Dissertations / Theses on the topic 'Radiographic simulation'

Depuis plus d’un siècle, la radiographie sur film est utilisée pour le contrôle non destructif (CND) de pièces industrielles. Avec l’introduction de méthodes numériques dans le domaine médical, la communauté du CND industriel a commencé à considérer également les techniques numériques alternatives au film. La radiographie numérique (en anglais Computed radiography -CR) utilisant les écrans photostimulables (en anglais imaging plate -IP) est une voie intéressante à la fois du point de vue coût et facilité d’implémentation. Le détecteur (IP) utilisé se rapproche du film car il est flexible et réutilisable. L’exposition de l’IP aux rayons X génère une image latente qui est ensuite lue et numérisée grâce à un système de balayage optique par laser. A basse énergie, les performances du système CR sont bonnes ce qui explique son utilisation importante dans le domaine médical. A haute énergie par contre, les performances du système CR se dégradent à la fois à cause de la mauvaise absorption de l’IP mais également de la présence de rayonnement diffusé par la pièce qui, étant d’énergie plus faible, est préférentiellement absorbée par l’IP. Les normes internationales préconisent l’utilisation d’écrans métalliques pour améliorer la réponse des systèmes CR à haute énergie. Néanmoins, la nature et l’épaisseur de ces écrans n’est pas clairement définie et la gamme des configurations possibles est large. La simulation est un outil utile pour prévoir les performances d’une expérience et déterminer les meilleures conditions opératoires. Les méthodes Monte Carlo sont communément admises comme étant les plus précises pour simuler les phénomènes de transport de rayonnement, et ainsi comprendre les phénomènes physiques en jeu. Cependant, le caractère probabiliste de ces méthodes implique des temps de calcul importants, voire prohibitifs pour des géométries complexes. Les méthodes déterministes au contraire, peuvent prendre en compte des géométries complexes avec des temps de calcul raisonnables, mais l’estimation du rayonnement diffusé est plus difficile. Dans ce travail de thèse, nous avons tout d’abord mené une étude de simulation Monte Carlo afin de comprendre le fonctionnement des IP avec écrans métalliques à haute énergie pour le contrôle de pièces de forte épaisseur. Nous avons notamment suivi le trajet des photons X mais également des électrons. Quelques comparaisons expérimentales ont pu être menées à l’ESRF (European Synchrotron Radiation Facility). Puis nous avons proposé une approche de simulation hybride, qui combine l'utilisation de codes déterministe et Monte Carlo pour simuler l'imagerie d'objets de forme complexe. Cette approche prend en compte la dégradation introduite par la diffusion des rayons X et la fluorescence dans l'IP ainsi que la diffusion des photons optiques dans l'IP. Les résultats de différentes configurations de simulation ont été comparés<br>For over a century, film-based radiography has been used as a nondestructive testing technique for industrial inspections. With the advent of digital techniques in the medical domain, the NDT community is also considering alternative digital techniques. Computed Radiography (CR) is a cost-efficient and easy-to-implement replacement technique because it uses equipment very similar to film radiography. This technology uses flexible and reusable imaging plates (IP) as a detector to generate a latent image during x-ray exposure. With an optical scanning system, the latent image can be readout and digitized resulting in a direct digital image. CR is widely used in the medical field since it provides good performance at low energies. For industrial inspection, CR application is limited by its poor response to high energy radiation and the presence of scattering phenomena. To completely replace film radiography by such a system, its performance still needs to be improved by either finding more appropriate IPs or by optimizing operating conditions. Guidelines have been addressed in international standards to ensure a good image quality supplied by CR system, where metallic screens are recommended for the case of using high energy sources. However, the type and thickness of such a screen are not clearly defined and a large panel of possible configurations does exist. Simulation is a very useful tool to predict experimental outcomes and determine the optimal operating conditions. The Monte Carlo (MC) methods are widely accepted as the most accurate method to simulate radiation transport problems. It can give insight about physical phenomena, but due to its random nature, a large amount of computational time is required, especially for simulations involving complex geometries. Deterministic methods, on the other hand, can handle easily complex geometry, and are quite efficient. However, the estimation of scattering effects is more difficult with deterministic methods. In this thesis work, we have started with a Monte Carlo simulation study in order to investigate the physical phenomena involved in IP and in metallic screens at high energies. In particular we have studied separately the behavior of X-ray photons and electrons. Some experimental comparisons have been carried out at the European Synchrotron Radiation Facility. Then, we have proposed a hybrid simulation approach, combining the use of deterministic and Monte Carlo code, for simulating the imaging of complex shapes objects. This approach takes into account degradation introduced by X-ray scattering and fluorescence inside IP, as well as optical photons scattering during readout process. Different simulation configurations have been compared

O diagnóstico radiográfico é baseado na análise das diferenças das densidades ópticas (DO) do filme, que deveriam ser provocadas apenas pelas estruturas anatômicas do paciente. Entretanto, a intensidade do feixe de raios X não é uniforme devido a um efeito intrínseco do equipamento de aquisição de imagem, conhecido como efeito \"heel\". Estas variações prejudicam tanto a análise visual quanto o processamento computacional (CAD) das pequenas estruturas anatômicas. O presente trabalho apresenta um método computacional que corrige as diferenças de densidades ópticas produzidas na radiografia pelo efeito \"heel\". Esse método foi implementado utilizando ambiente de programação Delphi, rotinas em C e Matlab. O método simula a distribuição da intensidade ao longo do campo de radiação, determinando o caminho de absorção que os fótons sofrem dentro do alvo utilizando os modelos de Kramers e Fritz Livingston. Calcula a correlação espacial entre a radiografia e a imagem simulada, localizando o eixo anodo/catodo e o centro do campo nas duas imagens, empregando a função de correlação estatística de Pratt e a função de mapeamento de Zitová e Flusser. Calcula tanto os percentuais de radiação recebidos para cada ponto simulado em relação à radiação ao centro do campo, quanto os percentuais dos níveis de cinza de cada pixel da radiografia e corrige esse valor em função do correspondente na simulação. O algoritmo desenvolvido permitiu determinar a posição do centro do campo de radiação com precisão em torno de 1% e eliminou aproximadamente 90% do efeito \"heel\" na radiografia permitindo que os objetos apresentassem densidades ópticas coerentes com suas absorções específicas. Um estudo preliminar mostrou que esse método poderá ser utilizado como pré-processamento dos sistemas CAD.<br>The radiographic diagnosis is based on the analysis of the film optical density differences that should be created only by the patient anatomical structures. However, the intensity of the x-ray beam is not uniform due to an intrinsic effect to the image acquisition equipment, known as heel effect. These variations damage the visual analysis as well the (CAD) computer processing of the small anatomical structures. The current work presents a computer method that corrects the optical densities differences generated in the radiography by heel effect. This method was implemented using Delphi Programming Environment, routines in C and Matlab. The method simulates the intensity distribution along the radiation field, determining the absorption path that photons suffer inside the target using the models of Kramers and Fritz and Livingston. It calculates the space correlation between the radiography and the simulated image, determining the anode/cathode axis and the field center in the two images, using the statistics function of Pratt and the mapping function of Zitová and Flusser. It calculates as much the received radiation, percentage for each simulated point in relation the field center radiation, as the gray scales percentage of each radiography pixel and corrects their values as function of the correspondent in the simulation. The developed algorithm has allowed to determine the center position of the radiation field with about 1% precision and approximately eliminated 90%of the heel effect in the radiography, allowing the objects to present optical densities coherent with their specific absorptions. A preliminary study has showen that this method can be used as preprocessing of CAD systems.

Thesis (M.S.)--University of Florida, 2002.<br>Title from title page of source document. Document formatted into pages; contains xii, 91 p.; also contains graphics. Includes vita. Includes bibliographical references.

La radiographie éclair par faisceau X intense est spécifique en ce sens qu'elle doit permettre de photographier la matière soumise à des conditions extrêmes de densification, de température et de vitesse de déplacement. Le succès de ce type de radiographie repose sur la qualité de la source X qui doit nécessairement être pénétrante (quelques MeV), intense (plusieurs rads), brève (quelques dizaines de ns) et de petite dimension (quelques mm). L'impulsion X est ainsi générée à partir du rayonnement de freinage émis lors de l'interaction avec une cible en métal d'un faisceau focalisé d'électrons de haute énergie (MeV) et de haute intensité (kA). Ce procédé lie très fortement les propriétés du faisceau d'électrons à ceux du faisceau X et donc à la qualité de la radiographie. Dans ce contexte, la thèse porte sur la compréhension de la dynamique du faisceau dans la diode à l'électron (c'est-à-dire juste avant son entrée dans la ligne accélératrice) ainsi que sur la caractérisation du plasma de velours dont sont issus les électrons qui composent le faisceau. Dans un premier temps, la dynamique du faisceau intense d'électrons a été étudiée à l'aide du code LSP reposant sur la méthode " Particle-In-Cell ". Les simulations réalisées ont été comparées avec des mesures effectuées sur l'injecteur d'un accélérateur linéaire à induction, implanté au CEA Valduc sur l'installation Epure. Grâce au modèle de simulation développé, une nouvelle diode à électrons mono-impulsion a été conçue, dimensionnée et réalisée pendant ce travail de thèse afin d'augmenter l'intensité du faisceau d'électrons de 2,0 kA à 2,6 kA permettant ainsi d'améliorer les performances radiographiques de l'installation. Dans un second temps, un modèle permettant d'étudier les mécanismes mis en jeu dans la production du faisceau d'électrons au niveau de plasma de cathode a été développé. Ce dernier est un modèle collisionnel-radiatif (MCR) 0D qui permet de décrire l'évolution de la densité des espèces d'un plasma dont la composition est directement liée aux molécules et atomes désorbés par la cathode de velours. Trois différents mélanges ont été étudiés impliquant de l'hydrogène, de l'oxygène et du carbone dont les proportions ont été estimées par des mesures LIBS (spectroscopie de plasma induit par laser).[...]<br>Intense X-ray flash radiography is used to take a stop-action picture of a material under extreme conditions like high densification, high temperature and high movement speed. The success of this kind of radiography is based on the quality of the X-ray source which must necessarily be penetrating (some MeV), intense (several rads), short (a few tens of ns) and small (a few mm). The X-ray pulse is generated from the bremsstrahlung radiation emitted during the interaction with a metal target of a focused electron beam of high energy (MeV) and high intensity (kA). This process strongly links the properties of the electron beam to those of the X-ray beam and thus to the quality of the radiography picture. In this context, the thesis is about the electron beam dynamics in the electron diode (i.e. just before electrons move towards the accelerator) as well as about the characterization of the velvet plasma from which electrons are extracted to form the beam. Firstly, the dynamics of the intense electron beam was studied using the LSP code based on the "Particle-In-Cell" method. The simulations were compared to measurements made on the injector of a linear induction accelerator, at the CEA Valduc center on the Epure facility. Based on the developed simulation model, a new single-pulse electron diode was designed, sized and realized during this thesis to increase the intensity of the electron beam from 2.0 kA to 2.6 kA, thus improving the radiographic performances of the facility. In a second step, a model allowing to study the mechanisms involved in the production of the electron beam from the cathode plasma was developed. This latter is a collisional-radiative model (CRM) 0D describing the evolution of the plasma species density of a plasma whose composition is directly related to the molecules and atoms desorbed by the velvet cathode. [...]

A presente tese demonstra, a partir da avaliação convencional pelo método das Funções de Transferência de sistemas de imagem radiológica, que é necessário obter imagens de fenda em diversas orientações no campo para que essa análise tenha um significado mais real no caso de sistemas não isotrópicos. Isso provém da não linearidade na variação entre as FTMs obtidas para diversas direções 0 e 90&#176C relativas ao eixo do tubo de raios-X. Essa verificação, entretanto, representa um sério problema prático, pois indica um aumento no grau de complexidade de um método que, embora considerado o mais preciso pela maioria dos pesquisadores, tem sido utilizado apenas por laboratórios muito bem equipados. Assim, visando solucionar esse problema, esta tese propõe um novo método de simulação por computador que calcula a FEL e a FTM devidas ao ponto focal, dispensando, portanto, todo o complexo aparato experimental convencionalmente utilizado, o que contribui para tornar acessível à avaliação pelas funções de transferência a qualquer unidade radiológica. Por fim, faz parte desse trabalho também uma investigação do significado físico das variações registradas entre as FTMs e um estudo formal desenvolvido acerca dos conceitos da característica de campo e da magnificação lateral.<br>From the conventional evaluation by the radiological systems Transfer Functions, this work shows that it is necessary to obtain slit images at several field orientations so that this annalysis has a more real significance for non-isotropic systems. This is achieved from the non-linearity on the variations among the MTFs obtained in several directions between 0 and 90&#176C relative to the X-ray tube axis. This notification, however, represents a serious practical matter, because it shows an increase on the complexity of a method which has been used just by well structured laboratories, although many researchers have considered it the most accurate. Hence, in order to solve this problem, we present a new computer simulation method which calculates the LSF and the MTF due to the focal spot, without all the conventional complex experimental apparatus. This makes the evaluation by the transfer functions suitable to any radiological unit. Finally, it is also part of this work an investigation of the physical meaning of the variations among the MTFs and a formal study about the field characteristics and the lateral magnification concepts.

Thèse doctorat : Génie des Matériaux : Villeurbanne, INSA : 2003.<br>Chap. 2 et 4 rédigés en anglais. En appendice, 1 article rédigé en anglais intitulé "Optimal calibration via virtual X-ray imaging for dual-energy techniques : application to glass wool", issu du Colloque "Six international Conference on quality control by artificial vision" et paru dans la revue "SPIE", vol. 5132, 2003, p. 422-432. Titre provenant de l'écran-titre. Bibliogr. p. 143-155.

Cette thèse est consacrée à la mise au point d'un code de calcul permettant de simuler rapidement des images radiologiques réalistes en prenant en compte les principaux paramètres physiques à l'œuvre dans une chaîne d'imagerie par rayons X ou gamma. Dans la première partie, nous effectuons un tour d'horizon de l'état de l'art en matière de simulation des phénomènes de transport de rayonnement. Cette étude nous conduit à choisir une approche déterministe et à rechercher des solutions algorithmiques spécifiques, dédiées à la simulation d'images radiologiques et limitées dans un premier temps à la prise en compte du rayonnement directement transmis. Les solutions proposées, qui mettent l'accent sur la vitesse d'exécution et la robustesse, sont implémentées dans un code appelé VXI (Virtual X-ray Imaging). VXI permet d'effectuer aisément des simulations dans des configurations d'imagerie réalistes (spectre polychromatique, objets de géométrie complexe. . . ). La deuxième partie de cette thèse aborde la simulation du rayonnement diffusé par les objets inspectés. Nous proposons une méthode déterministe pour simuler le rayonnement diffusé d'ordre 1 sans recourir à une architecture de calcul parallèle. Cette méthode est validée en comparant les résultats qu'elle fournit avec ceux que donne le code de Monte Carlo Geant4<br>This PhD thesis is devoted to the development of a computer code enabling to simulate in a short time realistic radiological images, taking into account the main physical parameters acting in an X- or gamma-ray imaging chain. In the first part, we carry out a general survey of the state of the art in the field of radiation transport simulation. This study leads us to choose a deterministic approach and to seek specific algorithms, devoted to the simulation of radiological images and, at first, accounting only for the directly transmitted radiation. The proposed solutions, which emphasize execution speed and robustness, are implemented in a code named VXI (Virtual X-ray Imaging). VXI makes it easy to carry out simulations in realistic imaging configurations (polychromatic spectrum, objects with complex geometry. . . The second part of this thesis broaches the simulation of the radiation scattered by the inspected objects. We propose a deterministic method to simulate first-order photon scattering without having recourse to a parallel computing architecture. This method is validated by comparing its results with the ones given by the Monte Carlo code Geant4

To optimise any medical digital imaging system for chest radiography, it is vital that the images used for optimisation contain projected anatomy, or in other words, anatomical noise. In this thesis, a method to produce and validate a digitally reconstructed radiograph (DRR) computer algorithm that utilises real patient computed tomography (CT) data sets is presented. The algorithm uses a ray casting DRR calculation method to project X-ray pencil beams through CT data and derive the photon energy absorbed in a virtual computed radiography (CR) phosphor. Radiation scatter and CR system noise are added post DRR calculation. Quantitative and qualitative validation has shown the algorithm simulates chest CR images of average and obese patients with realistic anatomical and system noise. This has allowed images to be generated using various X-ray exposure parameters, i.e. tube potential, scatter rejection and receptor dose, which can then be used in the optimisation exercise. However, the algorithm is not without limitations; the impact of these on the resulting images is discussed. Simulated images reconstructed at the various X-ray exposure parameters and techniques were scored by experienced image evaluators; optimum tube potential, scatter rejection technique and receptor doses for clinical CR chest radiography have been derived. At the outset of this work, CR chest exposure factors across the Hull & East Yorkshire Hospitals NHS Trust (HEY) were not standardised, and therefore not optimised; this thesis concludes with recommendations to the HEY Radiology Department for optimum exposure factors and technique for chest radiography. These were implemented across the Trust as a result of this work. In summary, a DRR computer algorithm has been produced (and validated) that adequately simulates anatomical and system noise; image evaluators are able to grade simulated chest images presented at different X-ray exposure parameters in order to optimise radiographic technique for clinical CR chest radiography, without the need for repeat patient exposures.

L'évaluation de la qualité des images médicales est essentielle pour optimiser un système d'imagerie. Dans le cadre d'approches basées sur la tâche, les modèles numériques proposés ont des limites: pour la tâche de localisation, la plupart des modèles ont besoin de la connaissance a priori des paramètres du signal; les modèles dédiés à la détection de signaux dans une image 3D se limitent aux signaux symétriques. Dans cette thèse, nous proposons de nouveaux modèles numériques: le CJO pour la détection sur une coupe d'un signal dont l'amplitude, l'orientation et la taille ne sont pas connus a priori; le PCJO et le msPCJO pour la détection-localisation de plusieurs signaux paramétriques d'amplitude, d'orientation, de taille et d'emplacement variables sur une coupe (PCJO) ou en 3D (msPCJO). Deux expériences ont été conçues et mises en oeuvre pour la validation de nos modèles sur des images IRM présentant des lésions de SEP. Les résultats indiquent que nos modèles sont performants et prometteurs pour l'évaluation de systèmes d'imagerie<br>Medical image quality assessment is critical for comparing and optimizing medical imaging system. In the frame of task-based approach, numerical observers proposed for the objective medical image quality assessment have several limitations: e. G. For the multi-signal localization, most need a priori knowledge of signal parameters; multi-slice MOs narrow the task down to the detection of one symmetrical signal at the image centre. In this thesis, we propose novel numerical observers: the CJO for the detection of one parametric signal with random amplitude, orientation and size on single-slice; the PCJO and the msPCJO for the detection-localization of multiple parametric signals with random amplitude, orientation, size and location on single-slice and multi-slice, respectively. Two observer studies were designed and implemented for their validation on MR images with MS lesions, and the JAFROC FOM results indicate that they are promising for predicting radiologists' task performance, ultimately for evaluating medical imaging systems

Electrical parameter profiles of human breast images can be used to simulate and analyze the anticipated effects on tissue from its interaction with electromagnetic fields involved in the cancer treatment exposure. In part, the success of this approach depends on the accuracy and precision in identifying the different tissue types. In this work, we propose two methods of segmenting human breast images with malignant tumors. The first method of algorithmic partitioning of the image involves manual color-edge contouring of the tissues using a cursor and subsequent identification of the tissue types. For the second method, MRI T1 values and thresholds are used to perform segmentation and we investigate the potential of incorporating edge detection. The first method is effective, while the second lacks precision, but eliminates the need of manual contouring. The images are imported as BMP files into SEMCAD, an electromagnetic simulation tool based on finite-difference time-domain method, which recognizes the grouped tissues and creates a model of the image. The model allows the user to easily assign electrical parameter values to the grouped tissues, according to the measured values reported in the literature.

Os métodos de controle de qualidade tradicionais aplicados ao radiodiagnóstico, é a melhor maneira de garantir a boa qualidade das imagens produzidas. No entanto, a investigação de particularidades oriundas do processo de formação de imagens radiológicas requer ferramentas computacionais complementares, em função do número de variáveis envolvidas. Entretanto, os fantomas computacionais baseados em voxels não conseguem representar as variações morfométricas necessárias para a simulação de exames cujo diagnóstico é baseado em imagem. Neste trabalho foi desenvolvido um novo tipo de fantoma computacional, baseado em modelagem 3D, que possui as vantagens apresentadas pelos fantomas computacionais tradicionais sem os problemas encontados nestes. A ferramenta de modelagem utilizada, o Blender, é disponibilizada gratuitamente na internet. A técnica utilizada foi a box modeling, que consiste na deformação de uma primitiva básica, nesse caso um cubo, até que apresente a forma da estrutura que se deseja modelar. Para tanto, foram utilizadas como referencia, imagens obtidas de atlas de anatomia e fotografias de um esqueleto fornecido pela Universidade de Mogi das Cruzes. Foram modelados o sistema ósseo, os órgãos internos e a anatomia externa do corpo humano. A metodologia empregada permitiu a alteração de parâmetros do modelo dentro da ferramenta da modelagem. Essa possibilidade foi mostrada através da variação, dos formatos do intestino e do aumento da quantidade de tecido adiposo da malha referente a pele. A simulação das imagens radiológicas foi realizada a partir de coeficientes de atenuação de massa de materiais, ossos e tecidos e de modelos com diversas características físicas. Essa versatilidade permite prever a influência que as diferenças morfométricas entre os indivíduos provocam nas imagens, propriciando dessa forma, uma ferramenta relevante complementar aos métodos de controle de qualidade tradicionais.<br>The conventional methods of quality control applied to radio diagnosis are the best way to have assured good quality of the produced images. Due the amount of variables to consider, the study of particular issues of the process of formation of radiological images requires complementary computational tools. However, the computational voxel based phantoms are not suitable to represent the morphometrical variations, intended for test simulations with image based diagnosis. This work developed a new type of computational phantom, based on 3D modelling. It has the same advantages of the conventional ones, without some of their restrictions. The modeling tool employed, Blender, is available on internet for free download. The project uses the technique called box modeling, which consists in the deformation of a primitive form (a cube, in this case) until it presents a similar form to that it is wanted to model. In order to achieve it, some images, obtained from anatomy atlas and a skeleton pictures obtained from University of Mogi das Cruzes, were used as reference. Were built models from skeletal system, internal organs and external human body anatomy. The applied methodology allowed model´s parameter settings on the modelling tool. This option was presented by means of intestine format variation and increase of adipose tissue on the mesh that represents skin. The simulation of radiological images was done by means of x-ray mass attenuation coefficients, bones and tissues and models with diferent physical characteristics. This flexibility allows the analysis and forecasting of the influences that morphometrical differences of individual implies on images, revealing an important tool that complements the conventional quality control tools.

Este trabalho foi motivado por um estudo maio, proposto pela Agência Internacional de Energia Atômica (IAEA). O objetivo do trabalho foi estimar o kerma incidente (Ki) e a dose glandular média (Dg) em pacientes submetidas a mamografias nas projeções crânio-caudal (CC) e médio-lateral-oblíqua (MLO), nos sistemas de aquisição de imagens analógico e digital CR. A estimativa foi realizada através de cálculos matemáticos, tendo como base estudos computacionais utilizando simulações de Monte Carlo. Este estudo fez parte de um estudo maior em mamografia, coordenado pela IAEA, cujos dados fizeram parte da representação dos dados brasileiros no estudo da América Latina. Foram escolhidas mamografias cuja mama exposta tinha aproximadamente 50% de glandularidade e 50% de adiposidade. Dados da técnica radiográfica utilizada nos exames, assim como a espessura da mama comprimida, foram registrados em planilhas especificas e o cálculo foi realizado com base em testes de rendimento do tubo de raios X dos mamógrafos. Além disso, as mamografias acompanhadas foram analisadas quanto a critérios de qualidade da imagem. Foram acompanhadas 100 exposições, sendo 50 analógicas e d 50 digitais. Os resultados mostraram que as imagens foram predominantemente aceitas quanto aos critérios de qualidade da imagem estabelecidos pala IAEA, com melhores resultados do sistema digital. Os valores de Ki e Dg encontrados para o objeto simulador estavam de acordo com o recomendado, mas os calculados para as pacientes estavam mais altos, principalmente o Ki. O sistema digital obteve doses mais baixas e melhor qualidade de imagem. Os resultados sugerem que os exames devem ser realizados com o devido cuidado, de maneira a oferecer qualidade de imagem para o diagnóstico em doses tão baixas quanto razoavelmente exeqüíveis. O treinamento contínuo dos profissionais da mamografia deve ser realizado de maneira e oferecer exames adequados ao diagnóstico, protegendo os pacientes e mantendo a qualidade do serviço a todos.<br>This work has been motivated by a larger study, proposed by the International Atomic Energy Agency (IAEA). The aim of this work was to estimate the incident kerma (Ki) and mean glandular dose (Dg) in patients who underwent mammograms in craniocaudal projections (CC) and mediolateral-ablique (MLO), at analog and digital systems. The estimation was performed using mathematical calculations based on computational studies using Monte Carlo simulations. This study was part of a larger study in mammography, coordinated by the IAEA, whose data were included in the representation of Brazilian data in the study of Latin America. Mammograms with 50% of glandular tissue and 50% of fat were selected. Data from the radiographic technique used in the examinations, as well as the compressed breast thickness were recorded in spreadsheets and specific calculation was made based on output tests of the X-ray tube for mammography. In addition, the mammograms were analysed with the criteria of image quality. The amount of examinations was 100 mammograms, 50 analog and 50 digital. The results showed that the images were largely accepted as the criteria for image quality established by the IAEA, with better results from digital system. The values of Ki and Dg for the phantom were according to the recommended, but those calculate for the patients were higher, especially the Ki. Digital system produced lower dosis and better results on image quality. The results suggests that the examinations must be conducted with due care in order to offer image quality in dosis as low as reasonably achievable. The permanent training of mammographers should be conducted in a manner to provide exams with quality, protecting the patient and maintaining the quality of service to all patients treated.

L'imagerie par rayons X est fortement développée dans notre société et notamment dans les domaines industriels, médicaux ou sécuritaires. L'utilisation de cette méthode d'imagerie des structures internes (pour la détection d'irrégularité, de contrôle non destructif de pièces ou de menaces) est quotidienne. En radiographie, le contraste produit sur les images est relié à la variation de l'atténuation du flux de rayons X, qui est fonction de la densité, de l'épaisseur du matériau étudié ainsi que de la longueur d'onde utilisée. Ainsi par exemple, des gaines métalliques, des os ou des armes amènent du contraste sur l'image. Mais en plus de leur atténuation, les rayons X vont subir un déphasage qui est d'autant plus important que le matériau est peu atténuant. Ce phénomène va amener du contraste, dit de phase, permettant d'imager des matériaux peu denses tels que des plastiques, composites, tissus mous ou explosifs. Ce travail de thèse présente le développement et l'adaptation, dans le domaine des rayons X, de méthodes d'imagerie par contraste de phase sur des équipements de laboratoire. Le but est de compléter, d'une manière plus accessible et quotidienne, les demandes d'évaluation non destructives. Ce manuscrit se découpe suivant deux axes portant sur la simulation d'une part et sur le développement instrumental d'autre part. Un outil de simulation a été développé portant sur une description hybride alliant optique géométrique et optique ondulatoire. Les limites du modèle et des validations sont présentées. La partie instrumentale se focalise sur l'étude de deux techniques d'imagerie différentielle de phase. La première technique est de l'interférométrie à décalage multilatéral, dont l'adaptation sur tube à rayons X est réalisée pour la première fois. Une exploitation intéressante de la redondance de la mesure que produit la technique sera notamment introduite. La deuxième approche est une technique d'interférométrie de suivi de tavelure, dont nous présenterons une nouvelle exploitation<br>X-ray imaging is widely used in non-destructive testing dedicated to industry, medical or security domain. In most of the radiographic techniques, the image contrast depends on the attenuation of the X-ray beam by the sample. This attenuation is function of the density and thickness of the object and of the wavelength. Therefore, objects like metal covers, bones or weapons bring contrast on the image. In addition to attenuation, phase shifting happens, in particular for low-attenuating material. This phenomenon brings contrast, called phase contrast, and allows a X-ray image of low-attenuating material as plastics, composites, soft tissues or explosives. This work presents development and adaptation, in the X-ray domain, of phase contrast imaging techniques on laboratory equipment. The goal is to bring phase contrast imaging in daily use. This manuscript is split in two parts, simulation and instrumentation. A simulation tool has been developed, mixing geometrical optic and wave optic. Limits of the model and validation are presented. For the instrumental part, two interferometric techniques have been considered. The first one is multi-lateral interferometry where adaptation on X-ray tube is presented for the first time. Interesting use of the measurement recurrence will be introduced. The second one is speckle tracking interferometry, recently adapted on X-ray tube, for which we present new advancements

La présence des structures normales du sein (texture) dans les images mammographiques numériques dégrade la pertormance de détection des radiologues. Nous proposons de nouveaux outils et de nouvelles méthodes pour étudier l'impact de la présence de texture sur la pertormance de détection des lésions dans ces images. Nous développons un outil permettant de synthétiser des images mammographiques avec une texture d'apparence réaliste. Nous évaluons l'impact de la texture sur le scoring d'un fantôme contraste-détail couramment utilisé en mammographie. Nous montrons que c fantôme non texturé est pertinent d'un point de vue clinique pour évaluer la détectabilité des microcalcifications, mais pas des lésions plu grandes comme les opacités. Nous proposons un nouvel observateur mathématique, basé sur une approche a-contrario introduite récemment. Avec une stratégie de détection proche de celle des radiologues dans leur tâche clinique, notre observateur est une alternative pertinente pour modéliser la perception visuelle humaine. Nous élargissons la méthode a-contrario au cas des images texturées et démontrons la capacité de cet observateur à prédire les lois de détectabilité dans diverses textures dont les textures mammographiques. Nous appliquons ces outils et méthodes à l'optimisation des paramètres d'acquisition d'un système de mammographie numérique. Nous montrons que le système actuel a atteint un optimum de fonctionnement. De plus, la variabilité induite par la texture ne peut pas être compensée en optimisant les paramètres d'acquisition du système, ce qui met en évidence l'intérêt des technologies d'imagerie du sein permettant de réduire l'influence de la texture<br>The presence of normal breast structure (texture) in digital mammography images decreases the detection pertormance of radiologists. We propose new tools and methods to study the impact of the presence of breast texture on the detectability of les ions in such images. We develop an image simulation tool which enables to synthesize mammography images with realistic texture appearance. We evaluate the impact of the texture on the scoring of a contrast-detail phantom commonly used in mammography. We show that this flat-field phantom is clinically relevant to assess the detectability of microcalcifications, but not of larger lesions such as masses. We propose a new mathematical observer, based on the recenUy introduced a-contrario framework. With a detection strategy close to th one of radiologists when looking for les ions in real clinical situations, this observer is a relevant alternative to model human visual perception. Extending the a-contrario framework to the case of textured backgrounds, we demonstrate the ability of the a-contrario observer to predict detectability laws in various textures and iparticular in mammography textures. We apply these new tools to the optimization of the acquisition parameters in digital mammography in presence of texture. We show that the current digital mammography systems have reached an optimal usage. Furthermore, we demonstrate that the variability on detection pertormance induced by breast texture can not be overcome by optimizing the acquisition techniques, highlighting the benefit of any breast imaging technology reducing the influence of breast texture

CAPES<br>Desde os primeiros trabalhos propondo o uso dos prótons para a construção da imagem, a principal vantagem, sobre os raios X, foi esperada como um resultado da propriedade específica do fluxo de próton em diminuir acentuadamente no fim do alcance da partícula. Esta ideia foi declarada, porém não foi checada. No presente trabalho, esta hipótese foi investigada usando simulação Monte Carlo, com o auxílio do software SRIM-2008, para o caso de um detector que registra pequena parte da energia de saída do próton (detector fino). Primeiramente, foi determinada a espessura do detector para a investigação da perda de energia dos prótons em um objeto de água com 20 cm. Foi estudada a dose absorvida necessária para distinguir dois objetos de água com espessura de 20 cm e 20,1 cm, como função da energia inicial. O resultado foi comparado com os dados obtidos por um detector grosso, que registra energia total de saída. A investigação realizada mostrou que a ideia principal de se trabalhar no fim de alcance dos prótons para a construção da imagem, não é correta quando se utiliza detector fino bem como detector grosso. Em geral, os sistemas de detecção, baseados em medidas de energia total de saída, são preferíveis àqueles que se baseiam em detectores finos.<br>Since the earliest works proposing the use of protons for imaging, the main advantage of protons over X-rays was expected to be a result of the specific property of the proton flux dropping off very steeply at the end of the particle range. This idea was declared but was not checked. In the present work, this assumption was investigated using the Monte Carlo simulation, with the help of SRIM-2008 software, for the case of a detector that registers only a small part of the proton exit energy (thin detector). First, a thickness of such detector was determined for the investigated water object of 20 cm. Then, it was studied the minimum dose absorbed by an object that is necessary to distinguish two water objects with the thicknesses of 20 cm and 20.1 cm as a function of the initial energy. The results were compared with the case of thick detector that registers the total exit energy. The carried out investigation shows that the principal idea of proton imaging to work near the end of the proton range is not correct in the case of thin detector as well as for thick one. In general, detector systems based on the measurement of the total exit energy are preferable to systems based on thin detectors.

Le développement de nouvelles techniques en radiothérapie externe ouvre de nouvelles voies dans la recherche de gain de précision dans la distribution de dose en passant notamment par la connaissance du mouvement pulmonaire. La simulation numérique NEMOSIS (Neural Network Motion Simulation System) basée sur les Réseaux de Neurones Artificiels (RNA) développée ici permet, en plus de déterminer de façon personnalisée le mouvement, de réduire les doses nécessaires initiales pour le déterminer. La première partie présente les techniques actuelles de traitement, les mouvements pulmonaires ainsi que les méthodes de simulation ou d'estimation du mouvement déjà existantes. La seconde partie décrit le réseau de neurones artificiels utilisé et les étapes de son paramétrage pour répondre à la problématique posée. Une évaluation précise de notre approche a été réalisée sur des données originales. Les résultats obtenus sont comparés avec une méthode d'estimation du mouvement. Les temps de calcul extrêmement faibles, de l'ordre de 7 millisecondes pour générer une phase respiratoire, ont permis d'envisager son utilisation en routine clinique. Des modifications sont apportées à NEMOSIS afin de répondre aux critères de son utilisation en radiothérapie externe et une étude sur le mouvement de contours tumoraux est effectuée. Ces travaux ont mis en place les bases de la simulation du mouvement pulmonaire par RNA et ont validé cette approche. Son exécution en temps réel couplé à la précision de la prédiction fait de NEMOSIS un outil prometteur dans la simulation du mouvement synchronisé avec la respiration.

A mamografia é a principal técnica radiográfica utilizada para imagens das mamas. A mama é um órgão radio-sensível e existe um risco de câncer induzido pela radiação associado com a mamografia. A dose glandular média (DG) é a grandeza dosimétrica aceita para caracterizar o risco de câncer induzido pela radiação. Estudos anteriores concluíram que a DG depende não somente do conteúdo glandular, mas também da distribuição do tecido glandular dentro da mama. Neste trabalho, é proposto um novo método para a geração de modelos computacionais de mama com tecido epitelial e distribuição espacial do tecido glandular de pacientes submetidos a exames de mamografia digital. Cento e setenta modelos computacionais de mama com diferentes glandularidades foram simulados e os resultados foram comparados com valores de coeficientes de conversão para DG recomendados. As diferenças encontradas sugerem que os coeficientes de conversão recomendados internacionalmente podem superestimar a dose glandular média para mamas menos densas e subestimar a dose glandular média para as mamas mais densas. A metodologia descrita neste trabalho constitui uma ferramenta poderosa para a dosimetria em mamografia, principalmente para estudos de risco.<br>Mammography is the main radiographic technique used for breast imaging. A major concern with mammographic imaging is the risk of radiation-induced breast cancer due to the high sensitivity of breast tissue. The mean glandular dose (DG) is the dosimetric quantity widely accepted to characterize the risk of radiation induced cancer. Previous studies have concluded that DG depends not only on the breast glandular content but also on the spatial distribution of glandular tissue within the breast. In this work, a new method for generating breast models featuring skin composition and glandular tissue distribution from patients undergoing digital mammography is proposed. A hundred and seventy breast models with different breast glandularity were simulated and the results were compared with those obtained from recommended DG conversion factors. The results show that the internationally recommended conversion factors may be overestimating the mean glandular dose to less dense breasts and underestimating the mean glandular dose for denser breasts. The methodology described in this work constitutes a powerful tool for breast dosimetry, especially for risk studies.

En neuroradiologie interventionnelle, la machine d'angiographie RX est le moyen d'imagerie par excellence pour guider le médecin dans l'accomplissement de son<br />geste thérapeutique. Cette machine permet l'acquisition d'une image 3D montrant les artères du patient (ou 3DXA). Cependant, le contrôle visuel du déploiement des<br /> outils (guide, cathéter...) est effectué en 2D avec une image temps réel (ou fluoroscopie). Cette thèse a pour ambition de contribuer à l'amélioration des techniqu<br />es de guidage en proposant des outils permettant une utilisation de l?image 3D durant le traitement.<br /><br />Les images étant acquises avec la même machine d'angiographie, nous avons consacré une partie de notre travail au développement de méthodes fiables de calibrage de<br /> la chaîne image portée par l'arceau rotatif de la machine. Le but était de comprendre si l'arceau se déformait ou non sous l'influence de son poids.<br /><br />Tirant parti du fait que les images sont acquises avec la même machine, nous avons ensuite proposé une méthode de recalage 3D2D entre l'image 3DXA et la fluorosc<br />opie. Cette méthode exploite les capteurs de position du système et incorpore les déformations subies par le système.<br /><br />Suite à ces travaux, un système permettant la fusion de l'image 3DXA avec la fluoroscopie a été développé en collaboration avec GE Healthcare et évalué au CHU de Nanc<br />y pour le traitement des anévrismes cérébraux.<br /><br />Enfin, un nouveau système doté de deux chaînes images (ou système bi-plan) a été installé à Nancy durant notre thèse. Après avoir développé une méthode de détection 2D du g<br />uide dans les images fluoroscopiques, nous avons initié une première étude de la reconstruction 3D du guide à partir des images bi-plan.

Les modèles humains numériques sont devenus des outils indispensables à l’étude de la posture est du mouvement dans de nombreux domaines de la biomécanique visant des applications en ergonomie ou pour la clinique. Ces modèles intègrent une représentation géométrique de la surface du corps et un squelette filaire interne composé de segments rigides et d’articulations assurant leur mise en mouvement. La personnalisation des mannequins s'effectue d’abord sur les dimensions anthropométriques externes, servant ensuite de données d’entrée à l’ajustement des longueurs des segments du squelette en interne. Si les données externes sont de plus en plus facilement mesurables à l’aide des outils de scanning 3D actuels, l’enjeu scientifique est de pouvoir prédire des points caractéristiques du squelette en interne à partir de données uniquement externes. L’Institut de Biomécanique Humaine Georges Charpak (Arts et Métiers ParisTech) a développé des méthodes de reconstruction des os et de l’enveloppe externe à partir de radiographies biplanes obtenues avec le système basse dose EOS. En s’appuyant sur cette technologie, ces travaux ont permis de proposer de nouvelles relations statistiques externes-internes pour prédire des points du squelette longitudinal, en particulier l’ensemble des centres articulaires du rachis, à partir d’une base de données de 80 sujets. L'application de ce travail pourrait permettre d’améliorer le réalisme des modèles numériques actuels en vue de mener des analyses biomécaniques, principalement en ergonomie, nécessitant des informations dépendant de la position des articulations comme les mesures d’amplitude de mouvement et de charges articulaires<br>Digital human models have become instrumental tools in the analysis of posture and motion in many areas of biomechanics, including ergonomics and clinical settings. These models include a geometric representation of the body surface and an internal linkage composed of rigid segments and joints allowing simulation of human movement. The customization of human models first starts with the adjustment of external anthropometric dimensions, which are then used as input data to the adjustment of internal skeletal segments lengths. While the external data points are more readily measurable using current 3D scanning tools, the scientific challenge is to predict the characteristic points of the internal skeleton from external data only. The Institut de Biomécanique Humaine Georges Charpak (Arts et Métiers ParisTech) has developed 3D reconstruction methods of bone and external envelope from biplanar radiographs obtained from the EOS system (EOS Imaging, Paris), a low radiation dose technology. Using this technology, this work allowed proposing new external-internal statistical relationships to predict points of the longitudinal skeleton, particularly the complete set of spine joint centers, from a database of 80 subjects. The implementation of this work could improve the realism of current digital human models used for biomechanical analysis requiring information of joint center location, such as the estimation of range of motion and joint loading

CAPES<br>A dosimetria em tomografia computadorizada (TC) envolve desde a determinação de grandezas dosimétricas específicas de TC até a estimativa de dose absorvida e dose efetiva. Entretanto, deve-se considerar que por envolver radiação ionizante no seu processo, este procedimento apresenta riscos inerentes e sua utilização deve ponderar o custo e o benefício propiciado pelo procedimento. A proteção de pacientes submetidos a exames radiológicos, de uma maneira geral, é determinada pelos princípios da “justificação” e “otimização”. Desta forma, torna-se importante o conhecimento dos níveis de radiação nas exposições durante um procedimento tomográfico. Estes níveis foram observados através da estimativa das grandezas específicas para tomografia computadorizada, tais como o Índice de kerma no ar (C100,ar), o Índice de kerma ponderada (Cw) e o produto kerma comprimento (PKL,CT), e em estimados os níveis de dose efetiva e risco para o estudo tomográfico computadorizado de crânio, tórax e abdômen realizado com múltipla varredura. Os valores obtidos foram comparados com os obtidos por simulação computacional por Monte Carlo. Eles foram utilizados, neste estudo, no cálculo da Dose Efetiva e risco e para comparação com o nível de referência de dose estabelecido pela Comunidade Européia. Utilizando-se o programa de simulação computacional Dosecal X_CT e o protocolo ICRP 103, foram determinadas as grandezas de radioproteção relevantes para o estudo que são os valores de dose efetiva referente ao procedimento. O nível do Produto kerma comprimento (PKL,CT), utilizando-se o C100,ar previamente estabelecido, foi também obtido e comparado com o nível de referência de dose estabelecido pela comunidade europeia. Os valores encontrados até agora estão dentro dos limites dos Níveis de Referência.<br>The computed tomography (CT) dosimetry involves measurements of specific quantities of CT, which are part of CT quality control procedures, as well as calculation of absorbed and effective doses to a patient submitted to CT examinations. Since CT uses ionizing radiation, it should be considered that a precise balance between risks and benefits must be achieved in order to justify the adoption of such technique. Radiation protection of patients undergoing radiological exams is established based on the justification and optimization principles. Nowadays, it is important to know the dose radiation levels to which a patient is exposed during a tomographic procedure. Those are given by the estimation of specific dosimetric quantities called the computed tomography kerma index in air, C100,air, the weighted computed tomography kerma index Cw, kerma length product, PKL,CT and then the levels of effective dose and risk to the computerized CT scan study of skull, thorax and abdomen with a multiple scanning. The values obtained were compared with those obtained by computer simulation using Monte Carlo method. The protection quantities organ absorbed dose, effective dose and risk for comparison with the reference dose level established by the European Community. Using computational simulation program of the Dosecal X_CT and the ICRP 103 Protocol, were certain quantities of radiation protection relevant to study which are the values of effective dose for the procedure. The dose length product PKL,CT level was calculated from the C100,air and compared to the reference dose level established by the European Community. The values found so far are within the limits of the reference levels.

This paper describes a MATLAB application with the main purpose of the simulation of noise components and noise elimination methods of Digital Radiography. The main parts of simulator are the model of a scene, procedures for loading the noise components to image data and methods for image processing. Various methods are employed depending on the type of noise. Subtraction techniques are used for the elimination of structural noise. The physical noise suppression is obtained using several methods of cumulation and Pixel Shift is used to reduce motion artifacts caused by the existence of moving noise. The techniques of superposition highlight the areas of interest in an image. Included are also auxiliary procedures for simulator running and presentation of final data. The model and the presented application can be used mainly for educational purposes as a powerful didactic tool.

L'un des principaux défis de l'interprétation radiographique réside dans la compréhension de l’anatomie radiographique, laquelle est intrinsèquement liée à la disposition tridimensionnelle des structures anatomiques et à l’impact du positionnement du tube radiogène vis-à-vis de ces structures lors de l'acquisition de l'image. Traditionnellement, des radiographies obtenues selon des projections standard sont employées pour enseigner l'anatomie radiographique en médecine vétérinaire. La tomodensitométrie − ou communément appelée CT (Computed Tomography) − partage plusieurs des caractéristiques de la radiographie en ce qui a trait à la génération des images. À l’aide d'un plug-in spécialement développé (ORS Visual ©), la matrice contenant les images CT est déformée pour reproduire les effets géométriques propres au positionnement du tube et du détecteur vis-à-vis du patient radiographié, tout particulièrement les effets de magnification et de distorsion. Afin d'évaluer le rendu des images simulées, différentes régions corporelles ont été imagées au CT chez deux chiens, un chat et un cheval, avant d'être radiographiées suivant des protocoles d'examens standards. Pour valider le potentiel éducatif des simulations, dix radiologistes certifiés ont comparé à l'aveugle neuf séries d'images radiographiques simulées aux séries radiographiques standard. Plusieurs critères ont été évalués, soient le grade de visualisation des marqueurs anatomiques, le réalisme et la qualité radiographique des images, le positionnement du patient et le potentiel éducatif de celles-ci pour différents niveaux de formation vétérinaire. Les résultats généraux indiquent que les images radiographiques simulées à partir de ce modèle sont suffisamment représentatives de la réalité pour être employées dans l’enseignement de l’anatomie radiographique en médecine vétérinaire.<br>Understanding radiographic anatomy, which is intimately linked to the comprehension of tridimensional anatomy and the impact of patient, radiographic tube and x-ray detector positioning, represents a challenge for students. Traditionally, radiographs obtained under specific angles of projection have been used for teaching radiographic anatomy. Computed tomography (CT) shares several features with radiography with regard to image production. A plug-in was developed for a DICOM viewer (ORS visual ©) simulating radiographs using CT datasets. This plug-in distorts the CT image matrix to reproduce the magnification and distortion effects that take place in radiographs due to the variations in radiographic tube, patient and detector positioning and angulation. In order to test this model, specific body parts of two dogs, one cat and one horse were radiographed and CT-scanned. The CT datasets were used to generate a total of nine series of radiographic simulations that could be compared to corresponding standard radiographic projections. Ten board-certified veterinary radiologists blindly scored several parameters in these image series, including the visualization of specific anatomical landmarks, image realism and quality, patient positioning, and the educational potential for students and veterinarians of variable degree of veterinary training Overall results indicate that simulated radiographs are representative enough to be used to teach several concepts of image formation and radiographic anatomy in veterinary radiology.

碩士<br>國立中央大學<br>物理研究所<br>100<br>Proton Range Radiography (PRR) is an image quality control tool in proton therapy. Before a patient begins proton irradiation, a real-time imaging can be generated by the same proton beam. By comparing with the computer tomography image; the doctor can know any changes of tumor to evaluate the validity of the treatment. The PRR detector includes a pair of gas electron multiplier and a range telescope made with a stack of 30 scintillators. The gas electron multiplier is to record the trajectory of the particle and the scintillator is to record the energy loss of particle passing through scintillator. In this thesis, we used these two conditions to reconstruct the image and analysis image resolution from these images. In order to find the minimum dosage require for clean picture, we through a simple algorithm to search the minimum dosage.

No<br>Objective: The use of Computer-based simulation (CBS), a form of simulation which utilises digital and web based platforms, is widely acknowledged in healthcare education. This literature review explores the current evidence relating to CBS activities in supporting radiographer education in CT and MRI. Key findings: Journal articles published between 2010 and 2020 were reviewed (n ¼ 663). The content was evaluated and summarised with the following headings; current utility, overview of CBS types, knowledge acquisition and evaluation, and student perspective. CBS utility in CT and MRI radiography education is limited. Its current use is for pre-registration education, and the interfaces used vary in design but are predominantly used as a preclinical learning tool to support the training of geometric scan planning, image acquisition and reconstruction, and associated technical skills. CBS was positively acknowledged by student radiographers; based on its inherent flexibility, self-paced learning and the ability to practice in a safe environment. Nonetheless, the educational validation of CBS in CT and MRI education pertaining to knowledge and skill acquisition has not been fully assessed through rigorous academic assessments and metrics. Conclusion: The current use of CBS in CT and MRI education is limited. The development of software programmes with functionality and capability that correlates with current clinical practice is imperative; and to enable more research in CBS utility to be undertaken to establish the efficacy of this pedagogical approach. Implications for practice: Due to limited placement opportunities, the use of simulation is increasing and evolving; in line with the approach to design and deliver high quality Simulation Based Education (SBE) in Diagnostic Radiography education. The continued development, utility and evaluation of CBS interfaces to support student radiographers at pre and post registration level is therefore essential.

Yes<br>Introduction: Whilst there is increasing demand on radiology services in the UK, pressures are restricting the expansion of the multi-professional workforce. A pilot academy for radiography reporting was established to augment the traditional university and clinical education in a simulated environment using focussed teaching and real image worklists in a dedicated environment away from departments. Methods: Located at a facility to replicate the clinical reporting environment, the emphasis of the nine-month pilot was to provide extensive ‘hands-on’ training to eight trainees. Evaluation of the academy was undertaken through focus groups, telephone interviews, and online surveys to consider the experiences of the trainees and their managers and mentors. Results: There was overwhelming support for the academy from trainees, mentors, and managers. Key benefits included relieving pressures on department and mentors; providing an intense, structured, and safe environment to learn; and, perhaps most importantly, an extensive and cohesive peer-support network. Issues identified included conflict within departments due to differences in reporting style and the need for greater collaboration between the university, academy, and departments. Conclusion: The use of simulation in education is widely researched, however, there are a number of key factors that need to be considered when implementing it into practise. Peer-support and reflection is seen as essential for its success. Extensive dedicated time to focus on reporting alongside peers can support the development of these skills away from the clinical environment and as such can reduce pressure on service delivery and positively influence learner outcomes.<br>The pilot academy received funding from the NHS Vanguard scheme (Working Together NHS Vanguard) for purchase of equipment and the salaries of the clinical educator (pro-rata). The university fees and backfill payments were funded by Health Education England.

Inledning: Att utbilda röntgensjuksköterskestudenter att ta bilder med i första hand konventionell röntgen har alltid varit svårt då man har varit hänvisad till fantomer eller rollspel på grund av strålningen. Med utvecklingen av VR uppkommer dock frågan om det kan bidra som komplement till utbildningen. Syfte: Syftet med uppsatsen var att beskriva om VR kan användas i röntgensjuksköterskeutbildningen. Metod: Studien utfördes som en allmän litteraturöversikt där en kvalitativ och nio kvantitativa vetenskapliga artiklar ingick i analysen efter sökningar i 3 databaser. Resultat: Det mesta pekade på att VR kan vara ett bra komplement i röntgensjuksköterskeutbildningen då studenterna uppvisade större säkerhet i positionering och ökat självförtroende vid bildtagning. Dock upplevdes inlärningskurvan av VR som tidskrävande och behov av mer forskning inom området finns. Resultatet visade även att utbildningsanordnaren måste anpassa och utveckla hur de lär ut. Slutsats: Många studenter är efter avslutade studier osäkra på röntgenbildtagning när de kommer till sin första arbetsplats. Kan användandet av VR i röntgensjuksköterskeutbildningen leda till en högre nivå av självsäkerhet och potentiellt ökad kunskap om positionering får det anses vara fördelaktigt.<br>Introduction: Teaching radiography students to be able to take good quality x-ray images has always been a problem since they have been referred to phantoms or roleplay given the radiation. With the development of VR, the question has been raised if that could serve as a complement. Aim: The purpose of this study was to describe if VR could be used during radiography education. Method: The study was conducted as a literature review where one qualitative and nine quantitative scientific articles were analysed after searching 3 databases. Result: Most findings point to that VR might be a good compliment to regular radiography education since the students exhibit improved patient positioning and enhanced self-confidence while taking x-ray images. The learning curve with VR was however deemed as somewhat time consuming and more research seems necessary. The result also shows that the provider of the education must adapt and develop the way they teach. Conclusion: Many students feel insecure about x-ray imaging upon starting at their first place of work. If the use of VR in radiography education might raise the level of self-confidence as well as potentially increase the knowledge of positioning it might be considered beneficial.

Trabalho de Projeto do Mestrado Integrado em Engenharia Biomédica apresentado à Faculdade de Ciências e Tecnologia<br>Cancer is one of the biggest public health issues, being the second leading cause of death worldwide. More than 100 different types of cancer have been reported until the present day, and there are several approaches to combat them. One of them, a type of radiation therapy, is proton therapy. Image-guidance is crucial for improving the precision of experiments such as small-animal proton irradiation. This project investigated the feasibility of a novel small-animal proton radiographic imaging setup based on a pixelated silicon detector and the concept of energy stacking. The method for position verification, and direct measurement of the two-dimensional Water Equivalent Thickness (WET) distribution of a small-animal sized object, consisted of scanning a 2 x 2 x 2 cubic centimeter water phantom, including four different material inserts (0.3 x 0.3 x 2 cubic centimeter each), with 61 proton energies ranging from 45 to 75 MeV. Different beam shapes were tested. Energy deposition was scored in a simplified model of a commercially available pixelated CMOS detector (1024 x 512 pixels with 48 μm pixel pitch, 2 μm sensitive thickness), placed at varying distances downstream of the phantom. For each detector pixel, the energy deposition versus initial proton beam energy was recorded and converted to WET values by a signal decomposition using a Monte Carlo-based lookup-table (LUT) matrix. For a set of Monte Carlo simulation parameters, including the use of a combination of two perpendicular rectangular proton beams, WET values were obtained with an accuracy of about 96.1%. Sub-millimeter spatial resolutions of (0.59 ± 0.02) mm and (0.67 ± 0.01) mm were obtained in both dimensions. With a total dose to the object of 69.7 mGy, the entire radiographic acquisition time for this approach was estimated to range from 14 to 33 minutes, depending on the used accelerator.<br>O cancro é um dos maiores problemas de saúde pública, sendo a segunda maior causa de morte no mundo. Mais de 100 tipos diferentes de cancro foram reportados até ao presente, existindo diversas abordagens para os combater. Uma delas, um tipo de radioterapia, é a terapia com protões. O conceito de image-guidance é crucial para melhorar a precisão de estudos como a irradiação protónica de pequenos animais. Este projeto investigou a viabilidade de uma nova configuração para imagiologia de radiografias protónicas de pequenos animais baseada num detetor de silício pixelizado e na noção de empilhamento de energias. O método para verificação de posição, e medição direta da distribuição bidimensional da Espessura Equivalente a Água (WET) de um objeto com o tamanho de um pequeno animal, consistiu no varrimento de um fantoma com 2 x 2 x 2 centímetros cúbicos de água, includindo quatro inserções de diferentes materiais (0,3 x 0,3 x 2 centímetros cúbicos cada), com 61 feixes de protões com energias entre 45 e 75 MeV. A deposição de energia foi registada num modelo simplificado de um detetor CMOS pixelizado comercialmente disponível (1024 x 512 pixels com 48 μm de lado e uma espessura sensível de 2 μm), colocado em diferentes distâncias a jusante do fantoma. Para cada pixel do detector, a deposição de energia versus a energia inicial do feixe de protões foi registada e convertida em valores de WET através de uma decomposição de sinal, usando uma matriz de consulta baseada em Monte Carlo (LUT). Para um conjunto de parâmetros de simulação de Monte Carlo, os valores de WET foram obtidos com uma precisão de cerca de 96,1%. Resoluções espaciais sub-milimétricas de (0,59 ± 0,02) mm e (0,67 ± 0,01) mm foram obtidas em ambas as dimensões. Com uma dose total para o objeto de 69,7 mGy, o tempo total de aquisição radiográfica para esta abordagem foi estimado em 14 a 33 minutos, dependendo do acelerador utilizado.

Given a known signal and perfect knowledge of the environment there exist few detection and estimation problems that cannot be solved. Detection performance is limited by uncertainty in the signal, an imperfect model, uncertainty in environmental parameters, or noise. Complex environments such as the ocean acoustic waveguide and the human anatomy are difficult to model exactly as they can differ, change with time, or are difficult to measure. We address the uncertainty in the model or parameters by incorporating their possibilities in our detection algorithm. Noise in the signal is not so easily dismissed and we set out to provide cases in which what is frequently termed a nuisance parameter might increase detection performance. If the signal and the noise component originate from the same system then it might be reasonable to assume that the noise contains information about the system as well. Because of the negative effects of ionizing radiation it is of interest to maximize the amount of diagnostic information obtained from a single exposure. Scattered radiation is typically considered image degrading noise. However it is also dependent on the structure of the medium and can be estimated using stochastic simulation. We describe a novel Bayesian approach to signal detection that increases performance by including some of the characteristics of the scattered signal. This dissertation examines medical imaging problems specific to mammography. In order to model environmental uncertainty we have written software to produce realistic voxel phantoms of the breast. The software includes a novel algorithm for producing three dimensional distributions of fat and glandular tissue as well as a stochastic ductal branching model. The image produced by a radiographic system cannot be determined analytically since the interactions of particles are a random process. We have developed a particle transport software package to model a complete radiographic system including a realistic x-ray spectrum model, an arbitrary voxel-based medium, and an accurate material library. Novel features include an efficient voxel ray tracing algorithm that reflects the true statistics of the system as well as the ability to produce separable images of scattered and direct radiation. Similarly, the ocean environment includes a high degree of uncertainty. A pressure wave propagating through a channel produces a measurable collection of multipath arrivals. By modeling changes in the pressure wave front we can estimate the expected pattern that appears at a given location. For this purpose we have created an ocean acoustic ray tracing code that produces time-domain multipath arrival patterns for arbitrary 3-dimensional environments. This iterative algorithm is based on a generalized recursive ray acoustics algorithm. To produce a significant gain in computation speed we model the ocean channel as a linear, time invariant system. It differs from other ocean propagation codes in that it uses time as the dependent variable and can compute sound pressure levels along a ray path effectively measuring the spatial impulse response of the ocean medium. This dissertation also investigates Bayesian approaches to source localization in a 3-D uncertain ocean environment. A time-domain-based optimal a posteriori probability bistatic source localization method is presented. This algorithm uses a collection of acoustic time arrival patterns that have been propagated through a 3-D acoustic model as the observable data. These replica patterns are collected for a possible range of unknown environmental parameters. Receiver operating characteristics for a bistatic detection problem are presented using both simulated and measured data.<br>Dissertation

Law Tsui Ying.<br>Thesis (M.Phil.)--Chinese University of Hong Kong, 2000.<br>Includes bibliographical references (leaves 64-70).<br>Abstracts in English and Chinese.<br>Acknowledgments --- p.ii<br>Chapter 1 --- Introduction --- p.1<br>Chapter 1.1 --- Structure of Bronchus --- p.3<br>Chapter 1.2 --- Existing Systems --- p.4<br>Chapter 1.2.1 --- Virtual Endoscope System (VES) --- p.4<br>Chapter 1.2.2 --- Virtual Reality Surgical Simulator --- p.4<br>Chapter 1.2.3 --- Automated Virtual Colonoscopy (AVC) --- p.5<br>Chapter 1.2.4 --- QUICKSEE --- p.5<br>Chapter 1.3 --- Organization of Thesis --- p.6<br>Chapter 2 --- Three Dimensional Visualization in Medicine --- p.7<br>Chapter 2.1 --- Acquisition --- p.8<br>Chapter 2.1.1 --- Computed Tomography --- p.8<br>Chapter 2.2 --- Resampling --- p.9<br>Chapter 2.3 --- Segmentation and Classification --- p.9<br>Chapter 2.3.1 --- Segmentation by Thresholding --- p.10<br>Chapter 2.3.2 --- Segmentation by Texture Analysis --- p.10<br>Chapter 2.3.3 --- Segmentation by Region Growing --- p.10<br>Chapter 2.3.4 --- Segmentation by Edge Detection --- p.11<br>Chapter 2.4 --- Rendering --- p.12<br>Chapter 2.5 --- Display --- p.13<br>Chapter 2.6 --- Hazards of Visualization --- p.13<br>Chapter 2.6.1 --- Adding Visual Richness and Obscuring Important Detail --- p.14<br>Chapter 2.6.2 --- Enhancing Details Incorrectly --- p.14<br>Chapter 2.6.3 --- The Picture is not the Patient --- p.14<br>Chapter 2.6.4 --- Pictures-'R'-Us --- p.14<br>Chapter 3 --- Overview of Advanced Segmentation Methodologies --- p.15<br>Chapter 3.1 --- Mathematical Morphology --- p.15<br>Chapter 3.2 --- Recursive Region Search --- p.16<br>Chapter 3.3 --- Active Region Models --- p.17<br>Chapter 4 --- Overview of Centerline Methodologies --- p.18<br>Chapter 4.1 --- Thinning Approach --- p.18<br>Chapter 4.2 --- Volume Growing Approach --- p.21<br>Chapter 4.3 --- Combination of Mathematical Morphology and Region Growing Schemes --- p.22<br>Chapter 4.4 --- Simultaneous Borders Identification Approach --- p.23<br>Chapter 4.5 --- Tracking Approach --- p.24<br>Chapter 4.6 --- Distance Transform Approach --- p.25<br>Chapter 5 --- Automated Extraction of Bronchus Area --- p.27<br>Chapter 5.1 --- Basic Idea --- p.27<br>Chapter 5.2 --- Outline of the Automated Extraction Algorithm --- p.28<br>Chapter 5.2.1 --- Selection of a Start Point --- p.28<br>Chapter 5.2.2 --- Three Dimensional Region Growing Method --- p.29<br>Chapter 5.2.3 --- Optimization of the Threshold Value --- p.29<br>Chapter 5.3 --- Retrieval of Start Point Algorithm Using Genetic Algorithm --- p.29<br>Chapter 5.3.1 --- Introduction to Genetic Algorithm --- p.30<br>Chapter 5.3.2 --- Problem Modeling --- p.31<br>Chapter 5.3.3 --- Algorithm for Determining a Start Point --- p.33<br>Chapter 5.3.4 --- Genetic Operators --- p.33<br>Chapter 5.4 --- Three Dimensional Painting Algorithm --- p.34<br>Chapter 5.4.1 --- Outline of the Three Dimensional Painting Algorithm --- p.34<br>Chapter 5.5 --- Optimization of the Threshold Value --- p.36<br>Chapter 6 --- Automatic Centerline Determination Algorithm --- p.38<br>Chapter 6.1 --- Distance Transformations --- p.38<br>Chapter 6.2 --- End Points Retrieval --- p.41<br>Chapter 6.3 --- Graph Based Centerline Algorithm --- p.44<br>Chapter 7 --- Experiments and Discussion --- p.48<br>Chapter 7.1 --- Experiment of Automated Determination of Bronchus Algorithm --- p.48<br>Chapter 7.2 --- Experiment of Automatic Centerline Determination Algorithm --- p.54<br>Chapter 8 --- Conclusion --- p.62<br>Bibliography --- p.63

近年来，血管类疾病已经成为人类健康的第一杀手。每年有成百上千万人死于血管疾病。血管介入术是一种非常有前景的血管类疾病的治疗手段。血管介入术是一种微创手术，它已经被广泛的用于治疗中风，血管狭窄，血管瘤等疾病。相对于传统的开放式手术，它具有风险低，恢复快，住院时间短等优点。该疗法通常在透视影像的引导下由导管和导线在血管内协同完成手术过程。因为介入术的复杂性和特殊性，作为介入手术医生的必要技能，掌握手术中手眼协同，各种手术器具的使用和复杂细致的手术流程无疑是一个巨大的挑战。因此，迫切地需要一种高效、安全的训练系统。相对于传统的训练方法，基于虚拟现实技术的训练系统是一种非常好的训练手段。<br>为了建立一套高仿真的介入手术训练模拟器，首先，我们要为病人的血管网重建三维模型。我们提出了一种自动的提取中心线的方法，用来从分割好的CTA/MRA体数据中获取病人血管网的中心线。基于改进的平行传递算法，沿着这些中心线，生成了一系列连续的标架。根据这些标架，我们构造了血管的横截面，并在此基础上生成了光滑连续的三维血管模型。<br>其次，作为血管介入术中最基础和最重要的手术器械，我们为导管和导线建立了物理模型。我们提出了一种基于最小势能原理的可变形的模型用于模拟导管和导线对于受力的反应。我们还提出了一个快速并且稳定的多网格算法来保证模拟的真实性和严格的实时交互要求。另外，我们做了几组实验。通过这些实验，验证了多网格算法在稳定性、实时性、模拟的真实性等方面满足了我们对于训练用模拟系统的要求。<br>再次，为了模拟血管栓塞术的手术过程，我们提出了一种模拟线圈填充血管瘤的过程的新方法。通过加总线圈弯曲变形的弹性势能、血管瘤变形的弹性势能以及外力做的功，我们建立了在血管栓塞术的环境下的总势能模型。为了求解这个模型，我们提出了一个基于有限元方法的求解器。从而模拟了线圈在介入医生的操作下慢慢的进入血管瘤，并缠绕起来的过程。<br>另外，我们提出了一个分层圆柱网格模型（LCGM）用于模拟在血管网中血流的运动。这一模型在几何上和拓扑结构上都非常适合我们的应用。我们将血液在血管中的流动近似为一维的层流，并用一组线性等式描述了血管网中流速与血压的关系。通过求解这一线性系统，得到了在分层圆柱网格模型下血流的速度场。依据这个血流的速度场，我们采用平流-扩散模型来模拟造影剂在血管中的传播的过程。<br>Vascular diseases have been becoming the number one cause of death worldwide in recent years. Millions of people were killed by vascular diseases each year. An increasingly promising therapy for treating vascular diseases is Vascular Interventional Radiology (VIR). VIR is a minimally invasive surgery (MIS) procedure, which has been widely used to cure stroke, angiostenosis, aneurysm and etc. A low risk, an accelerated recovery and a shorter stay in hospital are important advantages over the traditional vascular surgery. This therapy is performed by a guidewire-catheter combination inside the blood vessels under the guidance of the fluoroscopic imaging. Because of the complexity and particularity of these procedures, it is a great challenge to master hand-eye coordination, instrument manipulation and procedure protocols for each radiologist mandatory. An efficient and safe training system is needed urgently. In contrast to these traditional training methods, virtual reality (VR) based simulation systems is a pretty good surrogate.<br>In order to build a high fidelity interventional simulator for physician training, firstly, we reconstructed the three dimensional (3D) model for the vascular network of the patients. An method of automatic skeleton extraction was proposed to acquire the centerline of the vascular network from the segmented volume data from CTA/MRA. A series of continuing frames were generated along with the centerline based on improved parallel transporting method. According to these frames we built the crossections of the vessels and further the 3D vascular model with the smooth meshes.<br>Secondly, as the most basic and important instruments in the VIR procedure, the catheter and guidewire were modeled and simulated physically. We developed a deformable model to simulate complicated behaviors of guidewires and catheters based on the principle of minimum total potential energy. A fast and stable multigrid solver was proposed to ensure both realistic simulation and real time interaction. A series of experiments were conducted to evaluate our multigrid solver in terms of stability, time performance, the capability of simulating catheter behaviors and the realism of catheter deformation.<br>Thirdly, to simulate the procedure of embolization, we proposed a novel method to simulate the motion of coil and their interactions with the aneurysm. We formulated the total potential energy in the embolization circumstance by summing up the elastic energy deriving from the bending of coils, the potential energy due to the deformation of the aneurysm and the work by the external forces. A novel FEM-based approach was proposed to simulate the deformation of coils. And the motion of coils and their responses to every input from the interventional radiologist can be calculated globally.<br>Fourthly, we proposed our Layered Cylindrical Gird Model (LCGM) for simulating blood flow in vascular network, which is pretty suitable for sampling the vascular network geometrically and topologically. The blood flow in vessels was regarded as 1D laminar flow and formulated into a set of linear equations based on the Poiseuille law to describe the relationship between the speed of flow and the pressure. Solving those equations, we got the velocity fields in the blood flow. In terms of the velocity fields, an advection-diffusion model was adopted to simulate the propagation of contrast agent with the blood flow.<br>Finally, all above techniques and procedures were implemented and integrated into a simulation system for training the medical students to acquire the endovascular skill, and an empirical study was also designed based on a typical selective catheteriza- tion procedure to assess the feasibility and effectiveness of the proposed system.<br>Detailed summary in vernacular field only.<br>Detailed summary in vernacular field only.<br>Detailed summary in vernacular field only.<br>Detailed summary in vernacular field only.<br>Detailed summary in vernacular field only.<br>最后，我们将所有以上提到的技术和方法集成到模拟系统中用于训练医学院的学生，并使他们获得血管介入术的技能。并且，我们基于一个典型的导管插入术过程，使用经验分析的方法对模拟系统的可用性和效率进行了评估。<br>Li, Shun.<br>Thesis (Ph.D.)--Chinese University of Hong Kong, 2012.<br>Includes bibliographical references (leaves 105-116).<br>Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web.<br>Abstracts in also Chinese.<br>Abstract --- p.i<br>Acknowledgement --- p.vi<br>Chapter 1 --- Introduction --- p.1<br>Chapter 2 --- Vascular Modeling --- p.14<br>Chapter 2.1 --- Introduction and Related Work --- p.14<br>Chapter 2.2 --- Vascular Skeleton Graph Construction --- p.15<br>Chapter 2.2.1 --- Chamfer distance transform and Dijkstra's shortest-path algorithm --- p.17<br>Chapter 2.2.2 --- End vertices retrieval --- p.19<br>Chapter 2.2.3 --- The algorithm of vascular skeleton extraction --- p.21<br>Chapter 2.3 --- Vascular Modeling --- p.21<br>Chapter 2.3.1 --- Tubular Model --- p.21<br>Chapter 2.3.2 --- Bifurcation Model --- p.23<br>Chapter 3 --- Catheter Simulation --- p.28<br>Chapter 3.1 --- Introduction and Related Works --- p.28<br>Chapter 3.2 --- Catheter Simulation --- p.31<br>Chapter 3.2.1 --- Kirchhoff Theory of Elastic Rod --- p.32<br>Chapter 3.2.2 --- Problem Formulation --- p.34<br>Chapter 3.2.3 --- The Multigrid Iterative Solver --- p.38<br>Chapter 3.3 --- Collision detection --- p.45<br>Chapter 3.4 --- Validation of the Catheter Simulation Method --- p.47<br>Chapter 3.4.1 --- Stability --- p.49<br>Chapter 3.4.2 --- Time Performance --- p.50<br>Chapter 3.4.3 --- Preservation of Curved Tip --- p.51<br>Chapter 3.4.4 --- The realism of catheter deformation --- p.53<br>Chapter 4 --- Coil Embolization Simulation --- p.59<br>Chapter 4.1 --- Introduction and Related Work --- p.59<br>Chapter 4.2 --- Methodology --- p.61<br>Chapter 4.2.1 --- Total potential energy of a coil --- p.61<br>Chapter 4.2.2 --- The FEM-based numeric solver for interactive coil simulation --- p.61<br>Chapter 5 --- Angiography Simulation --- p.70<br>Chapter 5.1 --- Introduction and related works --- p.70<br>Chapter 5.2 --- The Equations of Fluid --- p.72<br>Chapter 5.3 --- Layered Cylindrical Gird Model --- p.73<br>Chapter 5.4 --- Numerical Method --- p.76<br>Chapter 5.4.1 --- Evaluation of the velocity field of blood flow --- p.76<br>Chapter 5.4.2 --- Evaluation of the density field --- p.78<br>Chapter 5.5 --- Results --- p.81<br>Chapter 6 --- System Implementation and Evaluation --- p.84<br>Chapter 6.1 --- Introduction and Related Work --- p.84<br>Chapter 6.2 --- System Construction --- p.85<br>Chapter 6.3 --- Empirical Study of the Training System --- p.89<br>Chapter 7 --- Conclusion and Discussion --- p.98<br>Chapter 7.1 --- Geometric Modeling of Vasculature --- p.99<br>Chapter 7.2 --- Catheterization Simulation --- p.99<br>Chapter 7.3 --- Embolization Simulation --- p.100<br>Chapter 7.4 --- Angiography Simulation --- p.101<br>Chapter 7.5 --- System and Evaluation --- p.102<br>Publication List --- p.103<br>Bibliography --- p.105