Personalized, or patient-specific, healthcare is gaining ground in the clinical community as a paradigm allowing for the enhancement of treatment specificity and efficacy. A particularly promising track along this line involves the application of adequately sophisticated computational simulation models for the evaluation of the progress of disease, for the determination of long-term risk and for the planning of optimal treatment strategies. Here, we focus on the modeling of cerebral aneurysms (one of the leading causes for stroke). We discuss basic features of the disease and present techniques and results that open the way for predicting the evolution of a specific aneurysm, for a particular patient: We argue that questions like “Will it grow further?”, “Will it stabilize?”, “Will it rupture?” can be addressed successfully via computational models. Beyond diagnostics, we discuss interventions and their rationalization via computer modeling. Simulation can be a central tool in discovering, designing and using devices and procedures: “How should a microcatheter tip be shaped to improve guidance?”, “Will a particular platinum coil packing configuration lead to thrombosis in a clinically meaningful time-frame?” Simulation of blood flow, of wall fiber remodeling, of the biochemistry of thrombosis and of their interplay are some of the tools we employ. Techniques ranging from porous media and finite volume to finite elements and reaction kinetics are brought into play and coupled together to account for the complex multiphysics and multiscale phenomena involved. We argue that if models of this nature are to find practical and routine use in a clinical setting, very challenging requirements of comprehensiveness, integration and computational efficiency have to be met.