Wall shear stress (WSS) has long been identified as a factor in the development of atherosclerotic lesions. Autopsy studies have revealed a strong tendency for lesion development at arterial branch sites and along the inner walls of curvature areas that, in theory, should experience low WSS. Calculations of coronary artery WSS have typically been based upon average models of coronary artery geometry with average flow conditions and then compared to average lesion distributions. With all the averaging involved, a more detailed knowledge of the correlation between WSS and atherosclerotic lesion development might be obscured. Recent advancements in hemodynamic modeling now enable the calculation of WSS in individual subjects. An image-based approach for patient-specific calculation of in vivo WSS using computational fluid dynamics (CFD) would allow a more direct study of this correlation. New state-of-the-art technologies in multi-detector computed tomography (CT) and 3.0 Tesla magnetic resonance imaging (MRI) offer potential improvements for the measurement of coronary artery geometry and blood flow. The overall objective of this research was to evaluate the quantitative accuracy of multi-detector CT and 3.0 Tesla MRI and incorporate those imaging modalities into a patient-specific CFD model of coronary artery WSS. Using a series of vessel motion phantoms, it has been shown that 64-detector CT can provide accurate measurements of coronary artery geometry for heart rates below 70 beats per minute. A flow phantom was used to validate the use of navigator-echo gated, phase contrast MRI at 3.0 Tesla to measure velocity of coronary blood flow. Patient-specific, time-resolved CFD models of coronary WSS were created for two subjects. Furthermore, it was determined that population-average velocity curves or steady state velocities can predict locations of high or low WSS with high degrees of accuracy compared to the use of patient-specific blood flow velocity measurements as CFD boundary conditions. This work is significant because it constitutes the first technique to non-invasively calculate in vivo coronary artery WSS using image-based, patient-specific modeling.