Up to 50% of strokes may be caused by rupture of atheromatous plaque in the carotid arteries in the neck. Such strokes are potentially preventable by carotid endarterectomy, an operation during which the plaque is surgically excised [1]. Appropriate selection of patients for surgery is a crucial clinical issue and is currently based on the severity of carotid luminal stenosis. There is, however, growing evidence to suggest that luminal stenosis per se may not be the best predictor of subsequent strokes. Consequently, other factors such as intrinsic plaque morphology and biomechanical stress are increasingly thought to be more important risk factors for plaque rupture [2, 3]. This study explores the role of stress as a marker of plaque vulnerability using a combination of magnetic resonance imaging (MRI) and finite element analysis (FEA). Ten patients (5 symptomatic, 5 asymptomatic) with carotid stenosis underwent high-resolution in vivo pre-operative multi-sequence MRI. Intact carotid plaques were retrieved after surgery for histological analysis. Segmentation was based primarily on MRI for fibrous cap and lipid core, when compared with histology. Each patient axial MR slices were segmented and all contours were traced to generate the boundaries of lipid core, fibrous cap, vessel lumen and wall based on net intensity characteristics. The mesh was generated for each slice and finite element analysis was conducted in each case to determine the stresses within every plaque. The plaque components were modelled as elastic materials with different Young’s moduli and Poisson’s ratios. The internal luminal pressure was assigned at 15kPa, and large deformation analysis was performed for plaque stress simulation. Statistical analysis using a non-paired t test was carried out for the comparison of symptomatic and asymptomatic patients. High stress concentration is found at the shoulder region of the symptomatic patient, while in asymptomatic patient the stress distribution is much more averaged. The mean principal stresses in the plaques of symptomatic patients are significantly higher than those of asymptomatic patients (533.4±185.3 versus 203.5±121.3 kPa, p<0.05). This study illustrates the potential application of state-of-the-art computational modelling techniques to solving important clinical problems in medicine. This FEA suggests that the maximum stresses within the plaques of symptomatic patients are higher than those of asymptomatic patients. Mechanical stress in the carotid plaque, therefore, ought to be regarded as a complementary indicator of plaque rupture risk alongside the traditional measure of stenosis. A combination of high resolution MR and FEA could potentially act as a useful tool for assessment of risk in patients with carotid disease.