It has been hypothesised that the stress distribution within the arterial wall may provide an indicator of atherosclerotic plaque rupture. This study presents an automated method for the generation of finite element models of the carotid bifurcation from in vivo computerised tomographic angiography. Models generated using this method have been used to investigate plaque vulnerability, assessing the influence of geometric factors and the stress distribution within the wall.Structured hexahedral meshes of the carotid bifurcation were created using a custom built automated system. Systolic pressure and appropriate boundary conditions were applied to each of the models. Six symptomatic patients were considered, generating models from each of their left and right carotid bifurcations (12 in total). Results from their symptomatic bifurcation were compared with their contralateral asymptomatic side.K δ , a measure of the difference in curvature of the inner and outer plaque surfaces, was found to be significantly higher in symptomatic arteries than in asymptomatic arteries (p<0.05). The location of maximum K δ (the plaque shoulders) was also found to be a region of high von Mises stress yet no significant difference was found in maximum von Mises stress between the symptomatic and asymptomatic groups. Results suggest that K δ is an important factor in the development of a symptomatic plaque and with further investigation could be a useful indicator of plaque rupture.