The mortality caused by cardiovascular diseases is dramatically increasing [1-4]. Atherosclerosis is among the main contributors to this extremely high cardiovascular disease mortality [1]. Atherosclerosis is controlled by mechanical forces exerted by the flow of blood on the inner lining of arteries, the endothelium [5-9]. In order to fight this lethal disease a realistic computational model is required that offers an accurate understanding of the effect of blood flow on the arterial wall. In order to realistically describe complex blood flow patterns and their interaction with the arterial wall, we have developed an integrated computational technique that takes into account both the particulate cellular composition of the blood and the interactions between the particulate blood and the vessel wall, in macro-circulation. The cellular composition of the blood was modelled using a multiphase fluid dynamics method by computing both an Eulerian fluid domain for modelling blood plasma, and a Lagrangian solid domain that represented the blood cells. Interactions of the blood with the vessel wall were realistically modelled using a novel iterative immersed boundary method. Both the multiphase technique and the immersed boundary method were validated by comparing obtained numerical results with the literature [10-1], and a high degree of similarity was found. Moreover, our multiscale integrated model was applied to characterize clinically relevant flow patterns from atherosclerotic arterial segments. Our model revealed that under realistic non- Newtonian multiphase cellcontaining blood flow conditions, the pressure at the stenosis site is c. 30% higher than predicted by single-phase Newtonian fluid dynamics models [10,12]. This effect is most probably explained by a decrease of c. 28% in the peak velocity. The latest seems to be caused by the momentum interchange due to the collision of the particles (blood cells) with each other, with the vessel wall, and with the fluid phase (plasma). The most significant difference in velocities between the single-phase and the two-phase results was registered at the stenosis site. It was also found that high blood cell content (peak particle concentration) correlates with increasing flow laminarization. This means a decrease in velocity, which is more evidenced at the site of the stenosis. Additionally, our findings indicated that wakes forming downstream of the stenosis might be much weaker in non-Newtonian blood flow simulations compared to the Newtonian models. Taken together, our mathematical model and the resulting computational results might offer a more accurate understanding of the effects of realistic blood flow patterns on atherosclerotic vessels. This is of crucial importance since this deadly disease is initiated and progressed by blood flow related mechanical factors, such as the wall shear stress [8-9,22-25].