................. 59 5.2 Comparison with Computational Fluid Dynamics (CFD) ....................................... 68 5.3 Comparison of Artificial Neural Networks with the Developed Classical Digital Image Processing Approach ............................................................................................... 70 6 Summary, Conclusion and Outlook ............................................................................... 74 7 References ........................................................................................................................ 76 / Die vorliegende Arbeit stellt eine auf Computertomografie und digitaler Bildverarbeitung basierte Methodik für die Partikelverfolgung dar. Diese ermöglicht es, mittels Labor- Microcomputertomografie (μCT) Anlagen, welche nicht dazu in der Lage sind, zeitaufgelöste Partikelverfolgung zu realisieren, Partikel in opaken Strukturen zu verfolgen. Durch die Methodenentwicklung ergibt sich, dass eine Röntgenquelle mit Kegelstrahlgeometrie sowie die Durchführungsmöglichkeit von schnellen, dynamischen Scans Voraussetzungen sind, um parabelförmige Bewegungsartefakte zu erzeugen. Dafür werden durch experimentelle Untersuchungen Daten erzeugt, die sowohl von einfachen Partikelsedimentationen als auch von eigens entwickelten Filterstrukturen, die sich an der Tiefenfiltration orientieren, abgeleitet werden. Diese Experimente bestätigen, dass sich die Partikelposition am Scheitelpunkt der Bewegungsartefakte befindet. Auf Grundlage der ersten Messungen werden verschiedene Möglichkeiten für die Partikelkoordinatenbestimmung diskutiert und Ansätze kritisch betrachtet. Dabei hat sich eine Kombination aus dem Random Sample Consensus (RANSAC) Algorithmus und der Methode der kleinsten Quadrate als am sinnvollsten für die Bestimmung der Partikelposition ergeben. Zudem wird die entwickelte Methodik anhand von künstlich erzeugten Daten validiert, bei welchen die Bewegungsartefakt-Parameter Größe, Raumorientierung und Krümmung sowie Rauschen variiert werden. Zusätzlich werden auch Daten manuell ausgewertet, um einen Vergleich ziehen zu können. Neben der Präsentation und Diskussion der Anwendung der Methodik wird außerdem ein Vergleich zu künstlichen neuronalen Netzen (KNNs) und die Vor- und Nachteile beider Methoden diskutiert. Abschließend wird ein erster Vergleich einer extrahierten Partikelspur mit einer Strömungssimulation durch die komplexe Struktur durchgeführt, welche zeigt, dass die Partikelspur der Strömung folgt.:Table of Contents List of Figures ............................................................................................................................. i List of Tables ............................................................................................................................. vi List of Formula Symbols ......................................................................................................... vii List of Abbreviations ................................................................................................................. x 1 Introduction ....................................................................................................................... 1 2 Fundamentals .................................................................................................................... 5 2.1 Methods for Particle Tracking and Tracing .............................................................. 5 2.2 Computed Tomography .......................................................................................... 10 2.2.1 Tomography Design and Functional Principle ................................................ 10 2.2.2 Data Reconstruction ......................................................................................... 15 2.3 Digital Image Processing .......................................................................................... 18 3 Material............................................................................................................................. 30 3.1 Laboratory X-ray Tomography System TomoTU ................................................... 30 3.2 Experimental Setup .................................................................................................. 33 3.3 Choice of Particles and Medium ............................................................................. 34 4 Method development ...................................................................................................... 36 4.1 Characterisation of the Motion Artefacts ............................................................... 38 4.2 Method Consideration ............................................................................................. 45 4.3 Pre-processing .......................................................................................................... 46 4.4 Combination of Random Sample Consensus and Least Squares Method.......... 48 4.5 Multiple Particle Tracing .......................................................................................... 51 4.6 Coordinate Processing ............................................................................................. 53 4.7 Method Validation .................................................................................................... 53 5 Results and Discussion .................................................................................................... 59 5.1 Evaluation experimental data ................................................................................. 59 5.2 Comparison with Computational Fluid Dynamics (CFD) ....................................... 68 5.3 Comparison of Artificial Neural Networks with the Developed Classical Digital Image Processing Approach ............................................................................................... 70 6 Summary, Conclusion and Outlook ............................................................................... 74 7 References ........................................................................................................................ 76

Introdução: A pressão crítica de fechamento da faringe (PCrit) é a pressão nasal em que há colapso da faringe. Conceitualmente a Pcrit reflete a contribuição anatômica na gênese da apnéia obstrutiva do sono (AOS). No entanto, a relação entre a PCrit e a anatomia das vias aéreas superiores (VAS) tem sido pouco estudada. A PCrit determinada durante o sono requer pesquisadores experientes durante a noite. A indução do sono com midazolam é usada na prática clínica para a realização de procedimentos ambulatoriais e poderia ser uma alternativa mais conveniente para se determinar a Pcrit. Porém, o midazolam pode provocar sedação além de simples indução do sono, reduzir a atividade muscular das VAS e aumentar a colapsabilidade quando comparado com o sono normal. Objetivos: 1. validar a determinação da PCrit durante o dia após a indução do sono com midazolam; 2. comparar a arquitetura do sono induzido com baixa dose de midazolam com o sono natural; 3. correlacionar a PCrit com a anatomia das VAS. Métodos: Homens com graus variados de sintomas sugestivos de AOS foram submetidos a polissonografia completa noturna, determinação da PCrit durante o sono natural e após a indução do sono com midazolam bem como tomografia computadorizada de cabeça e pescoço para avaliação das VAS. Resultados: Foram estudados 15 sujeitos com idade (média±DP) de 54 ± 10 anos, índice de massa corporal de 29,9 ± 3,9 kg/m2 e índice de apnéiahipopnéia (IAH) de 38 ± 22 (variação: 8-66 eventos/h). A indução do sono foi obtida em todos os sujeitos, utilizando doses mínimas de midazolam (mediana [intervalo interquartil]) (2,4 [2,0-4,4] mg). A PCrit durante o sono natural e induzido foram semelhantes (-0,82 ± 3,44 e -0,97 ± 3,21 cmH2O, P = 0,663) e se associaram (coeficiente de correlação intraclasse=0,92 (IC 95% 0,78-0,97 P<0,001). A distribuição das fases do sono durante sono natural e induzido foi similar, com excessão da fase 1 (10,5 ± 5,1% vs. 20,6 ± 8,1, respectivamente; P=0,001). A Pcrit determinada durante o sono natural e induzido se correlacionaram com o IAH (r=0,592, P=0,020 e r=0,576, P=0,025, respectivamente). Além disso tanto a Pcrit determinada por sono natural e induzido se correlacionaram com diversas variáveis tomográficas de VAS, incluindo a posição do osso hióide, ângulo da base do crânio e as áreas seccionais da velofaringe e hipofaringe (r variando de 0,577 a 0,686, P<0,05). A regressão linear múltipla revelou que o IAH foi independentemente associado com a Pcrit durante sono induzido, circunferência da cintura e idade (r2 = 0,785, P = 0,001). Conclusão: A PCrit determinada durante o dia com indução do sono é semelhante à determinada durante o sono natural e é um método alternativo promissor para determinar a PCrit. O sono induzido por doses baixas de midazolan promove um sono similar ao sono natural. A Pcrit determinada tanto durante o sono natural e induzido correlaciona-se com várias características anatômicas das VAS / Introduction: The pharyngeal critical closing pressure (Pcrit) is the nasal pressure at which the airway collapses. Pcrit is thought to reflect the anatomical contribution to the genesis of obstructive sleep apnea (OSA). However, the relationship between Pcrit and upper airway anatomy has been poorly investigated. Pcrit determined during sleep requires experienced investigators at night. Sleep induction with midazolam is frequently used in clinical practice during ambulatory procedure and could be a more convenient alternative to assess Pcrit. On the other hand, midazolam could induce sedation rather than sleep, decrease upper airway muscle activity and increase collapsibility compared with natural sleep. Objectives: 1. validate Pcrit determination during the day after sleep induction with midazolam; 2. compare the sleep architecture of induced sleep after low doses of midazolam with natural sleep; 3.correlate Pcrit with upper airway anatomy. Methods: Men with different severity of OSA symptoms underwent baseline full polysomnography, Pcrit determination during natural sleep and after sleep induction with midazolam and head and neck computed tomography. RESULTS: Fifteen men aged (mean±SD) 54±10ys, body mass index=29.9 ± 3.9 Kg/m2 and apnea hypopnea index=38±22, range: 8-66 events/h were studied. Sleep induction was obtained with minimum doses of midazolam (median[interquartile range] (2.4 [2.0-4.4] mg). Sleep phase distribution during natural and induced sleep was similar, except for stage 1 (10.5 ± 5.1% vs. 20.6 ± 8.1, respectively; P=0.001). Natural and induced sleep Pcrit were similar (-0.82 ± 3.44 and -0.97 ± 3.21 cmH2O, P=0.663) and closely associated (intraclass correlation coefficient=0.92 (95%CI 0.78-0.97, P<0.001). Natural and sleep induced Pcrit correlated with AHI (r=0.592, P=0.020; r=0.576, P=0.025, respectively). Pcrit determined both during natural and induced sleep were significantly associated with several tomographic variables, including hyoid position, cranial base angle and cross sectional areas of the velopharynx and hypopharynx (r range: 0.577 to 0.686, P<0.05). Multiple linear regression revealed that AHI was independently associated with induced sleep Pcrit, waist circumference and age (r2=0.785, P=0.001). Conclusion: Pcrit determined during the day with sleep induction is similar to natural sleep and is a promising alternative method to determine Pcrit. Sleep induction with small doses of midazolam promoted sleep similar to natural sleep. Pcrit determined both during natural and induced sleep correlates with several anatomical characteristics of the upper airway

Introdução: A utilização da posição prona melhora significativamente a oxigenação de pacientes com síndrome do desconforto respiratório agudo (SDRA). Estudos prévios sugerem que o recrutamento das regiões pulmonares colapsadas e pobremente aeradas é um dos possíveis mecanismos responsáveis pela melhora da oxigenação, no entanto, os mesmos ainda não foram comprovados. Objetivos: Quantificar a distribuição regional da aeração e da perfusão pulmonar, em ambas as posições prona e supina, através da tomografia de impedância elétrica (TIE) e da tomografia computadorizada multislice (TC), correlacionando-as com as respectivas trocas gasosas. Métodos: Foram estudados 21 suínos, da raça Ladrasse anestesiados e em ventilação mecânica controlada. Os animais foram divididos em dois grupos, de acordo com o método de imagem. 13 animais foram estudados com a TIE (grupo TIE) e 8 animais foram estudados com a TC (grupo TC). Após a indução do modelo de lesão pulmonar (infusão intermitente de solução salina e ventilação lesiva por 3 horas), os animais foram submetidos a uma manobra de recrutamento alveolar máxima (MR) seguida por uma manobra de titulação da PEEP (MTP), realizada em passos decrementais de 2 em 2 cmH2O PEEP. Onze animais (7 no grupo TIE e 4 no grupo TC) foram randomizados para iniciar o estudo na posição supina, seguida de uma segunda MR e MTP na posição prona. Dez animais (6 no grupo TIE e 4 no grupo TC) receberam as manobras na ordem inversa. Para o estudo da perfusão foram adicionados mais sete animais (2 no grupo TIE e 5 no grupo TC) que foram submetidos à injeção rápida de solução salina hipertônica e/ou de contraste iodado respectivamente. Resultados: Não foram encontradas diferenças significativas na quantidade de tecido pulmonar colapsado e hiperdistendido, entre as posições estudadas em ambos os grupos TIE e TC (p= 0.06). Entretanto, as trocas gasosas foram consistentemente melhores durante a posição prona (p<0.05), com shunt pulmonar significativamente menor (<55%, p<0.001) para níveis equivalentes de PEEP entre ambas as posições. A análise da perfusão pulmonar evidenciou a presença de grande fluxo sanguíneo pulmonar (2x mais perfusão específica) nas áreas de colapso pulmonar durante a posição supina, assim como a ausência de maiores efeitos gravitacionais na distribuição da perfusão entre as posturas. Conclusão: A análise quantitativa da TIE e da TC evidenciou que, para a mesma quantidade de tecido pulmonar colapsado, a oxigenação arterial foi sempre mais comprometida durante a posição supina, com aumento significativo do shunt pulmonar e com uma região de colapso pulmonar mais perfundida, sugerindo que a posição prona melhora a oxigenação e a relação ventilação perfusão, mas não atenua os efeitos gravitacionais sobre o parênquima pulmonar, o que não confirma os possíveis efeitos protetores associados à posição prona / Introduction: Prone position has been shown to consistently improve oxygenation in patients with acute respiratory distress syndrome (ARDS). Previous studies suggested some improvement in lung recruitment or a better ventilation of poorly aerated areas as possible mechanisms for such oxygenation benefits. Objective: To quantify the regional distribution of aeration (collapse and hyperdistend lung tissue) and lung perfusion by Computer Tomography (CT) and electrical impedance tomography in supine and prone positions and to correlate them with pulmonary gas exchange. Methods: We studied 21 anesthetized Landrace pigs under controlled mechanical ventilation. These animals were divided in two groups: Thirteen (13) animals in the EIT group and eight (8) in the CT group. After lung injury (saline lavage + VILI during 3 hours), animals were recruited and submitted to two sequential PEEP trials, both consisting of decremental PEEP steps (2 cmH2O steps). Seven (n=7) animals in the EIT group and four (4) in the CT group were allocated to a PEEP trial under supine position, followed by a second PEEP trial in prone. Six (6) animals in the EIT group and four (4) in the CT group received PEEP trials in reverse order. Seven (7) additional animals were studied for lung perfusion distribution, by analyzing the first pass kinetics of hypertonic solution (2 animals - EIT group) and iodine contrast (5 animals - CT group). Results: No differences in the amount of collapsed and hyperdistended lung tissue were found between both postures (p= 0.12 vs. p = 0.41 respectively) in both the EIT and CT groups. However, the gas exchange was consistently better (p <0.05), with much lower (55% lower) pulmonary shunt during prone position (p=0.001), at equivalent PEEP levels in both groups. The perfusion studies confirmed a higher perfusion ( 2 times increment in specific perfusion) of the atelectatic lung tissue in supine position, without majors gravitational effects between both positions. Conclusions: The quantitative analysis of EIT and CT showed that for the same amount of collapsed lung tissue, the PaO2 was always lower in supine position, with higher pulmonary shunt and higher perfusion of the collapsed lung areas. We could not demonstrate any lung protective effect associated with prone positioning. Thus, these results suggest that prone position improves oxygenation and V/Q imbalances, but it does not attenuate the effects of gravity on the lung

Introdução: No momento, pouco se conhece sobre o desempenho diagnóstico da angiotomografia de coronárias com 64 colunas de detectores (Angio-TC 64) frente aos achados do ultrassom intracoronário com técnica de histologia virtual (USIC-HV). Este estudo compara a capacidade diagnóstica de ambos os métodos para a avaliação das dimensões vasculares e para avaliação da composição da placa aterosclerótica através da análise de toda extensão dos três vasos epicárdicos principais. Métodos e Resultados: Um total de 21 pacientes com diagnóstico de doença arterial coronária obstrutiva foi incluído neste estudo. Angio-TC 64 foi realizada em todos os pacientes antes da realização do USIC-HV, com intervalo<72horas entre os dois exames. No total, 70 vasos foram avaliados (3,3 vasos por paciente) e divididos em 641 subsegmentos de 4 mm de extensão cada. Um total de 5.972 cortes seccionais de USIC-HV e 5.233 cortes seccionais da Angio-TC 64 foram analisados. As medidas de área luminar, área do vaso, área da parede arterial e carga de placa à Angio-TC 64 e ao USIC-HV foram significativamente correlacionadas (r-Spearman: 0,81; 0,78; 0,55 e 0,49; respectivamente - p<0,001 para todas as correlações). A Angio-TC 64 subestimou a área luminar ao USIC-HV (em mediana: 0,4mm2 ,variando entre -5,6 mm2 e 10,2 mm2). A Angio-TC 64 superestimou a área do vaso, a área da parede arterial (placa+média) e a carga de placa (em mediana: 3,0 mm2; 3,2 mm2 e 13,9%, respectivamente). O aumento da densidade média da placa à Angio-TC 64 foi significativamente associado com o aumento da contribuição percentual dos componentes cálcio denso e núcleo necrótico ao USIC-HV. O aumento da densidade média da placa à Angio-TC 64 foi significativamente associado com a diminuição da contribuição percentual do componente fibro-lipídico ao USIC-HV. Parâmetros de qualidade da imagem (atenuação luminar, ruído da atenuação luminar e relação sinal ruído) influenciaram significativamente os resultados da Angio-TC 64. Conclusão: Nosso estudo demonstra que as imagens da Angio-TC 64 se correlacionam significativamente com as imagens do USIC-HV. Estes achados indicam que a Angio-TC 64 pode ser uma ferramenta útil para a avaliação quantitativa da luz arterial e da placa aterosclerótica; bem como para a avaliação da composição da placa aterosclerótica in vivo / Background: Currently, little is known about the comparative diagnostic performance for coronary assessment of 64-slice multidetector computed tomography (64-MDCT) versus virtual histology intravascular ultrasound technique (VH-IVUS). The present study compares the diagnostic ability of both methods for the evaluation of coronary lumen and vessel wall dimensions as well as plaque composition in a three-vessel whole-artery analysis protocol. Methods and Results: A total of 21 patients with diagnosed obstructive coronary artery disease was included. 64-MDCT was performed within 72 hours before the VH-IVUS examination. Overall, 70 vessels were imaged (3.3 vessels per patient), and divided into 641 subsegments of 4 mm each. A total of 5,972 VH-IVUS cross-sections and 5,233 64-MDCT cross-sections were analyzed. 64-MDCT and VH-IVUS measurements for luminar area, vessel area, arterial wall area (plaque plus media area) and plaque burden were significantly correlated (r-Spearman: 0.81; 0.78; 0.55 e 0.49; respectively - p<0,001 for all correlations). 64-MDCT underestimated VH-IVUS measurements for luminar area (median: 0.4mm2, range: -5.6 mm2 to 10,2 mm2). 64-MDCT overestimated VH-IVUS measurements for vessel area, arterial wall area, and plaque burden (median: 3.0 mm2; 3.2 mm2 e 13.9%, respectively). Increasing plaque density at 64-MDCT was significantly associated with increasing dense calcium and necrotic core percent composition at VH-IVUS. Increasing plaque density at 64-MDCT was significantly associated with decreasing fibrofatty percent composition, and decreasing necrosis-to-calcium ratio at VH-IVUS. Image quality parameters (i.e. lumen attenuation, image noise, signal-to-noise ratio) significantly influenced the results of 64-MDCT. Conclusion: Our study shows that 64-MDCT imaging significantly correlates with VH-IVUS. These findings indicate that 64-MDCT may be a useful non-invasive tool for quantitative evaluation of lumen and plaque parameters, as well as for the in vivo assessment of atherosclerotic plaque composition

Introdução: A utilização da posição prona melhora significativamente a oxigenação de pacientes com síndrome do desconforto respiratório agudo (SDRA). Estudos prévios sugerem que o recrutamento das regiões pulmonares colapsadas e pobremente aeradas é um dos possíveis mecanismos responsáveis pela melhora da oxigenação, no entanto, os mesmos ainda não foram comprovados. Objetivos: Quantificar a distribuição regional da aeração e da perfusão pulmonar, em ambas as posições prona e supina, através da tomografia de impedância elétrica (TIE) e da tomografia computadorizada multislice (TC), correlacionando-as com as respectivas trocas gasosas. Métodos: Foram estudados 21 suínos, da raça Ladrasse anestesiados e em ventilação mecânica controlada. Os animais foram divididos em dois grupos, de acordo com o método de imagem. 13 animais foram estudados com a TIE (grupo TIE) e 8 animais foram estudados com a TC (grupo TC). Após a indução do modelo de lesão pulmonar (infusão intermitente de solução salina e ventilação lesiva por 3 horas), os animais foram submetidos a uma manobra de recrutamento alveolar máxima (MR) seguida por uma manobra de titulação da PEEP (MTP), realizada em passos decrementais de 2 em 2 cmH2O PEEP. Onze animais (7 no grupo TIE e 4 no grupo TC) foram randomizados para iniciar o estudo na posição supina, seguida de uma segunda MR e MTP na posição prona. Dez animais (6 no grupo TIE e 4 no grupo TC) receberam as manobras na ordem inversa. Para o estudo da perfusão foram adicionados mais sete animais (2 no grupo TIE e 5 no grupo TC) que foram submetidos à injeção rápida de solução salina hipertônica e/ou de contraste iodado respectivamente. Resultados: Não foram encontradas diferenças significativas na quantidade de tecido pulmonar colapsado e hiperdistendido, entre as posições estudadas em ambos os grupos TIE e TC (p= 0.06). Entretanto, as trocas gasosas foram consistentemente melhores durante a posição prona (p<0.05), com shunt pulmonar significativamente menor (<55%, p<0.001) para níveis equivalentes de PEEP entre ambas as posições. A análise da perfusão pulmonar evidenciou a presença de grande fluxo sanguíneo pulmonar (2x mais perfusão específica) nas áreas de colapso pulmonar durante a posição supina, assim como a ausência de maiores efeitos gravitacionais na distribuição da perfusão entre as posturas. Conclusão: A análise quantitativa da TIE e da TC evidenciou que, para a mesma quantidade de tecido pulmonar colapsado, a oxigenação arterial foi sempre mais comprometida durante a posição supina, com aumento significativo do shunt pulmonar e com uma região de colapso pulmonar mais perfundida, sugerindo que a posição prona melhora a oxigenação e a relação ventilação perfusão, mas não atenua os efeitos gravitacionais sobre o parênquima pulmonar, o que não confirma os possíveis efeitos protetores associados à posição prona / Introduction: Prone position has been shown to consistently improve oxygenation in patients with acute respiratory distress syndrome (ARDS). Previous studies suggested some improvement in lung recruitment or a better ventilation of poorly aerated areas as possible mechanisms for such oxygenation benefits. Objective: To quantify the regional distribution of aeration (collapse and hyperdistend lung tissue) and lung perfusion by Computer Tomography (CT) and electrical impedance tomography in supine and prone positions and to correlate them with pulmonary gas exchange. Methods: We studied 21 anesthetized Landrace pigs under controlled mechanical ventilation. These animals were divided in two groups: Thirteen (13) animals in the EIT group and eight (8) in the CT group. After lung injury (saline lavage + VILI during 3 hours), animals were recruited and submitted to two sequential PEEP trials, both consisting of decremental PEEP steps (2 cmH2O steps). Seven (n=7) animals in the EIT group and four (4) in the CT group were allocated to a PEEP trial under supine position, followed by a second PEEP trial in prone. Six (6) animals in the EIT group and four (4) in the CT group received PEEP trials in reverse order. Seven (7) additional animals were studied for lung perfusion distribution, by analyzing the first pass kinetics of hypertonic solution (2 animals - EIT group) and iodine contrast (5 animals - CT group). Results: No differences in the amount of collapsed and hyperdistended lung tissue were found between both postures (p= 0.12 vs. p = 0.41 respectively) in both the EIT and CT groups. However, the gas exchange was consistently better (p <0.05), with much lower (55% lower) pulmonary shunt during prone position (p=0.001), at equivalent PEEP levels in both groups. The perfusion studies confirmed a higher perfusion ( 2 times increment in specific perfusion) of the atelectatic lung tissue in supine position, without majors gravitational effects between both positions. Conclusions: The quantitative analysis of EIT and CT showed that for the same amount of collapsed lung tissue, the PaO2 was always lower in supine position, with higher pulmonary shunt and higher perfusion of the collapsed lung areas. We could not demonstrate any lung protective effect associated with prone positioning. Thus, these results suggest that prone position improves oxygenation and V/Q imbalances, but it does not attenuate the effects of gravity on the lung

Avec les récents développements des mesures de champs, il est devenu possible deréaliser des expériences sur un échantillon de sol en effectuant des mesures simultanémentaux différentes échelles pertinentes : celle de l'échantillon, et celle des grains, et ceci en3D volumique. Ces progrès ont permis une meilleure compréhension des processus qui seproduisent dans le sol, en les observant au cœur même des échantillons.Cette thèse vise à caractériser les comportements de rétention d'eau et hydro-mécaniquedu sable d'Hostun non saturé à l'échelle du grain. Pour ce faire, des essais de rétentiond'eau et des essais triaxiaux ont été effectués, qui ont nécessité le développement d'outils detraitement d'image et des montages expérimentaux spécifiques. L'installation de tomographie à rayons X du Laboratoire 3SR de l'Université des Grenoble a été utilisée pour générer des images en 3D de la microstructure du sol non saturé. Une technique de trinarization a été mise en œuvre pour permettre la séparation des phases (les grains, l'eau et l'air) dans les images 3D, et d'effectuer différentes mesures à l'échelle du grain. La porosité et le degré de saturation ont été mesurées macroscopiquement en utilisant les images segmentées, et microscopiquement en utilisant un code de cartographie spécialement développé. Les mesures microscopiques ont été réalisées sur un volume élémentaire représentatif (VER), dont la taille a été déterminée en utilisant une analyse développée sur la base d'outils statistiques. Plusieurs analyses discrets ont été développées pour étudier l'évolution de la phase fluide (l'eau et l'air) avec le chargement (succion, et chargement mécanique). La corrélation d'images numérique 3D volumique a été mise en œuvre systématiquement pour ces deux types d'essais.Pour analyser le comportement de rétention d'eau du sable d'Hostun pendant les processus de séchage et de mouillage, une cellule spécifique a été développée. La succion a été appliquée en utilisant la technique de tensiomètre. L'échantillon (cylindrique 1x1cm) a été préparé en utilisant la technique de la pluviation sous l'eau et scanné avec une résolution de 7,5micron. La courbe de rétention d'eau, surface et les domaines hydriques ont été obtenus et le phénomène d'hystérésis et la cohésion du sol ont été analysés à l'échelle du grain.Pour le comportement hydro-mécanique des sols non saturés, des essais triaxiaux ontété effectué sous trois conditions: saturé drainé, non saturé drainé et non saturé non drainé. Un appareil triaxial que l'on peut placer à l'intérieur de la cabine du tomographe à rayons X a été développé afin de permettre le suivi des changements dans la microstructure du sable et la distribution de l'eau lorsque l'on soumet l'échantillon à chargement mécanique. L'échantillon (cylindre de dimensions hxd=2x1cm) a été préparé en utilisant la technique de la pluviation sous l'eau (complétement saturé au début des trois tests), amené ensuite au niveau de saturation désiré (dans le cas des essais non saturés), puis chargé par étapes successives avec scan à chaque étape avec une résolution de 13micron, jusqu'à arriver à 21% de déformation axiale pour les trois types d'essai. Les courbes contrainte-déformation, la réponse volumétrique et les déformations ont été obtenus et le profil de déformation, l'hétérogénéité de l'échantillon, la cohésion et la distribution de l'eau ont été analysés. / With the recent developments of full field measurements, it has become possible to do experiments on a soil specimen, and perform measurements at different scales. This progress has allowed greater understanding for all the processes that occur within the soil, and also created better environment to observe these mechanisms in 3D. This PhD work aims to characterize water retention and hydro-mechanical behaviours of unsaturated Hostun sand at the grain scale. To achieve this goal, water retention and triaxial tests were performed and a set of experimental and image processing tools was developed and used for the two series of tests. X-ray computed tomography set up available at 3SR Laboratory in Grenoble University was used to generate 3D images of the unsaturated soil microstructure non-destructively. A developed region growing systematic technique (trinarization technique) was used to allow the separation of the phases (i.e., grains, water and air) in the 3D images, and performing different measurements at the grain scale. Porosity and degree of saturationwere measured macroscopically using the trinarized images and microscopically using a developed mapping code. The microscopic measurements were performed over an REV, whose size was determined using a developed analysis based on statistical tools. Several microscopic-discrete analyses were developed to investigate the evolution of fluid phase (water and air) with loading (suction and mechanical loading). Digital Image Correlation was performed for the two tests.To analyze water retention behaviour of Hostun sand during drying and wetting processes, a pressure plate apparatus was developed. Suction was applied using tensiometry technique. The specimen (cylindrical 1x1cm) was prepared using water pluviation technique and scanned with a resolution 7.5micron. As a result, water retention curve, surface and domains were obtained and hysteresis phenomenon and soil cohesion were analyzed at the grain scale.For the mechanical behaviour of unsaturated soils, triaxial test were performed at three conditions: saturated drained, unsaturated drained and unsaturated undrained. A triaxial apparatus that can be placed inside the x-ray cabinet was developed in order to facilitate monitoring the changes in sand microstructure and water distribution when subjecting the soil specimen to loading. The specimen (cylinder of hxd=2x1cm) was prepared using water pluviation technique (completely saturated at the beginning of the three tests), loaded then scanned with a resolution 13micron, repeatedly till reaching 21% axial strain forthe three tests.Consequently, stress-strain curves, volumetric response and deformations measurements were obtained and deformation pattern, specimen heterogeneity, cohesion and water distribution were analyzed.

Introdução: A pressão crítica de fechamento da faringe (PCrit) é a pressão nasal em que há colapso da faringe. Conceitualmente a Pcrit reflete a contribuição anatômica na gênese da apnéia obstrutiva do sono (AOS). No entanto, a relação entre a PCrit e a anatomia das vias aéreas superiores (VAS) tem sido pouco estudada. A PCrit determinada durante o sono requer pesquisadores experientes durante a noite. A indução do sono com midazolam é usada na prática clínica para a realização de procedimentos ambulatoriais e poderia ser uma alternativa mais conveniente para se determinar a Pcrit. Porém, o midazolam pode provocar sedação além de simples indução do sono, reduzir a atividade muscular das VAS e aumentar a colapsabilidade quando comparado com o sono normal. Objetivos: 1. validar a determinação da PCrit durante o dia após a indução do sono com midazolam; 2. comparar a arquitetura do sono induzido com baixa dose de midazolam com o sono natural; 3. correlacionar a PCrit com a anatomia das VAS. Métodos: Homens com graus variados de sintomas sugestivos de AOS foram submetidos a polissonografia completa noturna, determinação da PCrit durante o sono natural e após a indução do sono com midazolam bem como tomografia computadorizada de cabeça e pescoço para avaliação das VAS. Resultados: Foram estudados 15 sujeitos com idade (média±DP) de 54 ± 10 anos, índice de massa corporal de 29,9 ± 3,9 kg/m2 e índice de apnéiahipopnéia (IAH) de 38 ± 22 (variação: 8-66 eventos/h). A indução do sono foi obtida em todos os sujeitos, utilizando doses mínimas de midazolam (mediana [intervalo interquartil]) (2,4 [2,0-4,4] mg). A PCrit durante o sono natural e induzido foram semelhantes (-0,82 ± 3,44 e -0,97 ± 3,21 cmH2O, P = 0,663) e se associaram (coeficiente de correlação intraclasse=0,92 (IC 95% 0,78-0,97 P<0,001). A distribuição das fases do sono durante sono natural e induzido foi similar, com excessão da fase 1 (10,5 ± 5,1% vs. 20,6 ± 8,1, respectivamente; P=0,001). A Pcrit determinada durante o sono natural e induzido se correlacionaram com o IAH (r=0,592, P=0,020 e r=0,576, P=0,025, respectivamente). Além disso tanto a Pcrit determinada por sono natural e induzido se correlacionaram com diversas variáveis tomográficas de VAS, incluindo a posição do osso hióide, ângulo da base do crânio e as áreas seccionais da velofaringe e hipofaringe (r variando de 0,577 a 0,686, P<0,05). A regressão linear múltipla revelou que o IAH foi independentemente associado com a Pcrit durante sono induzido, circunferência da cintura e idade (r2 = 0,785, P = 0,001). Conclusão: A PCrit determinada durante o dia com indução do sono é semelhante à determinada durante o sono natural e é um método alternativo promissor para determinar a PCrit. O sono induzido por doses baixas de midazolan promove um sono similar ao sono natural. A Pcrit determinada tanto durante o sono natural e induzido correlaciona-se com várias características anatômicas das VAS / Introduction: The pharyngeal critical closing pressure (Pcrit) is the nasal pressure at which the airway collapses. Pcrit is thought to reflect the anatomical contribution to the genesis of obstructive sleep apnea (OSA). However, the relationship between Pcrit and upper airway anatomy has been poorly investigated. Pcrit determined during sleep requires experienced investigators at night. Sleep induction with midazolam is frequently used in clinical practice during ambulatory procedure and could be a more convenient alternative to assess Pcrit. On the other hand, midazolam could induce sedation rather than sleep, decrease upper airway muscle activity and increase collapsibility compared with natural sleep. Objectives: 1. validate Pcrit determination during the day after sleep induction with midazolam; 2. compare the sleep architecture of induced sleep after low doses of midazolam with natural sleep; 3.correlate Pcrit with upper airway anatomy. Methods: Men with different severity of OSA symptoms underwent baseline full polysomnography, Pcrit determination during natural sleep and after sleep induction with midazolam and head and neck computed tomography. RESULTS: Fifteen men aged (mean±SD) 54±10ys, body mass index=29.9 ± 3.9 Kg/m2 and apnea hypopnea index=38±22, range: 8-66 events/h were studied. Sleep induction was obtained with minimum doses of midazolam (median[interquartile range] (2.4 [2.0-4.4] mg). Sleep phase distribution during natural and induced sleep was similar, except for stage 1 (10.5 ± 5.1% vs. 20.6 ± 8.1, respectively; P=0.001). Natural and induced sleep Pcrit were similar (-0.82 ± 3.44 and -0.97 ± 3.21 cmH2O, P=0.663) and closely associated (intraclass correlation coefficient=0.92 (95%CI 0.78-0.97, P<0.001). Natural and sleep induced Pcrit correlated with AHI (r=0.592, P=0.020; r=0.576, P=0.025, respectively). Pcrit determined both during natural and induced sleep were significantly associated with several tomographic variables, including hyoid position, cranial base angle and cross sectional areas of the velopharynx and hypopharynx (r range: 0.577 to 0.686, P<0.05). Multiple linear regression revealed that AHI was independently associated with induced sleep Pcrit, waist circumference and age (r2=0.785, P=0.001). Conclusion: Pcrit determined during the day with sleep induction is similar to natural sleep and is a promising alternative method to determine Pcrit. Sleep induction with small doses of midazolam promoted sleep similar to natural sleep. Pcrit determined both during natural and induced sleep correlates with several anatomical characteristics of the upper airway