A theoretical fluid-structure interaction (FSI) model for the stress field of atherosclerotic coronary arteries are obtained and the influence of various characteristics on the stress distribution in diseased coronary arteries is highlighted. A reliable model is developed (and hence accurate heart attack prediction), the following factors are incorporated: (1) non-Newtonian blood flow; (2) artery’s tapered shape; (3) the micro-calcification of the plaque; (4) blood pulsation. Incorporating these factors in the model makes it possible to accurately predict plaque ruptures. The system is modelled based on a 3D fluid-structure interaction analysis via the finite element method (FEM). Experimental data from previous studies are used to generate a realistic material model. The generated model is utilised as a predictive model for plaque rupture and to determine high risk situations in the coronary arteries. It is shown that incorporating the physiological flow rate in the model, the wall shear stress (WSS) (stresses impose to the plaque from blood) and von Mises stresses (stresses in the plaque) are predicted accurately. Also it is shown that microcalcification increases the von Mises stress substantially in the plaque, when the WSS remains the same. Considering tapered shape of the artery is also shown to be important for predicting correct values of both shear and von Mises stresses.