A quantitative analysis of the impact of inflow velocity waveform in computational hemodynamic models of coronary arteries

Hemodynamics plays a crucial role in understanding vessel wall disorders, so a quantitative study of hemodynamics is of paramount importance. A strong medical research focus is put on the study of the hemodynamics of coronary arteries, since approximately one-third of the deaths related to cardiovascular issues are due to atherosclerosis affecting the coronary artery walls. Computational fluid dynamics (CFD) is a widely used method of performing a quantitative analysis of the hemodynamics inside blood vessels. However, CFD presents several uncertainties related to the imposition of non-fully specific boundary conditions, which has an impact on simulation results and, subsequently, on the wall shear stress (WSS)-based descriptors used for quantitatively assessing some features of the blood flow. The aim of this thesis is to assess and quantify the impact of boundary conditions, in terms of shape of velocity waveforms applied at the inlet section of patient-specific coronary artery models, on WSS-based descriptors of major clinical and biological interest. In detail, the WSS-based descriptors obtained using patient-specific velocity waveforms were compared to those obtained using the generalized velocity waveforms applied at the inlet section of the same coronary artery models. All the above was aimed at determining whether generalized velocity waveforms can be used instead of patient-specific velocity waveforms to be acquired through invasive techniques. Patient-specific coronary artery models were reconstructed from cine-angiographic images and unsteady fluid dynamics simulations were performed by imposing the following inlet boundary conditions: (1) patient-specific velocity waveforms measured by means of intravascular Doppler; (2) generalized velocity waveforms obtained from averaging all the patient-specific velocity signals with the same average flow rate of the patient-specific velocity waveforms. Several WSS-based descriptors were computed: Time Averaged WSS (TAWSS), Oscillatory Shear Index (OSI), transverse WSS (transWSS), axial TAWSS (TAWSSax), secondary TAWSS (TAWSSsc), and Topological Shear Variation Index (TSVI) and evaluated in terms of average value at the vessel wall and of luminal surface exposed to a disturbed blood flow. The main results of this thesis are as follows: (1) statistically significant differences were shown between the average values of all the hemodynamic descriptors considered, excluding TAWSSsc; (2) a rather high correlation was shown between the average values of all the WSS-based descriptors considered; (3) for all the hemodynamic descriptors considered, no statistically significant differences were shown between the percentages of luminal surface exposed to a disturbed flow; (4) for all the hemodynamic descriptors considered, a rather high correlation was shown between the percentages of luminal surface exposed to a disturbed flow; (5) for all the hemodynamic descriptors considered, excluding OSI and TSVI, good co-localization was shown between the luminal surfaces exposed to a disturbed flow. Thus, the results of this thesis suggest that, for some WSS-based descriptors, generalized velocity waveforms can be used instead of patient-specific velocity waveforms, while a greater attention should be put on other WSS-based descriptors such as OSI and TSVI. However, this issue should be explored through further studies, given the small patient population being investigated here.