To aid diagnosis, treatment planning, and patient education, clinicians require tools to anal- yse and explore the increasingly large three-dimensional (3-D) datasets generated by modern medical scanners. Direct volume rendering is one such tool finding favour with radiologists and surgeons for its photorealistic representation. More recently, volume illustration — or non-photorealistic rendering (NPR) — has begun to move beyond the mere depiction of data, borrowing concepts from illustrators to visually enhance desired information and suppress un- wanted clutter. Direct volume rendering generates images by accumulating pixel values along rays cast into a 3-D image. Transfer functions allow users to interactively assign material properties such as colour and opacity (a process known as classification). To achieve real-time framerates, the rendering must be accelerated using a technique such as 3-D texture mapping on commod- ity graphics processing units (GPUs). Unfortunately, current methods do not allow users to intuitively enhance regions of interest or suppress occluding structures. Furthermore, addi- tional scalar images describing clinically relevant measures have not been integrated into the direct rendering method. These tasks are essential for the effective exploration, analysis, and presentation of 3-D images. This body of work seeks to address the aforementioned limitations. First, to facilitate the research program, a flexible architecture for prototyping volume illustration methods is pro- posed. This program unifies a number of existing techniques into a single framework based on 3-D texture mapping, while also providing for the rapid experimentation of novel methods. Next, the prototyping environment is employed to improve an existing method—called tagged volume rendering — which restricts transfer functions to given spatial regions using a number of binary segmentations (tags). An efficient method for implementing binary tagged volume rendering is presented, along with various technical considerations for improving the classifi- cation. Finally, the concept of greyscale tags is proposed, leading to a number of novel volume visualisation techniques including position modulated classification and dynamic exploration. The novel methods proposed in this work are generic and can be employed to solve a wide range of problems. However, to demonstrate their usefulness, they are applied to a specific case study. Ischaemic heart disease, caused by narrowed coronary arteries, is a leading healthconcern in many countries including Australia. Computed tomography angiography (CTA) is an imaging modality which has the potential to allow clinicians to visualise diseased coronary arteries in their natural 3-D environment. To apply tagged volume rendering for this case study, an active contour method and minimal path extraction technique are proposed to segment the heart and arteries respectively. The resultant images provide new insight and possibilities for diagnosing and treating ischaemic heart disease.

L’imagerie du petit animal est incontournable pour le développement des recherches dans les secteurs de labiologie, de la médecine et de l’industrie pharmaceutique. Parmi les différentes modalités d’imagerie développéeschez l’humain et adaptées à l’animal, l’imagerie tomographique à rayons X est devenue une référencepour l’analyse des caractères anatomiques et phénotypiques chez la souris. Elle permet de réaliser des étudeslongitudinales in vivo et des analyses haute résolution ex vivo de façon non invasives et en 3D. L’analyse deces images 3D nécessite des outils spécifiques à chaque problématique.Dans ce contexte, notre travail de thèse a permis d’apporter des contributions sur les thématiques suivantes :1. le développement d’outils d’analyse, à la fois quantitatifs et qualitatifs, pour l’imagerie des tissusminéralisés et adipeux2. l’application des méthodologies développées à des problématiques de recherche biomédicale3. l’étude comparative et "multi-échelle" de différentes technologies de tomographie X pour l’imagerie dupetit animal4. la mise au point d’une méthode originale par résonnance paramagnétique électronique pour la dosimétried’un acte d’imagerie X chez le petit animalEn conclusion, les outils d’imagerie 3D que nous avons développés représentent un nouvel apport pour la dissectionvirtuelle de l’animal de laboratoire, permettant l’exploration de nombreux tissus et organes et rivalisantavec les méthodes d’histologie et de microscopie électronique.L’application de ces méthodes d’imagerie pour la recherche fondamentale et pré-clinique ouvre la perspectived’une alternative nouvelle dans l’expérimentation animale. / Small animal imaging is highly necessary for the development of biomedical research and pharmaceuticalapplications. Amongst various available imaging methods, X-Ray tomography is now considered as a goldstandard for anatomical and phenotypical analysis in mice. CT imaging allows non invasive longitudinal studiesin vivo and high resolution analysis ex vivo. The 3D image analysis requires the development of specific toolsdepending on the biomedical problematics.In this context, we have investigated the following research areas :1. Development of 3D image tools for qualitative and quantitative image analysis of mineralized andadipose tissues in murin models2. Application of our tools to biomedical investigations3. Comparative and multi-scale analysis of various tomography technologies for small animal imaging4. Development of an original method using Electronic Paramagnetic Resonance (EPR) for X-ray dosimetryin miceIn conclusion, our 3D imaging methods are potentially of high interest for the virtual dissection of laboratoryanimals, allowing extended analysis of various tissues and organs complementary to standard histological andmicroscopic approaches.

Spectral computed tomography (CT) is a novel medical imaging technique that involves simultaneously counting photons at several energy levels of the x-ray spectrum to obtain a single multi-variate dataset. Visualisation of such data poses significant challenges due its extremely large size and the need for interactive performance for scientific and medical end-users. This thesis explores the properties of spectral CT datasets and presents two algorithms for GPU-accelerated real-time rendering from compressed spectral CT data formats. In addition, we describe an optimised implementation of a volume raycasting algorithm on modern GPU hardware, tailored to the visualisation of spectral CT data.

Medical imaging is an important part of the clinical workflow. With the increasing amount and complexity of image data comes the need for automatic (or semi-automatic) analysis methods which aid the physician in the exploration of the data. One specific imaging technique is angiography, in which the blood vessels are imaged using an injected contrast agent which increases the contrast between blood and surrounding tissue. In these images, the blood vessels can be viewed as tubular structures with varying diameters. Deviations from this structure are signs of disease, such as stenoses introducing reduced blood flow, or aneurysms with a risk of rupture. This thesis focuses on segmentation and visualization of blood vessels, consituting the vascular tree, in angiography images. Segmentation is the problem of partitioning an image into separate regions. There is no general segmentation method which achieves good results for all possible applications. Instead, algorithms use prior knowledge and data models adapted to the problem at hand for good performance. We study blood vessel segmentation based on a two-step approach. First, we model the vessels as a collection of linear structures which are detected using multi-scale filtering techniques. Second, we develop machine-learning based level set segmentation methods to separate the vessels from the background, based on the output of the filtering. In many applications the three-dimensional structure of the vascular tree has to be presented to a radiologist or a member of the medical staff. For this, a visualization technique such as direct volume rendering is often used. In the case of computed tomography angiography one has to take into account that the image depends on both the geometrical structure of the vascular tree and the varying concentration of the injected contrast agent. The visualization should have an easy to understand interpretation for the user, to make diagnostical interpretations reliable. The mapping from the image data to the visualization should therefore closely follow routines that are commonly used by the radiologist. We developed an automatic method which adapts the visualization locally to the contrast agent, revealing a larger portion of the vascular tree while minimizing the manual intervention required from the radiologist. The effectiveness of this method is evaluated in a user study involving radiologists as domain experts.

[pt] Neste trabalho aplicamos o método de oclusão implícita para acelerar o tempo de cálculo e renderização de isosuperfícies em dados volumétricos regulares. Dado um campo escalar contínuo f sobre um domínio D (onde Dé convexo) e um isovalor w, a oclusão implícita explora a continuidadede f para determinar os limites de visibilidades sem a necessidade de calcular a isosuperfície explicitamente. Aplicamos esta técnica para obter também as silhuetas visíveis das isosuperfícies. / [en] In this work we apply the Implicit Occluders method for optimizing the computation and rendering of isosurfaces in regular volumetric data. Given a continuous scalar field f over a domain D and an isovalue w, Implicit Occluders exploits the continuity of f to determine visibility bounds without the need for computing the isosurface explicitly. We apply this technique to obtain also the visible silhouettes of isosurfaces.

Este trabalho apresenta uma nova técnica de visualização que aproveita as vantagens do rendering volumétrico direto e do rendering de superfícies em um ambiente híbrido. O método faz uso de uma pré-visualização sobre o bordo do volume que viabiliza uma interação em tempo real com objetos volumétricos modelados por meio de malhas não estruturadas. Além disso, essa nova abordagem de visualização é paralelizável e pode se acelerada com placas gráficas comuns. / This work presents a new visualization technique that exploits the advantages of direct volume rendering and surface rendering in a hybrid environment. The method developed here makes use of a pre-visualization on the volume boundary to enable real time interaction with unstructured volumetric meshes. Furthermore, this new visualization approach can be implemented on existing parallel architectures and speed up by conventional graphical hardware.

Este projeto de pesquisa trata da elaboração de um sistema de reconstrução de imagens obstétricas de fetos, em aparelhos de ultra-som convencionais, para a visualização dessas imagens em três dimensões utilizando a internet como meio de utilização do sistema, com o principal objetivo de proporcionar aos médicos ginecologistas melhor visualização do formato e das estruturas internas, e em especial da face do feto, através do processo de reconstrução tridimensional feito a partir de um conjunto de imagens bidimensionais capturadas em aparelhos convencionais de ultra-som. O uso clínico deste projeto está previsto para o setor de obstetrícia do Hospital das Clínicas de Ribeirão Preto. / This project of research deals with the laboration of a reconstruction system of obstetrics images of embryos in devices of ultrasound will be conventional the visualization of these images in three dimensions using the internet half of uses of the system, with the main objective provides to the medical gynecologists a better visualization of the format and the internal structures and in special the face of the embryo through the made process of three-dimensional reconstruction from a dataset of captured bi-dimensional images in conventional devices of ultrasound. The clinical uses of this project is foreseen will be the sector of obstetrics of the Hospital of the Clinics of Ribeirão Preto.

Técnicas de visualização volumétrica direta são utilizadas para visualizar e explorar volumes de dados complexos. Dados volumétricos provêm de diversas fontes, tais como dispositivos de diagnóstico médico, radares de sensoriamento remoto ou ainda simulações científicas assistidas por computador. Um problema fundamental na visualização volumétrica é a especificação de Funções de Transferência (FTs) que atribuem cor e opacidade aos valores escalares que compõem o volume de dados. Essas funções são importantes para a exibição de características e objetos de interesse do volume, porém sua definição não é trivial ou intuitiva. Abordagens tradicionais permitem a edição manual de pontos de controle que representam a FT a ser utilizada no volume. No entanto, essas técnicas acabam conduzindo o usuário a um processo de “tentativa e erro” para serem obtidos os resultados desejados. Considera-se também que técnicas automáticas que excluem o usuário do processo não são consideradas as mais adequadas, visto que o mesmo deve possuir algum controle sobre o processo de visualização. Este trabalho apresenta uma ferramenta semi-automática e interativa destinada a auxiliar o usuário na geração de FTs de cor e opacidade. A ferramenta proposta possui dois níveis de interação com o usuário. No primeiro nível são apresentados várias FTs candidatas renderizadas como thumbnails 3D, seguindo o método conhecido como Design Galleries (MARKS et al., 1997). São aplicadas técnicas para reduzir o escopo das funções candidatas para um conjunto mais razoável, sendo possível ainda um refinamento das mesmas. No segundo nível é possível definir cores para a FT de opacidade escolhida, e ainda refinar essa função de modo a melhorála de acordo com as necessidades do usuário. Dessa forma, um dos objetivos desse trabalho é permitir ao usuário lidar com diferentes aspectos da especificação de FTs, que normalmente são dependentes da aplicação em questão e do volume de dados sendo visualizado. Para o rendering do volume, são exploradas as capacidades de mapeamento de textura e os recursos do hardware gráfico programável provenientes das plácas gráficas atuais visando a interação em tempo real. Os resultados obtidos utilizam volumes de dados médicos e sintéticos, além de volumes conhecidos, para a análise da ferramenta proposta. No entanto, é dada ênfase na especificação de FTs de propósito geral, sem a necessidade do usuário prover um mapeamento direto representando a função desejada. / Direct volume rendering techniques are used to visualize and explore large scalar volumes. Volume data can be acquired from many sources including medical diagnoses scanners, remote sensing radars or even computer-aided scientific simulations. A key issue in volume rendering is the specification of Transfer Functions (TFs) which assign color and opacity to the scalar values which comprise the volume. These functions are important to the exhibition of features and objects of interest from the volume, but their specification is not trivial or intuitive. Traditional approaches allow the manual editing of a graphic plot with control points representing the TF being applied to the volume. However, these techniques lead the user to an unintuitive trial and error task, which is time-consuming. It is also considered that automatic methods that exclude the user from the process should be avoided, since the user must have some control of the visualization process. This work presents a semi-automatic and interactive tool to assist the user in the specification of color and opacity TFs. The proposed tool has two levels of user interaction. The first level presents to the user several candidate TFs rendered as 3D thumbnails, following the method known as Design Galleries (MARKS et al., 1997). Techniques are applied to reduce the scope of the candidate functions to a more reasonable one. It is also possible to further refine these functions at this level. In the second level is permitted to define and edit colors in the chosen TF, and refine this function if desired. One of the objectives of this work is to allow users to deal with different aspects of TF specification, which is generally dependent of the application or the dataset being visualized. To render the volume, the programmability of the current generation of graphics hardware is explored, as well as the features of texture mapping in order to achieve real time interaction. The tool is applied to medical and synthetic datasets, but the main objective is to propose a general-purpose tool to specify TFs without the need for an explicit mapping from the user.

<p>Contrast-enhanced magnetic resonance angiography (MRA) is used to obtain images showing the vascular system. To detect stenosis, which is narrowing of for example blood vessels, maximum intensity projection (MIP) is typically used. This technique often fails to demonstrate the stenosis if the projection angle is not suitably chosen. To improve identification of this region a color-coding algorithm could be helpful. The color should be carefully chosen depending on the vessel diameter. </p><p>In this thesis a segmentation to produce a binary 3d-volume is made, followed by a distance transform to approximate the Euclidean distance from the centerline of the vessel to the background. The distance is used to calculate the smallest diameter of the vessel and that value is mapped to a color. This way the color information regarding the diameter would be the same from all the projection angles. </p><p>Color-coded MIPs, where the color represents the maximum distance, are also implemented. The MIP will result in images with contradictory information depending on the angle choice. Looking in one angle you would see the actual stenosis and looking in another you would see a color representing the abnormal diameter.</p>

<p>Volume rendering (VRT) has been used with great success in studies of patients using computed tomography (CT), much because of the possibility of standardizing the rendering protocols. When using magnetic resonance imaging (MRI), this procedure is considerably more difficult, since the signal from a given tissue can vary dramatically, even for the same patient. This thesis work focuses on how to improve the presentation of MRI data by using VRT protocols including standardized transfer functions. The study is limited to exclusively examining data from patients with suspected renal artery stenosis. A total number of 11 patients are examined.</p><p>A statistical approach is used to standardize the volume rendering protocols. The histogram of the image volume is modeled as the sum of two gamma distributions, corresponding to vessel and background voxels. Parameters describing the gamma distributions are estimated with a Maximum-likelihood technique, so that expectation (E1 and E2) and standard deviation of the two voxel distributions can be calculated from the histogram. These values are used to generate the transfer function.</p><p>Different combinations of the values from the expectation and standard deviation were studied in a material of 11 MR angiography datasets, and the visual result was graded by a radiologist. By comparing the grades, it turned out that using only the expectation of the background distribution (E1) and vessel distribution (E2) gave the best result. The opacity is then defined with a value of 0 up to a signal threshold of E1, then increasing linearly up to 50 % at a second threshold E2, and after that a constant opacity of 50 %. The brightness curve follows the opacity curve to E2, after which it continues to increase linearly up to 100%.</p><p>A graphical user interface was created to facilitate the user-control of the volumes and transfer functions. The result from the statistical calculations is displayed in the interface and is used to view and manipulate the transfer function directly in the volume histogram.</p><p>A transfer function generated with the Maximum-likelihood VRT method (ML-VRT) gave a better visual result in 10 of the 11 cases than when using a transfer function not adapting to signal intensity variations.</p>