Development of particle image velocimetry (PIV) for wall shear stress estimation within a 50cc Penn State artificial heart ventricular chamber

Motivation. Clinical experiences with the Penn State Artificial Hearts have shown clot formation within the chamber is the major problem after implantation, which can cause malfunction of an artificial heart or other native organs---such as embolismic cerebral ischemia. Fluid dynamics, especially wall shear stress, is a prime suspect of undesirable clotting. The main goal of this thesis work is to extend the use of the Particle Image Velocimetry (PIV) for the wall shear rate estimation within a sac-type artificial heart.; Objective. In this study, the work focuses on a 50cc Penn State Artificial Heart Ventricular Chamber. The 50cc model has very recently been initiated by Hershey Medical Center for patients with small circulation volume capacity. The objectives are, first, to develop the PIV technique for wall shear rate estimation, and, second, to investigate the flow dynamics within the 50cc chamber using the extended PIV technique.; Methodologies and outcome. PIV has recently been used in artificial heart works and started to become a standard fluid experimentation in many applications. It has the advantages of being non-invasive, quantitative, and field measurement. However, very few investigators have attempted to estimate wall shear rate using PIV. In the aspect of fluid mechanics, the problem with fluid-wall interface is difficult to solve: high gradient, three dimensionality, and small flow structures, etc. Therefore, estimating wall shear rates from PIV results must be cautious. Furthermore, there are also experimental difficulties for a near wall PIV measurement: image distortion, defocusing, low particle density, wall identification and vibration, and wall image noise. We have developed two techniques to achieve accurate wall shear rate estimation. The first technique is to use conventional PIV. The technique does not require much more experimental and computational effort over standard processing and is very flexible to flow patterns and surface geometries, while providing wall shear-rate uncertainty within +/-20% and a resolution of approximately 480 mum from the wall in our application. The second technique increases the resolution to capture smaller flow structures. Displacement vectors are obtained by a Particle Tracking Algorithm which requires large computational effort. The technique is only limited to one dimensional displacement along a straight surface. The results show improving resolution to approximately 70 mum from the wall with a wall shear-rate uncertainty of 10%.; For the artificial heart work, we studied the motion of the diaphragm in relation to flow patterns within the heart. The study shows a close correlation between the diaphragm opening pattern and the flow structures within the chamber. The global and local flow characteristics and wall shear rates within 50cc chamber were examined under the physiological operating condition. We found the areas of clot depositions correspond to areas of low wall shear-rate along the bottom wall. However, wall shear-rate distribution patterns on a wall within the artificial heart may require more measurement through the depth of the chamber as the result of 3D flow structure.