Single ventricle congenital heart defects afflict 2 per every 1000 births. They are characterized by cyanotic mixing between the de-oxygenated blood coming back from the systemic circulation and the oxygenated blood from the pulmonary circulation. Prior to introduction of the Fontan procedure in 1971, surgical options for single ventricle patients were limited. The Fontan operation involves a series of three palliative procedures aimed at the separation of systemic and pulmonary circulations and reducing the long term effects of chronic hypoxia and ventricular volume overload. The total cavopulmonary connection (TCPC) is completed in the final stage of the surgery with the anastomosis of the inferior vena cava (IVC) and superior vena cava to the pulmonary arteries. Improved quantification and visualization of flow structures within the TCPC has the potential to aid in the planning and design of the Fontan operation. Despite significant development of phase contrast magnetic resonance imaging (PC MRI) for in vivo flow measurements, it is not routinely applied in children with single ventricle congenital heart disease. Limited technologies available for post-processing of PC MRI data has prevented clinicians and scientists from conducting the detailed hemodynamic analyses necessary to better understand the physiology of the single ventricle circulation. This thesis attempts to bridge the gap between PC MRI and fluid dynamics, by developing the necessary post-processing technologies for PC MRI, and then applying these techniques for characterizing single ventricle hemodynamics.

Magnetic resonance imaging (MRI) is the preferred imaging modality for visualization of intracranial soft tissues. Surgical planning, and increasingly surgical navigation, use high resolution 3-D patient-specific structural maps of the brain. However, the process of MRI is a multi-parameter tomographic technique where high resolution imagery competes against high contrast and reasonable acquisition times. Resolution enhancement techniques based on super-resolution are particularly well suited in solving the problems of resolution when high contrast with reasonable times for MRI acquisitions are needed. Super-resolution is the concept of reconstructing a high resolution image from a set of low-resolution images taken at dierent viewpoints or foci. The MRI encoding techniques that produce high resolution imagery are often sub-optimal for the desired contrast needed for visualization of some structures in the brain. A novel super-resolution reconstruction framework for MRI is proposed in this thesis. Its purpose is to produce images of both high resolution and high contrast desirable for image-guided minimally invasive brain surgery. The input data are multiple 2-D multi-slice Inversion Recovery MRI scans acquired at orientations with regular angular spacing rotated around a common axis. Inspired by the computed tomography domain, the reconstruction is a 3-D volume of isotropic high resolution, where the inversion process resembles a projection reconstruction problem. Iterative algorithms for reconstruction are based on the projection onto convex sets formalism. Results demonstrate resolution enhancement in simulated phantom studies, and in ex- and in-vivo human brain scans, carried out on clinical scanners. In addition, a novel motion correction method is applied to volume registration using an iterative technique in which super-resolution reconstruction is estimated in a given iteration following motion correction in the preceding iteration. A comparison study of our method with previously published methods in super-resolution shows favorable characteristics of the proposed approach.

The cell infiltration into the myocardial infarction (MI) site was studied using magnetic resonance imaging (MRI) with micrometer-sized iron oxide particles (MPIO) as cell labeling probes. MI is a leading cause of global death and disability. However, the roles of inflammatory cells and stem cells during the post-MI remodeling and repair processes are yet to be discovered. This study was to develop noninvasive MRI techniques to monitor and quantify the cellular infiltration into the MI site. MPIO can produce pronounced signal attenuation at regions of interest in MRI. Therefore, cells labeled with these particles can be detected after they are activated and home to the MI site. In the first project, MPIO of various doses were injected into the mouse blood stream 7 days before the MI surgery. Serial MRI was performed at various time points post-MI to monitor the inflammatory cell infiltration into the MI site. Significant signal attenuation caused by labeled cells, in particular macrophages, was observed at the MI site. The study suggests an optimal imaging window should be from 7 to 14 days post-MI, during which the MR signal was inversely proportional to the MPIO dose. The study also suggests an optimal MPIO dose should be between 9.1 and 14.5 µg Fe/g body weight. In the second project, mesenchymal stem cells labeled with MPIO were transplanted into the mouse bone marrow 14 days before the MI surgery. Serial MRI was performed at various time points post-MI to monitor the labeled cells, which mobilized from the bone marrow and homed to the MI site. All the MRI findings were further confirmed by histology. In addition to revealing the characteristics of cell infiltration during MI, this study also provides noninvasive MRI techniques to monitor and potentially quantify labeled cells at the pathological site. The technique can also be used to investigate the function of cells engaged in MI and to test the effect on cell infiltration caused by any treatment strategies.

Cardiac Resynchronization Therapy (CRT) is a treatment option for heart failure patients with ventricular dyssynchrony. CRT corrects for dyssynchrony by electrically stimulating the septal and lateral walls of the left ventricle (LV), forcing synchronous con- traction and improving cardiac output. Current selection criteria for CRT rely upon the QRS duration, measured from a surface electrocardiogram, as a marker of electrical dyssynchrony. Unfortunately, 30-40% of patients undergoing CRT fail to benefit from the treatment. A multitude of studies have shown that presence of mechanical dyssynchrony in the LV is an important factor in determining if a patient will benefit from CRT. Furthermore, recent evidence suggests that patient response can be improved by placing the LV pacing lead in the most dyssynchronous or latest contracting segment. The overall goal of this project was to develop methods that allow for accurate assessment and display of regional mechanical dyssynchrony throughout the LV and at the site of the LV pacing lead. To accomplish this goal, we developed a method for quantifying regional dyssynchrony from standard short-axis cine magnetic resonance (MR) images. To assess the effects of LV lead placement, we developed a registration method that allows us to project the LV lead location from dual-plane fluoroscopy onto MR measurements of cardiac function. By applying these techniques in patients undergoing CRT, we were able to investigate the relationship between regional dyssynchrony, LV pacing lead location, and CRT response.

An ever-increasing number of functional magnetic resonance imaging (fMRI) studies are now using information-based multi-voxel pattern analysis (MVPA) techniques to decode mental states. In doing so, they achieve a significantly greater sensitivity compared to when they use univariate analysis frameworks. Two most prominent MVPA methods for information mapping are searchlight decoding and classifier weight mapping. The new MVPA brain mapping methods, however, have also posed new challenges for analysis and statistical inference on the group level. In this thesis, I discuss why the usual procedure of performing t-tests on MVPA derived information maps across subjects in order to produce a group statistic is inappropriate. I propose a fully nonparametric solution to this problem, which achieves higher sensitivity than the most commonly used t-based procedure. The proposed method is based on resampling methods and preserves the spatial dependencies in the MVPA-derived information maps. This enables to incorporate a cluster size control for the multiple testing problem. Using a volumetric searchlight decoding procedure and classifier weight maps, I demonstrate the validity and sensitivity of the new approach using both simulated and real fMRI data sets. In comparison to the standard t-test procedure implemented in SPM8, the new results showed a higher sensitivity and spatial specificity. The second goal of this thesis is the comparison of the two widely used information mapping approaches -- the searchlight technique and classifier weight mapping. Both methods take into account the spatially distributed patterns of activation in order to predict stimulus conditions, however the searchlight method solely operates on the local scale. The searchlight decoding technique has furthermore been found to be prone to spatial inaccuracies. For instance, the spatial extent of informative areas is generally exaggerated, and their spatial configuration is distorted. In this thesis, I compare searchlight decoding with linear classifier weight mapping, both using the formerly proposed non-parametric statistical framework using a simulation and ultra-high-field 7T experimental data. It was found that the searchlight method led to spatial inaccuracies that are especially noticeable in high-resolution fMRI data. In contrast, the weight mapping method was more spatially precise, revealing both informative anatomical structures as well as the direction by which voxels contribute to the classification. By maximizing the spatial accuracy of ultra-high-field fMRI results, such global multivariate methods provide a substantial improvement for characterizing structure-function relationships.

Functional magnetic resonance imaging (fMRI) has emerged as the method of choice to non-invasively investigate brain function in humans. Though brain is known to act as a nonlinear system, here has not been much effort to explore the applicability of nonlinear analysis techniques to fMRI data. Also, recent trends have suggested that functional localization as a model of brain function is incomplete and efforts are being made to develop models based on networks of regions to understand brain function. Therefore this thesis attempts to introduce the twin concepts of nonlinear dynamics and network analysis into a broad spectrum of fMRI data analysis techniques. First, we characterized the nonlinear univariate dynamics of fMRI noise using the concept of embedding to explain the origin of tissue-specific differences of baseline activity in the brain. The embedding concept was extended to the multivariate case to study nonlinear functional connectivity in the distributed motor network during resting state and continuous motor task. The results showed that the nonlinear method may be more sensitive to the desired gray matter signal. Subsequently, the scope of connectivity was extended to include directional interactions using Granger causality. An integrated approach was developed to alleviate the confounding effect of the spatial variability of the hemodynamic response and graph theory was employed to characterize the network topology. This methodology proved effective in characterizing the dynamics of cortical networks during motor fatigue. The nonlinear extension of Granger causality showed that it was more robust in the presence of confounds such as baseline drifts. Finally, we utilized the integration of the spatial correlation function to study connectivity in local brain networks. We showed that our method is robust and can reveal interesting information including the default mode network during resting state. Application of this technique to anesthesia data showed dose dependent suppression of local connectivity in the default mode network, particularly in the frontal areas. Given the body of evidence emerging from our studies, nonlinear and network characterization of fMRI data seems to provide novel insights into brain function.

The extension from Quantitative Magnetic Resonance Imaging to synthetic imaging has the clear advantage of being able to continually image the patient after the exam. MR techniques such as DWI are commonly used but have some clear disadvantages resulting from the use of echoplanar imaging. It should then be asked whether one imaging technique is objectively better. If one technique is better, the incorporation in clinical settings could produce better diagnostic rates, and save valuable time. In order to quantitatively assess the quality of these techniques, the SNR and CNR values of similar tissues were compared. The pre-analysis discussion concentrating on the spatial resolution and artifacts, supports that synthetic images have an advantage over DWI due to higher resolution and absence of artifacts. The SNR and CNR values were calculated for each patient and image type for the comparison, initially assuming that the synthetic images would have a higher mean SNR and CNR. In most cases the differences between scan types was found to not be statistically significant. In conclusion, this analysis could not support the initial theory that the synthetic images had a higher SNR or CNR. The research shows that they are more likely to be comparable. An investigation of the diagnostic power of the synthetic in comparison to standard DWI would give clinical relevance to these results.

O Transtorno de Humor Bipolar (THB) acomete até 3% dos adultos e os filhos desses pacientes constituem uma população em risco para o desenvolvimento de transtornos psiquiátricos. No entanto, faltam marcadores que permitam a identificação precoce dos indivíduos que apresentam maior vulnerabilidade para o desenvolvimento de psicopatologia. Estudos preliminares com Ressonância Magnética (RM) indicaram que alterações em substância branca estariam presentes não apenas em pacientes em episódio de alteração de humor, mas também em pacientes eutímicos e em seus familiares saudáveis, sugerindo que tais alterações poderiam constituir um endofenótipo potencial deste transtorno. A Imagem por Tensor de Difusão (Diffusion Tensor Magnetic Resonance Imaging - DT-MRI) é uma aquisição de RM que permite análise mais completa e detalhada das características da substância branca cerebral que a RM tradicional. A investigação de alterações na estrutura cerebral, particularmente de substância branca, de jovens portadores de THB não medicados e familiares saudáveis criteriosamente avaliados pode ajudar a elucidar a neurobiologia subjacente ao THB e, conseqüentemente, a identificar marcadores de vulnerabilidade ao transtorno. Objetivo: Avaliar se havia alterações em substância branca em crianças e adolescentes com THB e crianças e adolescentes filhos saudáveis de portadores de transtorno do humor bipolar quando comparados a controles saudáveis, utilizando a técnica de neuroimagem de DTMRI. Nossas hipóteses eram que jovens com THB e filhos de pacientes com THB, quando comparados a controles saudáveis, apresentariam (i) diminuição da fração de anisotropia (FA) e (ii) essas alterações seriam mais pronunciadas em crianças acometidas por THB do que em crianças saudáveis filhas de pacientes com THB. Métodos: Obtivemos imagens de DT-MRI de boa qualidade de 16 crianças e adolescentes com THB (média de idade ± D.P.= 12,7 ± 2,5 anos), 15 filhos saudáveis de pacientes com THB tipo I (média de idade ± D.P.= 13,5 ± 2,7 anos) e 15 controles saudáveis (média de idade ± D.P.= 13,5 ± 2,5 anos). Os diagnósticos foram formulados de acordo com os critérios do DSM-IV, usando as entrevistas Kiddie-SADS-PL (crianças) e Structured Clinical Interview for DSM-IV (adultos). A RM foi realizada em um scanner Philips 3,0 Tesla, com os seguintes parâmetros de aquisição: TR = 6106,0 ms, TE = 65,0ms, FOV = 224x224mm, espessura de corte = 2.0mm, sem gap, matriz de aquisição = 112x112 pixels e 3 médias, resultando em tamanho de voxel isotrópico = 2,0x2,0x2,0mm. As imagens de DTI foram pré-processadas com programas oriundos do FMRIB\'s software library (FSL), de acordo com o pipeline sugerido para o processamento de substância branca com Tract-Based Spatial Statistics (TBSS) e análise estatística com o programa Randomise (ambos integrantes do FSL). Resultados: Os grupos não diferiram em idade, gênero, grau de puberdade ou QI. Valores de FA de pacientes pediátricos com THB foram significativamente menores em relação aos de controles saudáveis (p < 0,05, corrigido para múltiplas comparações) em um cluster de 695 voxels no hemisfério direito que abrange a porção superiora da corona radiata e o corpo do corpo. Não houve diferença significativa entre pacientes com THB e filhos saudáveis de pacientes com THB, ou com filhos saudáveis de pacientes com THB e controles saudáveis. Discussão: Nossos dados corroboram a literatura de diminuição de FA em crianças e adolescentes com THB e avançam em mostrar esta alteração em pacientes não medicados. Mas nossos resultados não apoiam a hipótese de alterações em substância branca como endofenótipo de THB. Estudos de seguimento com amostras maiores e rigorosamente caracterizadas são necessários para se elucidar o papel das alterações em substância branca no THB. / Up to 3% of adults are affected with Bipolar Disorder (BD) and the offspring of these patients constitute a population at risk for the development of psychiatric disorders. Nevertheless, there are still no vulnerability markers to allow the early identification of those who are at greater risk of developing psychopathology among this population. Preliminary data indicate that white matter abnormalities may precede the disease onset and be present even in unaffected relatives - suggesting they could be further explored as an endophenotype for BD. Diffusion Tensor Magnetic Resonance Imaging (DTMRI) is an MRI acquisition that allows a thorough and detailed analysis of brain white matter characteristics. The investigation of white matter alterations in young, nonmedicated, BD patients and healthy relatives may help us understand the underlying neurobiology of BD. Objectives: evaluate white matter alterations in children at-risk and with BD using DT-MRI. Our hypothesis were that BD offspring, compared to healthy controls, would exhibit (i) reduced fractional anisotropy (FA) and (ii) these alterations would be more pronounced in children with BD than in those at-risk. Methods: We successfully scanned 16 BD patients (mean age ± S.D.= 12,7 ± 2,5 years) 15 healthy offspring with at least one parent with BD I diagnosis (mean age ± S.D.= 13,5 ± 2,7 years) and 15 healthy controls (mean age ± S.D.= 13,5 ± 2,5 years) with no history of psychiatric disorder in first-degree relatives. Psychiatric diagnosis were established according to the DSM-IV criteria, using the Kiddie-SADS-PL interview (children) and Structured Clinical Interview for DSM-IV (adults). The MRI was conducted at a 3.0 Tesla Philips scanner. Acquisition: the parameters were TR = 6106,0ms, TE = 65,0ms, FOV= 224x224mm, slice thickness = 2.0mm/no gap, matrix acquisition = 112x112 pixels, 3 averages, resulting in an isotropic voxel = 2,0x2,0x2,0 mm. DT-MRI images were preprocessed according to do FMRIB\'s software library\'s (FSL) pipeline for Tract-Based Spatial Statistics\' (TBSS) and Randomise analyses. Results: Groups did not differ in age, gender or pubertal status. Voxelwise analyses showed significant differences in FA values between BD patients and healthy controls (p < 0,05, FDR corrected for multiple comparisons) in a 695 voxels cluster comprising right corona radiata and corpus callosum . Discussion: This study was the first to evaluate a sample of non-medicated BD children and adolescents with DT-MRI and it corroborates extant literature data of lower FA in BD children compared to healthy controls. Nevertheless, our data do not support white matter alterations as an endophenotype for BD. More studies, with larger and well characterized samples are necessary to advance our understanding of the role of white matter alteration in BD.

L’Imagerie par Résonance Magnétique (IRM) est une belle application de la physique et constitue sans aucun doute l’une des techniques les plus performantes d’imagerie médicale. Basée sur le phénomène de la Résonance Magnétique Nucléaire (RMN) du proton contenu dans les molécules d’eau, l’IRM permet d’investiguer en coupes les tissus mous du corps, sur base de contrastes différents. La méthode est non-invasive et n’utilise pas de radiations ionisantes. En plus des données morphologiques, l’IRM permet également d’obtenir des informations fonctionnelles et physiologiques.<p>De nos jours, plus de 10 000 unités IRM existent dans le monde et des millions d’examens sont réalisés chaque année. La technique est en constant développement et le domaine de recherches est multidisciplinaire. Il concerne aussi bien les développements méthodologiques (imagerie rapide, imagerie de diffusion, etc.) que technologiques (imagerie à haut champ, systèmes de gradients à commutation rapide, etc.), le point central des recherches étant l’amélioration de la qualité des images et la diminution du temps d’acquisition. Ceci implique l'optimisation des différentes séquences IRM (séries d'impulsions radiofréquence et de gradients de champ magnétique) tenant compte des contraintes imposées par le matériel, ainsi que le développement et l'optimisation du matériel lui-même. Cette thèse est consacrée au design avancé des séquences d’impulsions et contribue donc à l'optimisation des schémas d’acquisition en IRM.<p>En particulier, le présent travail est focalisé sur la compréhension et l’amélioration d’un certain type de séquences rapides, employant des échos de gradients :les séquences Steady-State Free Precession (SSFP) et plus précisément les séquences dites balanced-SSFP. Dans ce genre de schéma d’acquisition, le système est excité rapidement et périodiquement, conduisant à l’établissement d’un état stationnaire de l’aimantation. <p>La première partie de la thèse est consacrée à une analyse approfondie des propriétés du signal dans une séquence balanced-SSFP, à la fois à l’état stationnaire et à l’état transitoire. Ensuite, de nouveaux schémas d’acquisition sont développés sur base de calculs analytiques et de simulations numériques et sont ensuite testés expérimentalement. D’une part, une manipulation de l’état stationnaire est présentée en vue de supprimer le signal de la graisse sur les images (qui peut être gênant pour le diagnostic de certaines lésions ou maladies). D’autre part, l’application d’une phase de préparation en vue d’obtenir un contraste basé sur le degré de diffusion des molécules d’eau dans les tissus est analysée en détails, afin d’améliorer la qualité d’image produite par des séquences de diffusion existantes.<p>La présente thèse constitue donc un travail de recherches théoriques et expérimentales, allant de la conception de nouveaux schémas d’acquisition à leur expérimentation sur volontaires, en passant par leur implémentation sur un imageur IRM. Ce travail a été réalisé au sein de l’Unité d’IRM – Radiologie de l’Hôpital Erasme, sous la direction de Thierry Metens, Docteur en Sciences et Physicien IRM.<p> / Doctorat en Sciences / info:eu-repo/semantics/nonPublished

No description available.