<p>The pumping performance of the heart is affected by the mechanical properties of the muscle fibre part of the cardiac wall, the myocardium. The myocardium has a complex structure, where muscle fibres have different orientations at different locations, and during the cardiac cycle, the myocardium undergoes large elastic deformations. Hence, myocardial strain pattern is complex. In this thesis work, a computation method for myocardial strain and a detailed map of myocardial transmural strain during the cardiac cycle are found by the use of surgically implanted metallic markers and beads. The strain is characterized in a local cardiac coordinate system. Thereafter, non-invasive phase contrast magnetic resonance imaging (PC-MRI) is used to compare strain at different myocardial regions. The difference in resolution between marker data and PC-MRI data is elucidated and some of the problems associated with the low resolution of PC-MRI are given.</p>

The pumping performance of the heart is affected by the mechanical properties of the muscle fibre part of the cardiac wall, the myocardium. The myocardium has a complex structure, where muscle fibres have different orientations at different locations, and during the cardiac cycle, the myocardium undergoes large elastic deformations. Hence, myocardial strain pattern is complex. In this thesis work, a computation method for myocardial strain and a detailed map of myocardial transmural strain during the cardiac cycle are found by the use of surgically implanted metallic markers and beads. The strain is characterized in a local cardiac coordinate system. Thereafter, non-invasive phase contrast magnetic resonance imaging (PC-MRI) is used to compare strain at different myocardial regions. The difference in resolution between marker data and PC-MRI data is elucidated and some of the problems associated with the low resolution of PC-MRI are given.

Autosomal dominant polycystic disease (ADPKD) is the most common hereditary renal disease and is characterized by renal cyst growth and enlargement. Hypertension occurs early when renal function is normal and is characterized by decreased renal blood flow. Accordingly, the measurement of blood flow in the renal arteries can be a valuable tool in evaluating disease progression. In studies performed in conjunction with this work, blood flow was measured through the renal arteries using magnetic resonance imaging (MRI). In order to validate these in vivo measurements, a vascular phantom was created using polyvinyl alcohol (PVA) and also scanned using MRI under controlled steady flow conditions. Ranges of vessel diameters and flow velocities were used to simulate actual flow in a normal and diseased population of adults and children. With the vessel diameters studied in this experiment, minimization of field of view and an increase in spatial resolution is important in obtaining accurate data. However, a significant difference does not exist between the results when using the 160 or 200 mm FOV. An increase in the number of phase encodings provides improved results, although an increase in image acquisition time is observed. Velocity-encoding in all three orthogonal directions does not improve image data. This method of using MRI to measure flow through a vessel is shown to be both accurate and reproducible, and the protocol providing the most correct results is prescribed. Breath-hold phase-velocity encoded MRI proves to be an accurate and reproducible technique in capturing flow and has the potential to be used for the purpose of observing hemodynamic changes in the renal arteries with the progression of ADPKD.

La vélocimétrie par IRM est une modalité intéressante pour explorer des pathologies cardiovasculaires. La séquence d'IRM en contraste de phase a la particularité de fournir à la fois des informations anatomiques et des informations physiologiques, permettant ainsi de mesurer les propriétés géométriques des vaisseaux ainsi que leurs propriétés hémodynamiques. Le but de cette thèse est d'automatiser la segmentation des vaisseaux et les mesures de vélocimétrie, un traitement manuel étant inadapté à une exploitation de la vélocimétrie IRM à des fins diagnostiques en routine clinique. Les travaux menés ont conduit à proposer une méthode de segmentation basée sur les contours actifs guidés par une information région (approche " région "), contrairement aux approches classiques se focalisant uniquement sur les frontières inter-régions (approche " contour "). Cette approche " région " a été évaluée sur des données provenant d'un fantôme réalisé afin de disposer d'une référence objective. Une seconde évaluation a été réalisée sur une base de 28 carotides (14 patients) segmentées manuellement par un radiologue expert. Les résultats obtenus sur les données " fantôme " montrent que l'approche " contour " conduit à une erreur de mesure de l'aire de la lumière de la carotide segmentée et de la mesure du flux de respectivement 18.4 % et 3.6 %. Ces erreurs sont plus importantes que celles obtenues en utilisant l'approche proposée (respectivement 2.3 % et 0.7 %). Ce bénéfice apparaît encore bien supérieur sur la base de patients avec une sous-estimation des aires et débits sanguins de respectivement 40.5 % et 26.5 % pour l'approche " contour ", contre 14,7 et 6.4 % pour l'approche proposée.