Classification of Blood Rheological Models through an Idealized Symmetrical Bifurcation

The assumed rheological behavior of blood influences the hemodynamic characteristics numerical flow simulations. Until now, alternative specifications have been utilized, with uncertain implications for results obtained. This work aims to group sixteen models in homogeneous clusters, by exploiting data generated from simulations on an idealized symmetrical arterial bifurcation. Blood is be pulsatile and simulated using a commercial finite volume solver. An appropriate mesh convergence study performed, all are collected at three different time instants throughout cardiac cycle: peak systole, early diastole, late diastole. Six variables computed: average wall shear stress, oscillatory index, relative residence time, global local non-Newtonian importance factor, effect factor. resulting analyzed hierarchical agglomerative clustering algorithms, which constitute typical unsupervised classification methods. Interestingly, can partitioned into groups, whereas appear as outliers do not belong any partition. Our findings suggest that defined similar manner mathematical perspective may behave substantially differently terms they produce. On other hand, characterized formulations same statistical (cluster) thus considered interchangeably.