Neste trabalho é apresentada uma nova modalidade de imagens elastográficas baseada na emissão acústica, quando um meio é submetido à radiação ultrassônica. Esta técnica está sendo denominada de Acustografia por Pulso/Emissão (APE). Características não-lineares da propagação acústica de ondas ultrassônicas, e a resposta mecânica vibracional, foram utilizadas como artifício para geração de imagens com frequências da ordem de quilohertz (kHz), a partir da excitação com ondas ultrassônicas na ordem de megahertz. Para produzir imagens com essa nova modalidade, simuladores de tecido biológicos foram construídos com diferenças de rigidez localizadas, e submetidos a uma radiação ultrassônica focalizada (MHz). O som emitido devido a interação da onda ultrassônica com a região de interesse era gravado e processado de modo a associar a cada pequena porção do tecido a um valor relacionado a rigidez para a formação da imagem. Os resultados mostraram que o método pode produzir imagens associadas às alterações viscoelásticas das amostras. A resolução espacial mostrou-se fortemente ligada a morfologia do campo acústico de excitação, sendo possível detectar estruturas da ordem de 0,25 mm isoladamente. A técnica de aquisição, desenvolvida e apresentada neste trabalho, é similar a técnica de vibroacustografia todavia, com uma instrumentação reduzida e com a possibilidade de obtenção de mais informações da estrutura do meio material, a partir dos fenômenos não lineares observados. Estudos pilotos de aplicação desta nova técnica e com a vibroacustografia, foram realizados e comparados para a avaliação de potenciais aplicações, por exemplo, na avaliação do sinal acústico diante de mudanças nas propriedades viscoelásticas do meio induzidas por mudança de temperatura; formação de imagens em meios com inclusões isoecogênicas e com rigidez ligeiramente diferentes; geração de imagens de estrutura óssea in vitro. / In this work is presented a new modality of elastography images based on the acoustic emission when a material medium was subjected to a ultrasound radiation.This tecnique Nonlinearity behavior of the acoustic wave propagation and the vibrational mechanical response were used to produce images from kilohertz frequencies when the sample was excited by ultrasound waves in megahertz. To produce images with this modality, tissue mimicking phantoms were made with stiffness in homogeneities and subjected a focused ultrasound radiation pulses. The sound emitted due the interaction of the ultrasound wave with the region of interest was recorded and processed in order to associate each small portion of the tissue to a value for image formation. The results showed that this method can produce images associated to the viscoelastic changes of the samples. The spatial resolution have showed strongly linked to the morphology of the excitation acoustic field, this way was possible to detect isolated structures in order of 0.25 mm. The acquisiton technique developed and presented in this work is similar to the vibroacoustography technique, however with reduced instrumentation setup and with the possibility to acquire further information about the structure of the material from the nonlinear phenomenal. Preliminary studies of this new technique and the vibroacoustography were made and compared to evaluate the potential applications, for example, in the evalution of the acoustic signal behavior due changes in the viscoelastic properties changes induced by temperature variations; image formation in the medium with lightly stiffness inclusions; generation of the images of bone structure in vitro.

Este trabalho teve como objetivo estudar um caso específico de epilepsia refratária causada por uma má formação no tecido cerebral, denominada displasia cortical focal (DCF). Por ser uma má formação no cérebro, suas consequências aparecem desde a infância, em que ela, a DCF, é a principal causadora das epilepsias de caso refratário. O mapeamento da região com DCF geralmente é feito por meio de imagens de ressonância magnética em conjunto com outras técnicas, como, por exemplo, o PET (positron emission tomography), o EEG (eletroencefalograma) intracraniano, entre outras. Contudo, por serem técnicas muito caras, de difícil realização ou muito invasivas, e por sabermos que as regiões displásicas possuem uma rigidez diferente da do restante do cérebro, foi proposto nesta dissertação o estudo desses casos utilizando uma técnica barata, simples, não invasiva e sensível à rigidez tecidual, a elastografia por ultrassom, na qual, para causar a deformação do tecido cerebral, foram usadas próprias artérias internas do cérebro. Para tal estudo, criamos um algoritmo de processamento de dados com uma interface gráfica GUI (graphical user interface) capaz de mudar os parâmetros de processamento e ver seus resultados em tempo real. Em seguida, esse algoritmo foi estudado em um ambiente controlado em material mimetizador de tecido biológico (phantom), no qual construímos um bloco de 10 x 10 x 12cm3, preenchido com material que mimetiza as propriedades mecânicas e acústicas do tecido mole e inserimos nele uma bexiga canudo preenchida com um uido simulador de sangue e uma inclusão mais rígida do que a base do material, posicionada acima do canudo. Foi utilizado, também, um acionador mecânico pulsátil para simular a pulsação mecânica equivalente à pulsação sanguínea da artéria cerebral. Foram feitas imagens elastográcas e de velocidade utilizando somente a deformação causada pelo deslocamento da bexiga, no interior do phantom, e, através de uma transformada de Fourier, foi calculado o período de pulsação da bexiga. Vimos que as imagens elastográcas e de velocidade foram capazes de localizar a inclusão, e o processamento temporal pode nos mostrar com precisão a frequência de pulsação da bexiga canudo. Finalizada essa etapa laboratorial, zemos o mesmo procedimento, porém in vivo, para dois casos: um com DCF tipo III-B, no qual não enxergávamos nada no modo B; e outro com tipo II-B, no qual foi observado uma diferença de impedância mecânica pelo modo B. As imagens foram coletadas durante o procedimento cirúrgico pelo próprio cirurgião usando um transdutor microconvexo acoplado a uma plataforma de ultrassom, modelo Sonix RP, e processadas num segundo momento. Vimos, no primeiro caso, pelas imagens elastográcas, as regiões mais rígidas, supostamente displásicas, que não estavam aparecendo no modo B e, no segundo caso, uma região maior do que a apresentada no modo B. Nossos resultados das medidas de frequência da pulsação arterial, para ambas as situações, 61; 5BPM e 91BPM, caram bastante próximos do valor medido com o eletrocardiograma durante a coleta do sinal, 65BPM e 94BPM, respectivamente. Por meio dos resultados da análise histológica, pudemos conrmar que o que estávamos enxergando com nosso programa era realmente uma região displásica. Dessa forma, concluímos que nosso algoritmo funcionou bem para esses casos clínicos. / The mainly goal of this work was to study a specic case of refractory epilepsy generated by a malformation in the brain tissue, called focal cortical dysplasia (FCD). Due the fact it is a brain malformation its eects show up since the childhood where it is the principal epilepsy generator. The mapping of this region is usually made by magnetic resonance images with another technique, such as, for instance, the PET (position emition tomography), the EEG (electrocardiogram), and others. However, for the fact that these techniques are expensive, dicult to perform or invasive, and knowing that the dysplastic regions are stier than the regular brain tissue, it was proposed in this dissertation the use of ultrasound elastography as a cheaper, simpler and noninvasive image modality capable to detect dierences in the tissue stiness of the FCD region. To generate the strain in the brain tissue it was used the pulsation of the local arteries. To achieve our goal, we created a data processing algorithm in MATLAB with a graphic user interface (GUI) capable to change the processing parameters to see its results in real time. This algorithm was tested in phantom using a block of tissue mimicking material (10 x 10 x 12 cm3). A balloon of latex led with a blood mimicking uid was immersed in the middle of the phantom and a cylindrical inclusion of 1 cm of diameter was immersed above the balloon. The bulb of the balloon was keep outside of the phantom to be mechanically pressured by a dedicated magnetic actuator, simulating the mechanical pulsation of the brain arteries. The velocity and elastography images were studied using just the strain caused by the displacement of the wall of the balloon tube inside the phantom. The period of pulsation was precisely calculated from these images. After that, we did the same procedure in two in vivo cases: one with FCD type III-B; and the other with FCD type II-B. All our intraoperative images were acquired for the surgeons using a micro convex transducer linked to an Ultrasound platform (Sonix RP) and, then, processed o-line. In the B mode scanning we didnât see any formation inside the brain for the rst case, and for the second, we did. In the elastographic images we saw a clearly stiffer region in the rst case that was invisible in the B mode; and for the second case, we saw a bigger stiffer region than we saw in the B mode imaging too. And for both results, the arteria pulsation frequency, 61.5 BPM and 91 BPM, were veryclose to the measured value collected in the electrocardiogram during the surgery, 65 BPM and 94 BPM, respectively. Analyzing the histological results we could conrm that what we were showing in our elastographic images were FCD, indeed. Thereby we concluded that our algorithm had worked in these clinical data.

A magneto-acustografia ou, em uma nova terminologia mais abrangente, vibromagneto-acustografia (VMA) é uma técnica que combina técnicas magnéticas e acústicas, a fim de observar o comportamento visco-elástico de um meio material marcado com partículas magnéticas (alvo). A força magnética oscilatória de frequência () provoca, dentro do alvo, movimento das partículas e do meio material com frequência (2). O feixe acústico pulso-eco utilizado para detectar o movimento das estruturas, tem uma frequência de repetição de pulso (PRF), no mínimo, cinco vezes maior que a frequência do movimento do alvo. Assim, a proposta deste projeto foi desenvolver um protocolo de aquisição, processamento e análise da técnica VMA, por meio da implementação de uma plataforma automatizada de aquisição, que forneça dados estruturados para posteriormente através de uma plataforma de processamento analisar diferentes aplicações da técnica VMA em fantomas e in vivo. A primeira etapa foi desenvolver a plataforma de aquisição utilizando um equipamento de ultrassom diagnóstico por imagem. Nesta etapa foi desenvolvido um software, com interface gráfica de usuário, escrito em C++ e Qt com a função de automatizar a excitação magnética e a aquisição acústica de alta resolução temporal, foram definidos os protocolos padrão dos modos de excitação magnética e de aquisição acústica em termos da taxa de aquisição dos frames ultrassônicos para gerar movimentos oscilatórios ou ondas de cisalhamento. Em seguida, os dados provenientes das aquisições foram estruturados por meio de um cabeçalho para transferência para outros computadores através do protocolo TCP/IP. Depois foi desenvolvida um software, com interface gráfica de usuário, para a fase de processamento e análise. O processamento dos frames adquiridos em modo radiofrequência (RF) consistiu em obter mapas de deslocamentos ou velocidades das estruturas internas do meio material utilizado como alvo. Nesta etapa adaptamos o algoritmo de correlação cruzada normalizada com correção de pico de máxima correlação x e obtivemos mapas otimizados para a técnica VMA. A avaliação da otimização destes mapas foi baseada na avaliação entre a resolução espacial e tempo computacional. A partir do uso destes softwares em experimentos com fantomas e in vivo foram desenvolvidos métodos de análise da frequência dos deslocamentos e segmentação/caracterização de ondas de cisalhamento em termos da sua velocidade e atenuação. Por fim, foi implementado um sistema VMA automatizado composto, por uma plataforma de aquisição embarcada em equipamento de diagnóstico por ultrassom para aquisição de mapas de RF e, que posteriormente eram transferidos para um computador com hardware superior para geração de mapas de deslocamento do meio material. Os mapas foram analisados para determinação de parâmetros visco-elásticos. Além disso, através da implementação da técnica VMA, viabilizou-se estudos clínicos de maneira rápida e eficiente, por exemplo, análises gástricas após a digestão de alimentos marcados com partículas magnéticas e produção de ondas de cisalhamento para caracterização viscoelástica. / The magneto motive ultrasound (MMUs) is a novel technique, which combines magnetism and acoustics, in order to observe viscoelastic behavior of medium labeled with inserted magnetic particles. The magnetic force with modulation frequency () creates a mechanical disturbance in the object of study with twice the modulation frequency (2). Was applied a pulse-echo ultrasonic beamforming with pulse repetition frequency (PRF) at least five times greater than frequency of internal structure movement. The aim of this study was to develop an acquisition, processing and analysis protocol for MMUs through implementation of an automated platform for acquisition, which provides structured datasets for further processing and analysis of different applications (tissue mimicking phantoms or in vivo) of MMUs. First step was to develop the acquisition platform using an ultrasound research interface (URI). At this stage was developed a software with graphical user interface (GUI) written using C++ and Qt. This software automates magnetic excitation and acoustic acquisition, which has high frame rate. In addition, were defined default presets to provide oscillatory movement or shear waves. Then was created datasets acquired were structured and a header and transferred to an external personal computer through TCP/IP network. Second, the processing software with graphical user interface for processing and analyzing was developed. The frames acquired in radiofrequency (RF) mode were processed into displacement or velocity maps of medium internal structures. In this stage, we adapted a cross correlation algorithm to optimize for MMUs datasets. The evaluation of these maps was based on tradeoff between spatial resolution and computation time. Protocols for analysis of motion frequency and segmentation/characterization of shear waves, extracting velocity and attenuation for experiments performed with phantoms or in vivo were developed . In this work, it was implemented an automated MMUs system integrated with a software framework running on an ultrasound research interface (URI), which is used for acquisition of RF maps, further transferred for a PC with robust hardware to process into displacement maps. These maps were analyzed to obtain viscoelastic parameters of the medium. Moreover, this implementation of MMUs enables clinical fast and efficient trials for gastric evaluation of meals with magnetic particles and shear wave production for viscoelastic characterization.

Les maladies chroniques du foie constituent un problème majeur de santé publique. L’évaluation précise du degré de fibrose hépatique apporte au clinicien une estimation du pronostic dès la prise en charge initiale de ces patients, mais permet également la réalisation d’une surveillance évolutive, et intervient dans la décision et le choix d’un traitement. L’évaluation noninvasive de la fibrose hépatique par élastographie a permis de révolutionner la prise en charge des malades atteints d’une maladie chronique du foie. L’objectif de notre travail est d’évaluer les performances diagnostiques d’une nouvelle technique d’élastographie hépatique, appelée « Supersonic ShearImaging » (SSI), et d’analyser sa plus-value dans l’évaluation non-invasive des maladies chroniques du foie.Dans une première étude, nous avons prospectivement analysé et comparé les performances diagnostiques de l’élastographie SSI par rapport au FibroScan et l’ARFI pour le staging de la fibrose hépatique sur une série de 349 patients avec une maladie chronique du foie diagnostiquée et gradée par ponction-biopsiehépatique.Dans une seconde étude, nous avons prospectivement étudié l’impact diagnostique de l’élastographie SSIsur le foie et la rate chez une population de 401 patients avec cirrhose pour l’évaluation de la gravité de la maladie cirrhotique.Dans une troisième étude, nous avons réalisé une analyse prospective bicentrique (Angers et Bordeaux)de la performance diagnostique de l’élastographie SSI par rapport au FibroScan et l’ARFI dans l’évaluation non invasive de la fibrose hépatique sur une série de 291 patients avec stéatopathie non-alcoolique et ponction-biopsiehépatique. / Abstract :The management and the prognosis for chronic liver diseases are widely based on the presence and the development of a liver fibrosis. The progressive worsening of liver fibrosis leads in a certain number of patients to the development of cirrhosis and its complications. Thus, the development of non-invasive diagnostic tools for the diagnosis and the monitoring of the liver fibrosis is of crucial interest. Liver elastography is one of the most promising techniques that have recently emerged in the field of chronic liver diseases. In this study, we aim to assess the diagnostic accuracy of a new elastography technique, named “Supersonic Shear Imaging” (SSI), and toanalyse its added value in the non invasive diagnosis of chronic liver diseases.In a first study, we prospectively analysed and compared the diagnostic performances of SSI elastography versus FibroScan and ARFI for the staging of liver fibrosis in a cohort of 349 patients with chronic liver diseases that consecutively underwent a liver biopsy. In a second study, we prospectively analysed the impact of liver and spleen SSI elastography in a cohortof 401 cirrhotic patients for the non invasive diagnosis of cirrhosis severity and oesophageal varices.In a third study, we assessed the clinical use of liver stiffness measurement evaluated by SSI, FibroScan,and ARFI in a cohort of nonalcoholic fatty liver disease patients who underwent liver biopsy. A total of 291 NAFLD patients were prospectively enrolled at 2 French university hospitals (Angers and Bordeaux).

Les modifications induites par les cellules tumorales sur leur environnement ont pour but de permettre leur développement en remodelant le tissu la soutenant et en créant un nouveau réseau vasculaire (angiogenèse). Plusieurs thérapies anti-angiogéniques inhibant le développement du réseau vasculaire tumoral ont obtenu l’autorisation de mise sur le marché et sont actuellement utilisées en clinique. Ces thérapies induisent de fortes modifications fonctionnelles au sein de la tumeur mais le simple suivi de l’évolution du volume tumoral n’est pas suffisant pour rendre compte de ces modifications. L’objectif principal de la thèse a consisté à utiliser différentes modalités d’imagerie ultrasonore afin d’évaluer leur sensibilité aux modifications générées dans des tumeurs murine (carcinome colorectal et pulmonaire) au cours de plusieurs types de thérapie (chimique : cytotoxique, anti-angiogénique / physique : plasma froid, sono-sensibilisation). Les modifications de la distribution spatiale des micro-vaisseaux et leur fonctionnalité ont été caractérisées à l’aide de l’imagerie de contraste ultrasonore (CEUS), l’altération de la microstructure de la tumeur a été évaluée grâce à l’analyse spectrale des signaux radiofréquences, connu comme « quantitative ultrasound » (QUS) et enfin les variations des propriétés mécaniques des tissus tumoraux ont été mesurées en élastographie à l’aide de la technique « Shear Wave Elastography » (SWE). Afin de comprendre l’origine des modifications observées in vivo, des paramètres standard comme les niveaux de fibrose ou de nécrose ont été caractérisés ex vivo dans le tissu tumoral, grâce à l’immunohistochimie, une technique de référence. / Tumor development is complex process made possible thanks to the microenvironment surrounding tumor cell. Modifications induced by tumor cells on their environment enable their own development by remodeling tissues sustaining them and by creating a new vascular network (angiogenesis). The use of several antiangiogenic therapies, inhibiting the sprout of a new vascular network, has been authorized in clinic. These therapies induce strong modifications in tumors at the functional level and following tumor size changes are is not sufficient to fully characterize tumor modifications. The main goal of this thesis was to use different ultrasound-based imaging modalities in order to assess their sensitivity to modifications induced in murine tumor model (colorectal and lung carcinomas) during different type of therapy (chemical: cytotoxic, antiangiogenic / physical: cold plasma, sonosensitization). Modifications of the spatial distribution of microvessels and their functionality were characterized using contrast-enhanced ultrasound (CEUS), alteration of tumor microstructure was assessed using spectral analysis of radiofrequency signal, known as quantitative ultrasound (QUS) and finally variations of mechanical properties in tumor tissues were measured in shear wave elastography (SWE). In order to better understand the origin of the modifications observed in vivo, standard parameters such as level of fibrosis and necrosis were characterize ex vivo in tumor tissue using immunochemistry as gold standard.

京都大学 / 0048 / 新制・課程博士 / 博士(人間健康科学) / 甲第22387号 / 人健博第73号 / 新制||人健||5(附属図書館) / 京都大学大学院医学研究科人間健康科学系専攻 / (主査)教授 黒木 裕士, 教授 山田 重人, 教授 松田 秀一 / 学位規則第4条第1項該当 / Doctor of Human Health Sciences / Kyoto University / DFAM

This project concerns the computational solution of inverse problems formulated as partial differential equation (PDE)-constrained optimization problems with interior data. The areas addressed are twofold. First, we present a novel software architecture designed to solve inverse problems constrained by an elliptic system of PDEs. These generally require the solution of forward and adjoint problems, evaluation of the objective function, and computation of its gradient, all of which are approximated numerically using finite elements. The creation of specialized "layered"' elements to perform these tasks leads to a modular software structure that improves code maintainability and promotes functional interoperability between different software components. Second, we address issues related to forward model definition in the presence of boundary condition (BC) uncertainty. We propose two variational formulations to accommodate that uncertainty: (a) a Bayesian formulation that assumes Gaussian measurement noise and a minimum strain energy prior, and (b) a Lagrangian formulation that is completely free of displacement and traction BCs. This work is motivated by applications in the field of biomechanical imaging, where the mechanical properties within soft tissues are inferred from observations of tissue motion. In this context, the constraint PDE is well accepted, but considerable uncertainty exists in the BCs. The approaches developed here are demonstrated on a variety of applications, including simulated and experimental data. We present modulus reconstructions of individual cells, tissue-mimicking phantoms, and breast tumors.

L'élastographie par résonance magnétique (ERM) est une technique d'imagerie, reconnue comme une méthode pertinente pour la caractérisation mécanique des tissus humains in vivo. Celle-ci représente un intérêt fondamental en diagnostic clinique car le développement d'un processus pathologique s'accompagne la plupart du temps d'altérations des propriétés mécaniques des tissus atteints. L'ERM consiste à enregistrer le champ de déplacement induit au passage d'une onde de cisaillement généré dans le milieu étudié. Les paramètres mécaniques comme la vitesse, v, et les modules de viscoélasticité de cisaillement, G' et G'', peuvent être cartographiés. La quantification des paramètres mécaniques dépend à la fois de la fréquence mécanique, fexc, de la taille de voxel, a, de l'amplitude des champs de déplacement induits, A, de l'amplitude du rotationnel du champ de déplacement, q, des erreurs de mesure, ΔA et Δq, donc du rapport signal à bruit, RSB, et enfin de la méthode de reconstruction. En inversant les équations différentielles du champ de déplacement acquis selon les trois dimensions de l'espace, ces paramètres ont été considérés pour déterminer la précision et l'exactitude des modules mécaniques obtenus et établir les conditions de validité de l'ERM. Dans cette thèse, nous avons tout d'abord considéré A et A/ΔA afin de définir un premier seuil de validité pour l'ERM. Nous avons étudié l'influence de ces deux paramètres sur un fantôme hétérogène dans un appareil IRM 1,5 T avec deux types d'antennes. Dans une première étude, les champs de déplacement ont été acquis en fonction de A en utilisant deux séquences écho de spin (RFE) et écho de gradient (FFE) sensibilisées au mouvement pour une taille de voxel isotrope de 1 mm. Dans une seconde étude, ils ont été acquis en RFE en fonction de A pour trois résolutions spatiales différentes. Ces études ont révélé l'existence d'un seuil en A/ΔA au-delà duquel les paramètres extraits (G', G'') atteignent un plateau et l'ERM est fiable. Nous avons ensuite considéré le nombre de voxel par longueur d'onde, λ/a, comme paramètre déterminant des conditions de validité de l'ERM et nous avons caractérisé la qualité des données acquises par le rapport q/Δq. Sur des simulaitons dans un milieu élastique, homogène et isotrope avec un RSB variant entre 5 et 30, la précision et l'exactitude des mesures se sont avérées optimales pour 6 à 9 voxels par longueur d'onde. Nous avons reproduit expérimentalement à 2 kHz les conditions des siimulations sur un fantôme de PVA. Les champs de déplacement ont été acquis à 11,7 T en utilisant une séquence RFE sensibilisée au mouvement pour des résolutions spatiales de 150 μm à 300 μm afin de balayer le rapport λ/a de 1 à 20. Les résultats expérimentaux confirment pleinement les prédictions de la simulation. La vitesse de cisaillement diminue et tend vers la vitesse de référence attendue lorsque l'acquisition est réalisée dans le domaine optimal, à savoir ici lorsque a est inférieure ou égale à 200 μm. En outre la dispersion de la vitesse est réduite dans le domaine optimal et des estimations plus précises des paramètres mécaniques ont pu être déduites. Cette thèse montre d'une part que la précision et l'exactitude de l'ERM sont optimales lorsque les acquisitions sont réalisées ou traitées pour un domaine d'échantillonnage de la longueur d'onde déterminé par le RSB. Elle montre d'autre part que la comparaison des résultats obtenus doit être menée dans une gamme similaire de q/Δq. La prise en compte des conditions de validité de l'ERM, déterminées par les rapports λ/a et q/Δq, conduit à une mesure quantitative effective des paramètres mécaniques. Il est ainsi possible d'envisager un diagnostic clinique pertinent au sein d'un même organe, d'un même sujet, entre sujets ou au cours du temps. / Magnetic Resonance Elastography (MRE) is a imaging technique, recognized as a pertinent method for the mechanical characterization of human tissue in vivo. It offersa particular interest in clinical diagnosis because the development of a pathological process is often accompanied by modifications of the mechanical properties of diseased tissues. MRE consists of recording, along the three spatial dimensions, the displacement field induced by the propagation of a shear wave generated by excitation of the investigated tissue. Mechanical parameters such as shear wave velocity, v, and shear moduli, G' and G'', can then be mapped. The quantification of the mechanical parameters depends on the frequency of the mechanical excitation, fexc, the spatial resolution, a, the amplitude of the induced displacement field, A and the amplitude of the curl field displacement, q, with associated measurement errors, ΔA and Δq, (related to the signal-to-noise ratio, SNR) and finally the reconstruction method. All these parameters were considered to determine the precision and the accuracy of the estimated mechanical moduli and to establish the conditions of validity of MRE following the inversion of the differential equations of the displacement field. In this work, first A and A/ΔA were considered to define a validity threshold for MRE. The influence of A and A/ΔA was studied on a heterogeneous phantom acquired using a 1.5 T MRI with two different types of coils. In a first study, the displacement fields were acquired as a function of A using motion-sensitized spin-echo (REF) and gradient-echo (FFE) sequences for an isotropic spatial resolution of 1 mm. In a second study, the displacement field was acquired as a function of A using RFE for three different spatial resolutions. These studies revealed the existence of a threshold in A/ΔA beyond which the extracted parameters (G', G'') reach a plateau and the MRE is reliable. Then the number of voxels per wavelength, λ/a was considered as a parameter determining the conditions of validity of MRE. This parameter was studied according to the quality of the acquired data characterized by the ratio q//Δq. Simulations were carried in a homogeneous and isotropic elastic medium with a SNR between 5 and 30. The accuracy and the precision of the measurements were found optimal for 6 to 9 voxels per wavelength. The simulation conditions were experimentally reproduced at 2 kHz on a home-made polyvinyl alcohol phantom. The displacement fields were acquired at 11.7 T using a motion-sensitized RFE sequence with spatial resolutions ranging from 150 μm to 300 μm in order to obtain a λ/a ratio ranging from 1 to 20. The experimental results fully confirm the predictions of the simulation. The shear wave velocity decreases with λ/a. It tends towards the expected reference value when the acquisition is performed in the optimal condition, namely here when a is less than or equal to 200 μm. In addition, the standard deviation of the shear wave velocity is reduced for the optimal conditions. Therefore, accurate estimation of mechanical parameters could be deduced. This thesis first demonstrates that the precision and accuracy of MRE are optimal when the acquisitions are performed or processed for a certain wavelength sampling range determined by the SNR. We also showed that for fair comparison of the results, MRE must be carried out in a similar range of q/Δq. Taking into account the conditions of validity of MRE, determined by the ratios λ/a and q/Δq, leads to an effective quantitative measurement of the mechanical parameters making it possible to establish a relevant clinical diagnosis within the same organ, the same subject, between subjects or over time.

Airway hyperresponsiveness (AHR) is a hallmark of asthma where constriction of airway smooth muscle (ASM) causes excessive airway narrowing. Asthmatics, unlike healthy subjects, cannot prevent or reverse this narrowing by stretching their airways with a deep inspiration (DI). Since stretching of isolated ASM causes dramatic reductions in force generation and asthmatics tend to have stiffer airways, researchers hypothesize that reduced ASM stretching during breathing and DIs results in hyperreactive airways. However, counterintuitively, excised measurement on intact airways show narrowing is minimally reversed by pressure oscillations simulating breathing and DIs. We hypothesized that AHR does not result from reduced capacity to stretch the airways; furthermore, each constituent of the airway wall experiences different strain magnitude during breathing and DIs. To test this, we used an intact airway system which controls transmural pressure (Ptm) to simulate breathing while measuring luminal diameter in response to ASM agonists. An ultrasound system and automated segmentation algorithm were implemented to quantify and compare the ability of Ptm fluctuations to reverse and prevent narrowing in larger (diameter=5.72±0.52mm) relative to smaller airways (diameter=2.92±0.29mm). We found the ability of Ptm oscillations to reverse airway narrowing was proportional to strain imposed on the airway wall. Further, tidal-like breathing Ptm oscillations (5-15cmH2O) after constriction imposed 196% more strain in smaller compared to larger airways (14.6% vs. 5.58%), resulting in 76% greater reversal of narrowing (41.2% vs. 23.4%). However, Ptm oscillations applied before and during constriction resulted in the same steady-state diameter as when Ptm oscillations were applied only after constriction. To better understand these results, we optimized an ultrasound elastography technique utilizing finite element-based image registration to estimate spatial distributions of displacements, strains, and material properties throughout an airway wall during breathing and bronchoconstriction. This required we formulate and solve an inverse elasticity problem to reconstruct the distribution of nonlinear material properties. Strains and material properties were radially and longitudinally heterogeneous, and patterns and magnitudes changed significantly after induced narrowing. Taken together, these data show AHR likely does not emerge due to reduced straining of airways prior to challenge, but remodeling that stiffens airway walls might serve to sustain constriction during an asthmatic-like attack.

Each year million of babies are born pre-term, some of these pre-term births occur due to the motherhaving a too soft cervix which can not withstand the forces the baby exposes it to. The aim of thisstudy was to implement and evaluate a programmable shear wave elastography ultrasound system forcervical applications and investigate the optimal settings of shear wave elastography push voltage andshear wave elastography push focus depth. Shear wave elastography is an ultrasound based imagingmodality aiming to evaluate the tissue elasticity by using acoustic radiation forces to induce shear waves.The propagation of the shear waves through the tissue is then tracked in order to calculate the shearwave velocity which is related to the tissue elasticity. B-mode imaging, pushing sequence and planewave imaging have been implemented and measurements have been conducted on four cervical polyvinylalcohol phantoms. The acquired data has been post-processed using Loupas 2D-autocorrector to gainthe axial displacement and enabling tracking of the shear waves to allow evaluation and optimizationof the implemented method. The implemented shear wave technique showed to be able to distinguishcervical phantoms of dierent elasticity and a high pushing voltage and shallow focus push depth havebeen found to produce the most reliable results.