Compact, low cost Particle Image Velocimetry system for the in vitro analysis of cardiovascular flows.

In the presented thesis work, an experimental setup was designed and implemented for the in vitro study of haemodynamics in a model of a healthy coronary artery and of a stenotic coronary artery. The setup was fine-tuned with the aim of obtaining a compact and low cost test-bench that could be employed in various applications in the biomedical field. In particular, it can be used in the Artery in microgravity project in order to investigate the effect of microgravity on the fluid dynamic structures that are generated within the coronary models. The experimental equipment was chosen in such a way as to replace some characteristic elements of the setups for the traditional PIV (e.g. high frame rate cameras or pulsed light lasers) with low-cost ones. Specifically, the acquisitions were made using a mobile phone, while for the lighting, it was decided to use a continuous white LED light. The hydraulic circuit was suitably characterized in such a way as to reproduce a physiological flow-rate range (20-80 ml/min) relating to the coronary arteries. Subsequently, an in-depth study was carried out on flow visualization techniques and two different optical techniques were chosen. Both techniques made it possible to obtain the velocity fields, within the two models, of a blood-mimicking fluid inseminated with polyamide particles. The first optical technique chosen is Particle Image Velocimetry (PIV), which was implemented using the PIVlab algorithm in Matlab. The second technique taken into consideration is Optical Flow (OF), implemented in Matlab as well. The validation of the velocity fields obtained through the implementation of both approaches was carried out through a comparison with the results of the third PIV challenge, compared with the characteristic velocity values obtained from the imposed flow rates and through comparisons with numerical simulations. The main sources of uncertainty of the method were also discussed and analyzed, in particular those that led to an underestimation of the fluid velocity values in areas where high displacements are reached, such as in the narrowing caused by the presence of stenosis. From the velocity fields obtained from the two algorithms, it was possible to visualize fluid dynamic patterns of interest in the field of cardiovascular biomechanics. In fact, low speed recirculation areas downstream of the stenosis were detected, which are among the main causes involved in the genesis and development of cardiovascular diseases such as atherosclerosis. In order to investigate and deepen the analysis in the post-stenotic region, which is the region of interest, starting from the motion fields, further quantities and fluid dynamics parameters were also calculated and discussed. In conclusion, it can be stated that the proposed approach represents an easy to implement and low cost methodology that allows to obtain robust qualitative data to be able to investigate fluid dynamics in coronary arteries in vitro. The differences between the performances of both algorithms are highlighted in higher computational costs for the OF, while PIVlab has a lower maximum spatial resolution. A series of improvements have also been proposed which, if implemented in future applications of the presented method, would allow to reduce the sources of uncertainty, in order to further improve the quality of the results.