Date of Award


Degree Type



Chemical and Biomedical Engineering

First Advisor

Chatzimavroudis, George

Subject Headings

Fluid dynamic measurements, Nuclear magnetic resonance, Arteries -- Stenosis


Arterial stenosis (narrowing) is due to vascular disease that, if untreated, leads to death. The stenosis causes the development of complex flow which results in high energy losses and great forces between blood and tissue. Traditionally, turbulent flows lead to signal loss in magnetic resonance (MR) image acquisitions, resulting in loss of diagnostic information. With the advancement in MR hardware and software, some of the limitations of MR in turbulent flow fields may have been reduced or eliminated. This study evaluated the potential of MR phase velocity mapping (MRPVM), a clinical velocimetry technique, under turbulent flow conditions such as those developed in arterial stenosis. Specifically, the goal of this study was to determine how accurately MRPVM can quantify the flow rate in experimental models of arterial stenosis under a variety of flow conditions, ranging from laminar to turbulent. Three models were used: a non-occluded straight tube an occluded tube with a 75 area reduction stenosis and an occluded tube with a 94 area reduction stenosis. Flow experiments were conducted under steady-state flow conditions (flow rates of 1.4-10.4 L/min) inside a 1.5 Tesla whole-body clinical Siemens MR scanner. Of interest was to vary several imaging parameters which control the time of the acquisitions and the spatial resolution of the acquired images. Therefore, five different echo times (TE) (2.65-6.0 ms) and three different combinations of the field of view and matrix size, resulting in three different spatial resolutions, were studied. Axial MRPVM acquisitions were performed in all models upstream, at the throat, and downstream of the stenosis. The acquired phase images provided the velocity from which the flow rate was calculated. The MR-measured flow rates were compared with the true flow rates from rotameters to determine the accuracy of MRPVM. The results confirmed that MRPVM can measure the flow rate accurately under laminar flow conditions. In turbulent flow, the accuracy was reduced due to signal lo