Blood flow simulation in a grid environment

With computers reaching a level of performance that allows to run ever more complex simulations on less expensive hardware technical simulation becomes interesting for more and more fields outside the traditional realm of defense and aerospace. Biomedical simulation is one such field, which allows integrating computational fluid dynamics into the daily routine of research hospitals. The backbone for such an approach is GRID computing allowing to access distributed and remote resources in a simple way. The test case for this is the treatment of abdominal aorta aneurysms (AAA) a dilatation of the artery. AAA may ripe and spring a leak or ruptures putting the patient at risk of death with a mortality rate at currently about 20%. One way to treat such a AAA is to implant stent grafts in order to channel the blood at its way through the aneurysm. However, complications may occur after the treatment. In order to avoid such complications and to improve understanding and planning of the procedure, in advance simulation of the flow pattern and the mechanical behavior of the system is necessary. For that purpose, medical data received from CT or MRI scans has to be transformed into a computable mesh for CFD simulations on supercomputers and the results have to be transferred back to the surgeon preparing the procedure.