Assessing the risk of rupture of intracranial aneurysms is important for clinicians because the natural rupture risk can be exceeded by the small but significant risk carried by current treatments. To this end numerous investigators have used image-based computational fluid dynamics models to extract patient-specific hemodynamics information, but there is no consensus on which variables or hemodynamic characteristics are the most important. This paper describes a computational framework to study and characterize the hemodynamic environment of cerebral aneurysms in order to relate it to clinical events such as growth or rupture. In particular, a number of hemodynamic quantities are proposed to describe the most salient features of these hemodynamic environments. Application to a patient population indicates that ruptured aneurysms tend to have concentrated inflows, concentrated wall shear stress distributions, high maximal wall shear stress and smaller viscous dissipation ratios than unruptured aneurysms. Furthermore, these statistical associations are largely unaffected by the choice of physiologic flow conditions. This confirms the notion that hemodynamic information derived from image-based computational models can be used to assess aneurysm rupture risk, to test hypotheses about the mechanisms responsible for aneurysm formation, progression and rupture, and to answer specific clinical questions.