In this talk I will summarize our research at the Center for Computational Fluid Dynamics focusing on understanding the underlying mechanisms responsible for the development, progression and rupture of intracranial aneurysms using patient-specific computational models constructed from 3D medical images. Specifically, I will describe studies using longitudinal data of unruptured growing and stable aneurysms as well as cross-sectional data of ruptured and unruptured aneurysms. The objective of these studies is to identify the hemodynamic conditions that predispose aneurysms for growth and rupture with the goal of improving current diagnostics and evaluation of patients with cerebral aneurysms in order to avoid unnecessary interventions and treat only those aneurysms that are a high risk of bleeding.