Magnetic resonance imaging (MRI) is an imaging modality which provides good conditions for studies of flow and motion in the cardiovascular system. When using the phase contrast imaging technique, it is possible to perform velocity measurements in moving tissues and in blood. The acquired velocity data can be used for subsequent calculation of different parameters used for the assessment of cardiovascular function. These parameters rely on the accuracy and reproducibility of the velocity measurements.

This thesis includes an introduction to phase contrast imaging and how displacement artifacts may hamper the accuracy of velocity measurements using phase contrast imaging. A method for correction of this artifact was developed using the knowledge of the times at which velocity and spatial encoding are carried out in the pulse sequence. The elaboration of this correction method included modifications of a 3D phase contrast pulse sequence and development of post processing algorithms involving streamline calculations. An alternative approach for correction of displacement artifacts is also presented where a pulse sequence, based on tripolar waveforms, inherently compensates for the artifact and, consequently, do not require any post processing algorithms.

The use of the sensitivity encoding (SENSE) method in MRI implies that the scanning time can be reduced dramatically. The SENSE technique can be combined with phase contrast imaging. An important question to answer is how the use of SENSE influences the accuracy and reproducibility in phase contrast imaging. In this thesis expressions are derived showing how the reproducibility of velocity and flow measurements are influenced when applying SENSE at different reduction factors.

As a consequence of the introduction of parallel imaging techniques, such as SENSE, modem MRI systems have the capability of assessing individual coil sensitivities. This information implies new alternatives in the combination of multiple phase images using phased array coils. Conventionally multiple phase images are combined using the weighted mean method, where the squared magnitude values are used as weights. In order to compare these two combination methods, theoretical and experimental studies were performed.

The results of this thesis show that corrupted velocity measurements caused by displacement artifacts can be alleviated using suggested correction methods or pulse sequences. This is of importance for subsequent flow analysis and visualization of stenotic and oblique flow. Phase contrast imaging in combination with SENSE provides opportunities to obtain large reductions in scanning times while keeping the increase of the reproducibility in the velocity and flow measurements within reasonable limits. In terms of phase reproducibility it is shown that the use of individual coil sensitivities in the combination of multiple phase images is equivalent to the conventional weighted mean method. In the accompanying magnitude images, however, signal variations due to coil sensitivities are compensated for when including estimated coil sensitivities.