No description available.

No description available.

No description available.

In certain neuro-degenerative diseases likemultiple sclerosis (MS), the rate of brain atrophy can be measured by monitoring the brain parenchymal fraction (BPF) in such patients. The BPF is defined as the ratio of brain parenchymal volume (BPV, defined as the total volume of gray matter tissue, white matter tissue and other unidentified tissue) and intracranial volume (ICV, the total volume of the skull). It can be represented by the formula in equation 1: <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?%5Csmall%20BPF%20=%20%5Cfrac%20%7BPBV%7D%7BICV%7D%20%5C;%20%5C;%20%5C;%20%5C;%20(1)" /> A complication with this measure is that the BPF is affected by the presence of edema in the brain, which leads to swelling and hence may obscure the true rate of brain atrophy. This leads to uncertainty when establishing “normal values” of BPF when analyzing different magnetic resonance imaging (MRI) scans of the same patient. Another problem is that different MRI scans of the same patient cannot be compared directly, due to the fact that the head of the patient will be in a different position for every scan. The SyMRI software used in this master thesis has the functionality to perform brain tissue characterization and measurement of brain volume, given a number of MR images of a patient. Using tissue properties such as longitudinal relaxation time (T1), transverse relaxation time (T2) and proton density (PD), each voxel in a volume can be classified to belong to a certain tissue type. From these measurements, the intracranial volume, brain volume, white matter, gray matter and cerebrospinal fluid volumes can easily be estimated. In this master thesis, the BPF of several patients were analyzed based on quantitative MRI (qMRI) images, in order to identify the change of BPF due to the presence of edema over time. Volumes obtained from the same patients at different time points were aligned (registered), such that the BPF can be easily compared between years. A correlation analysis between the BPF and R1, R2 and PD was performed (R1 is the longitudinal relaxation rate defined as 1/T1 relaxation time and R2 Is transverse relaxation rate defined as 1/T2 relaxation time) to investigate if any of these variables can explain the change in BPF. The results show that due to image registration, and removing some of the slices from the top and bottom of the head, the BPF of the patients was corrected to a certain extent. The change in the mean BPF of each patient over four years was less than 1% post registration and slice removal. However, the decrease in standard deviation was between 6.9% to 52% after registration and removing of slices. The BPF of the follow-up years also came closer to the initial BPF value measured in the first year. The statistical analysis of the BPF and R1, R2 and PD, showed a very low correlation (0.1) between BPF and PD, and intermediate correlations between BPF and R1, R2 (0.385 and -0.51, respectively). Future work will focus on understanding how these results relate to edema.

In this research, multi-perspective image registration using LiDAR and visual images was considered. 2D-3D image registration is a difficult task because it requires the extraction of different semantic features from each modality. This problem is solved in three parts. The first step involves detection and extraction of common features from each of the data sets. The second step consists of associating the common features between two different modalities. Traditional methods use lines or orthogonal corners as common features. The third step consists of building the projection matrix. Many existing methods use global positing system (GPS) or inertial navigation system (INS) for an initial estimate of the camera pose. However, the approach discussed herein does not use GPS, INS, or any such devices for initial estimate; hence the model can be used in places like the lunar surface or Mars where GPS or INS are not available. A variation of the method is also described, which does not require strong features from both images but rather uses intensity gradients in the image. This can be useful when one image does not have strong features (such as lines) or there are too many extraneous features.

One of the fundamental problems in computer vision is point set registration. Point set registration finds use in many important applications and in particular can be considered one of the crucial stages involved in the reconstruction of models of physical objects and environments from depth sensor data. The problem of globally aligning multiple point sets, representing spatial shape measurements from varying sensor viewpoints, into a common frame of reference is a complex task that is imperative due to the large number of critical functions that accurate and reliable model reconstructions contribute to. In this thesis we focus on improving the quality and feasibility of model and environment reconstruction through the enhancement of multi-view point set registration techniques. The thesis makes the following contributions: First, we demonstrate that employing kernel density estimation to reason about the unknown generating surfaces that range sensors measure allows us to express measurement variability, uncertainty and also to separate the problems of model design and viewpoint alignment optimisation. Our surface estimates define novel view alignment objective functions that inform the registration process. Our surfaces can be estimated from point clouds in a datadriven fashion. Through experiments on a variety of datasets we demonstrate that we have developed a novel and effective solution to the simultaneous multi-view registration problem. We then focus on constructing a distributed computation framework capable of solving generic high-throughput computational problems. We present a novel task-farming model that we call Semi-Synchronised Task Farming (SSTF), capable of modelling and subsequently solving computationally distributable problems that benefit from both independent and dependent distributed components and a level of communication between process elements. We demonstrate that this framework is a novel schema for parallel computer vision algorithms and evaluate the performance to establish computational gains over serial implementations. We couple this framework with an accurate computation-time prediction model to contribute a novel structure appropriate for addressing expensive real-world algorithms with substantial parallel performance and predictable time savings. Finally, we focus on a timely instance of the multi-view registration problem: modern range sensors provide large numbers of viewpoint samples that result in an abundance of depth data information. The ability to utilise this abundance of depth data in a feasible and principled fashion is of importance to many emerging application areas making use of spatial information. We develop novel methodology for the registration of depth measurements acquired from many viewpoints capturing physical object surfaces. By defining registration and alignment quality metrics based on our density estimation framework we construct an optimisation methodology that implicitly considers all viewpoints simultaneously. We use a non-parametric data-driven approach to consider varying object complexity and guide large view-set spatial transform optimisations. By aligning large numbers of partial, arbitrary-pose views we evaluate this strategy quantitatively on large view-set range sensor data where we find that we can improve registration accuracy over existing methods and contribute increased registration robustness to the magnitude of coarse seed alignment. This allows large-scale registration on problem instances exhibiting varying object complexity with the added advantage of massive parallel efficiency.

Xiong, Wei. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2007. / Includes bibliographical references (leaves 56-61). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgement --- p.iv / Chapter 1 --- Introduction --- p.1 / Chapter 2 --- Background Study --- p.6 / Chapter 2.1 --- Stereo matching --- p.6 / Chapter 2.2 --- Digital image matting --- p.8 / Chapter 2.3 --- Expectation Maximization --- p.9 / Chapter 3 --- Model Definition --- p.12 / Chapter 4 --- Initialization --- p.20 / Chapter 4.1 --- Initializing disparity --- p.20 / Chapter 4.2 --- Initializing alpha matte --- p.24 / Chapter 5 --- Optimization --- p.26 / Chapter 5.1 --- Expectation Step --- p.27 / Chapter 5.1.1 --- "Computing E((Pp(df = d1̐ưجθ(n),U))" --- p.28 / Chapter 5.1.2 --- "Computing E((Pp(db = d2̐ưجθ(n),U))" --- p.29 / Chapter 5.2 --- Maximization Step --- p.31 / Chapter 5.2.1 --- "Optimize α, given {F, B} fixed" --- p.34 / Chapter 5.2.2 --- "Optimize {F, B}, given α fixed" --- p.37 / Chapter 5.3 --- Computing Final Disparities --- p.40 / Chapter 6 --- Experiment Results --- p.42 / Chapter 7 --- Conclusion --- p.54 / Bibliography --- p.56

The advancements in medical technologies have allowed more accurate diagnosis and quantitative assessments. Magnetic Resonance Imaging is one of the most effective and critical technologies in modern diagnosis. However, preprocessing tasks are required to perform various research topics basing on MR image. Registration is one of the those preprocessing tasks. In this research, a semi-automatic utility was developed for doing MRI registration of small animals. It focuses on 2D rigid body registration. The test results show that this developed utility can perform registration well for MRI of small animals in both intra-subject and inter-subjects.

Thesis, Dijon, 1911 / Bibliography

The main goal of the thesis was the development of a registration method able to match images containing both rigid and non-rigid structures for neck images acquired during the process of radiotherapy. We were interested in developing a volumetric registration method for multimodal images. Specifically we developped an articulated transformation we embedded in the current registration framework. We used physics-based material equations of elasticity to propagate the resulting deformations. This registration method has been used to register articulated structures and was applied to vertebral columns. We have also developed non-rigid registration methods with Mathieu De Craene. The first method is a volumetric and multimodal non-rigid algorithm taking into account the elasticity of the different bodies in the images to register. We used the elastic theory to regularize a deformation maximizing a mutual information criterion. A second method, equivalent of the optical flow but developed for multi-modal images is also described. This method has fewer parameters to adjust but does not take into account the stiffness of the tissues to match. These non-rigid methods have been applied to prostate images, histological sections and to finalize the registration initialized with the articulated registration in the neck area. To assign physical properties, we had to create meshes with good quality elements. We have implemented mesh generators to create the meshes needed to represent the objects to be registered.