No description available.

Segundo a definição original de MCINERNEY & TERZOPOULOS (1995), modelos deformáveis são curvas ou superfícies formadas por pontos conectados que simulam corpos elásticos. Por superarem muitas limitações associadas ao procedimento manual e às técnicas tradicionais de processamento, os contornos deformáveis têm se popularizado. Ainda que o uso dos contornos deformáveis seja vasto e crescente, aspectos relevantes da teoria ainda têm demandado atenção. Muitas referências têm sido feitas às limitações da técnica impostas sobretudo pelo seu processo evolutivo. A convergência a mínimos locais e o agrupamento indesejado de pontos, por exemplo, limitam o emprego da técnica em cenários ruidosos e complexos como os encontrados em reservatórios de petróleo. Esse trabalho apresenta uma abordagem inédita às limitações dos contornos deformáveis. Pela definição de um segundo problema de minimização são definidas distâncias ótimas dos pontos do contorno deformável segundo critério de optimalidade que contempla as particularidades do contorno buscado. Os resultados demonstram que a técnica proposta é provedora de maior enquadramento entre o contorno buscado e o identificado, define solução definitiva aos problemas do agrupamento e espalhamento indesejados de pontos, aumenta a efetividade dos contornos deformáveis em regiões côncavas e, em acréscimo, define metodologia unicamente capaz de dotar os contornos deformáveis de sensibilidade quanto às particularidades de contorno. / According to the original definition of Terzopoulos, deformable models are curves or surfaces formed by connected points that simulate elastic bodies. By overcoming many limitations associated with the manual procedure and the traditional techniques of processing, deformable contours have become popular. Although the use of deformable contours is vast and growing aspects of the theory still demand attention. Many references have been made to the limitations of the technique imposed by the process evolution process. The convergence to minimum and unwanted bundling points, for example, limit the use of the technic on noisy and complex scenarios as those found in oil reservoirs. This work presents a novel approach to the limitations of deformable contours. By the definition of a second problem of minimization are defined optimal distances of the points of deformable contour according to a optimality criterion that incorporates features of the contour sought. The results show that the proposed technique peovides a larger framework between the contour sought and identified, defines a permanent solution to the problems of grouping and unwanted scattering of points, increases the effectiveness of deformable contours in concave regions and, in addition, defines methodology only able to provide the contours deformable sensitivity about the peculiarities of the contour.

Segundo a definição original de MCINERNEY & TERZOPOULOS (1995), modelos deformáveis são curvas ou superfícies formadas por pontos conectados que simulam corpos elásticos. Por superarem muitas limitações associadas ao procedimento manual e às técnicas tradicionais de processamento, os contornos deformáveis têm se popularizado. Ainda que o uso dos contornos deformáveis seja vasto e crescente, aspectos relevantes da teoria ainda têm demandado atenção. Muitas referências têm sido feitas às limitações da técnica impostas sobretudo pelo seu processo evolutivo. A convergência a mínimos locais e o agrupamento indesejado de pontos, por exemplo, limitam o emprego da técnica em cenários ruidosos e complexos como os encontrados em reservatórios de petróleo. Esse trabalho apresenta uma abordagem inédita às limitações dos contornos deformáveis. Pela definição de um segundo problema de minimização são definidas distâncias ótimas dos pontos do contorno deformável segundo critério de optimalidade que contempla as particularidades do contorno buscado. Os resultados demonstram que a técnica proposta é provedora de maior enquadramento entre o contorno buscado e o identificado, define solução definitiva aos problemas do agrupamento e espalhamento indesejados de pontos, aumenta a efetividade dos contornos deformáveis em regiões côncavas e, em acréscimo, define metodologia unicamente capaz de dotar os contornos deformáveis de sensibilidade quanto às particularidades de contorno. / According to the original definition of Terzopoulos, deformable models are curves or surfaces formed by connected points that simulate elastic bodies. By overcoming many limitations associated with the manual procedure and the traditional techniques of processing, deformable contours have become popular. Although the use of deformable contours is vast and growing aspects of the theory still demand attention. Many references have been made to the limitations of the technique imposed by the process evolution process. The convergence to minimum and unwanted bundling points, for example, limit the use of the technic on noisy and complex scenarios as those found in oil reservoirs. This work presents a novel approach to the limitations of deformable contours. By the definition of a second problem of minimization are defined optimal distances of the points of deformable contour according to a optimality criterion that incorporates features of the contour sought. The results show that the proposed technique peovides a larger framework between the contour sought and identified, defines a permanent solution to the problems of grouping and unwanted scattering of points, increases the effectiveness of deformable contours in concave regions and, in addition, defines methodology only able to provide the contours deformable sensitivity about the peculiarities of the contour.

In bats, the directivity patterns for reception are shaped by the surface geometry of the pinnae. Since many bat species are capable of large ear deformations, these beampatterns can be time-variant. To investigate this time-variance using numerical methods, a digital model that is capable of representing the pinna geometry during the entire deformation cycle has been developed. Due to large deformations and occlusions, some of the surfaces relevant to sound diffraction may not be visible and the geometry of the entire pinna has to be computed from limited data. This has been achieved by combining a complete digital model of the pinna in one position with time-variant sparse sets of three dimensional landmark data. The landmark positions were estimated using stereo vision methods. A finite element model based on elasticity was constructed from CT scans of the pinna post mortem. This elastic model was deformed to provide a good fit to the positions of the landmarks and retain values of smoothness and surface energy comparable to life. This model was able to handle ratios of data to degrees of freedom around 1:5000 and still effect life-like deformations with an acceptable goodness of fit. / Master of Science

Medical imaging plays an increasingly important role in modern healthcare. In medical imaging, it is often relevant to relate different images to each other, something which can prove challenging, since there rarely exists a pre-defined mapping between the pixels in different images. Hence, there is a need to find such a mapping/transformation, a procedure known as image registration. Over the years, image registration has been proved useful in a number of clinical situations. Despite this, current use of image registration in clinical practice is rather limited, typically only used for image fusion. The limited use is, to a large extent, caused by excessive computation times, lack of established validation methods/metrics and a general skepticism toward the trustworthiness of the estimated transformations in deformable image registration. This thesis aims to overcome some of the issues limiting the use of image registration, by proposing a set of technical contributions and two clinical applications targeted at improved clinical efficiency. The contributions are made in the context of a generic framework for non-parametric image registration and using an image registration method known as the Morphon.  In image registration, regularization of the estimated transformation forms an integral part in controlling the registration process, and in this thesis, two regularizers are proposed and their applicability demonstrated. Although the regularizers are similar in that they rely on local structure analysis, they differ in regard to implementation, where one is implemented as applying a set of filter kernels, and where the other is implemented as solving a global optimization problem. Furthermore, it is proposed to use a set of quadrature filters with parallel scales when estimating the phase-difference, driving the registration. A proposal that brings both accuracy and robustness to the registration process, as shown on a set of challenging image sequences. Computational complexity, in general, is addressed by porting the employed Morphon algorithm to the GPU, by which a performance improvement of 38-44x is achieved, when compared to a single-threaded CPU implementation. The suggested clinical applications are based upon the concept paint on priors, which was formulated in conjunction with the initial presentation of the Morphon, and which denotes the notion of assigning a model a set of properties (local operators), guiding the registration process. In this thesis, this is taken one step further, in which properties of a model are assigned to the patient data after completed registration. Based upon this, an application using the concept of anatomical transfer functions is presented, in which different organs can be visualized with separate transfer functions. This has been implemented for both 2D slice visualization and 3D volume rendering. A second application is proposed, in which landmarks, relevant for determining various measures describing the anatomy, are transferred to the patient data. In particular, this is applied to idiopathic scoliosis and used to obtain various measures relevant for assessing spinal deformity. In addition, a data analysis scheme is proposed, useful for quantifying the linear dependence between the different measures used to describe spinal deformities.

Medical imaging modalities provide effective information for anatomic or metabolic activity of tissues and organs in the body. Therefore, medical imaging technology is a critical component in diagnosis and treatment of various illnesses. Medical image segmentation plays an important role in converting medical images into anatomically, functionally or surgically identifiable structures, and is used in various applications. In this study, some of the major medical image segmentation methods are examined and applied to 2D CT images of upper torso for segmentation of heart, lungs, bones, and muscle and fat tissues. The implemented medical image segmentation methods are thresholding, region growing, watershed transformation, deformable models and a hybrid method / watershed transformation and region merging. Moreover, a comparative analysis is performed among these methods to obtain the most efficient segmentation method for each tissue and organ in torso. Some improvements are proposed for increasing accuracy of some image segmentation methods.

In brachytherapy, radiation therapy is performed by placing the radiation source into or very close to the tumour. When calculating the absorbed dose, water is often used as the radiation transport and dose scoring medium for soft tissues and this leads to inaccuracies. The iterative reconstruction algorithm DIRA is under development at the Center for Medical Imaging Science and Visualization, Linköping University. DIRA uses dual-energy CT to decompose tissues into different doublets and triplets of base components for a better absorbed dose estimation. To accurately determine mass fractions of these base components for different tissues, the tissues needs to be identified in the image. The aims of this master thesis are: (i) Find an automated segmentation algorithm in CT that best segments the male pelvis. (ii) Implement a segmentation algorithm that can be used in DIRA. (iii) Implement a fully automatic segmentation algorithm. Seven segmentation methods were tested in Matlab using images obtained from Linköping University Hospital. The methods were: active contours, atlas based registration, graph cuts, level set, region growing, thresholding and watershed. Four segmentation algorithms were selected for further analysis: phase based atlas registration, region growing, thresholding and active contours without edges. The four algorithms were combined and supplemented with other image analysis methods to form a fully automated segmentation algorithm that was implemented in DIRA. The newly developed algorithm (named MK2014) was sufficiently stable for pelvic image segmentation with a mean computational time of 45.3 s and a mean Dice similarity coefficient of 0.925 per 512×512 image. The performance of MK2014 tested on a simplified anthropomorphic phantom in DIRA gave promising result. Additional tests with more realistic phantoms are needed to confirm the general applicability of MK2014 in DIRA.

Η παρούσα διδακτορική διατριβή αποτελεί μια μελέτη σύγκρισης και βελτιστοποίησης διάφορων αλγορίθμων ανακατασκευής δεδομένων, τα οποία προέρχονται από ένα πρότυπο σύστημα ΡΕΤ ικανό για απεικονίσεις μικρών ζώων. Η σύγκριση αφορά υπάρχοντες αλγορίθμους ανακατασκευής ενώ παρουσιάζεται και ένας νέος επαναληπτικός αλγόριθμος o ISWLS. Το δεύτερο μέρος της παρούσας διδακτορικής διατριβής ασχολείται με ένα άλλο σημαντικό στάδιο της επεξεργασίας ιατρικών δεδομένων την τμηματοποίηση της ιατρικής εικόνας. Παρουσιάζονται διάφορες τεχνικές παραμετρικών ελαστικών μοντέλων. Συγκεκριμένα παρουσιάζονται το κλασσικό μοντέλο φιδιού (snake), το μοντέλο gradient vector flow (gvf-snake) και τα t-snakes (topology-adaptive snakes). Επίσης παρουσιάζεται η μέθοδος self-affine mapping σαν μια εναλλακτική των παραπάνω παραμετρικών ελαστικών μοντέλων και εισάγεται ένα νέο κριτήριο σύγκλισής της. Όλες οι τεχνικές εφαρμόζονται σε οφθαλμικές εικόνες με σκοπό την τμηματοποίηση του οπτικού δίσκου / Small animal imaging is the conjunctive ring between experimental research and clinical implementation. Positron Emission Tomography (PET) has proven a valuable tool for in vivo small animal functional imaging. Image reconstruction in PET uses the collected projection data of the object/patient under examination. The purpose of this study is to assess the performance of iterative reconstruction methods, using phantom data from a prototype small-animal PET system. The algorithms being compared are the simultaneous versions of ART (SART), EM-ML, ISRA and WLS and a new iterative algorithm being introduced under the short name ISWLS. In the second part of this thesis elastic or deformable models are studied. Various methods of parametric elastic models are presented, namely the classical snake, the gradient vector field snake (gvf-snake) and the topogy-adaptive snake (t-snake). Also presented the method of self-affine mapping system as an alternative of elastic models. Further a new comparison criterion for the self affine mapping system method is introduced. All methods are applied to retinal images with the purpose of segmenting the optical disk. Moreover the aforementioned methods are compared in terms of segmentation accuracy.

Interest in hippocampus increased sharply after his significance in the process of learning and retention of information was published. In particular, considerable interest was in its volume changes and their effect on Alzheimer’s disease. Understanding the structure and function hippocampus would contribute to a more accurate diagnosis of this disease. In this work was created a method of hippocampal segmentation using active contours. With its help, the data composed of both healthy and a diseased patients was segmented and the results were then statistically analyzed using statistical methods such as Kruskal-Walis test, Mann-Whitney test. The level of significance given by results of analysis supports alternative hypothesis that attaches significance of the difference in volume of the hippocampus between studied groups.

The need for efficient interference detection frequently arises in computer graphics, robotics, virtual prototyping, surgery simulation, computer games, and visualization. To prevent bodies passing directly through each other, the simulation system must be able to track touching or intersecting geometric primitives. In interactive simulations, in which millions of geometric primitives may be involved, highly efficient collision detection algorithms are necessary. For these reasons, new adaptive collision detection algorithms for rigid and different types of deformable polygon meshes are proposed in this thesis. The solutions are based on adaptive bounding volume hierarchies. For deformable body simulation, different refit and reconstruction schemes to efficiently update the hierarchies as the models deform are presented. These methods permit the models to change their entire shape at every time step of the simulation. The types of deformable models considered are (i) polygon meshes that are deformed by arbitrary vertex repositioning, but with the mesh topology preserved, (ii) models deformed by linear morphing of a fixed number of reference meshes, and (iii) models undergoing completely unstructured relative motion among the geometric primitives. For rigid body simulation, a novel type of bounding volume, the slab cut ball, is introduced, which improves the culling efficiency of the data structure significantly at a low storage cost. Furthermore, a solution for even tighter fitting heterogeneous hierarchies is outlined, including novel intersection tests between spheres and boxes as well as ellipsoids and boxes. The results from the practical experiments indicate that significant speedups can be achieved by using these new methods for collision queries as well as for ray shooting in complex deforming scenes.