HPC-based uncertainty quantification for fluidstructure coupling in medical engineering

Abstract

In recent decades biomedical studies with living probands (in vivo) and artificial experiments (in vitro) have been complemented more and more by computation and simulation (in silico). In silico techniques for medical engineering can give for example enhanced information for the diagnosis and risk stratification of cardiovascular disease, one of the most occurring causes of death in the developed countries. Other use cases for in silico methods are given by virtual prototyping and the simulation of possible surgery outcomes. High reliability is a requirement for cardiovascular diagnosis and risk stratification methods especially with surgical decision-making. Given uncertainties in the input data of a simulation, this implies a necessity to quantify the uncertainties in simulation results. Uncertainties can be propagated within a numerical simulation by methods of Uncertainty Quantification (UQ).