|Speaker:|| Philippe Moireau|
|Affiliation:|| M3DISIM, INRIA - Paris - France|
|When:|| Thursday 7th June 2018|
|Abstract:|| When considering the modeling of the cardiovascular system and more specifically the heart, there is a need for the personalization of not only the geometry but various aspects of the physical model: uncertain initial conditions, constitutive parameters or boundary conditions. Indeed, identifying key parameters - using measurements of a type that is available in medical imaging - can provide patient specific simulations that can be used by clinicians in their diagnosis. In other engineering fields where large amounts of data are available - like weather forecasting or climatology - it is now common to tackle these uncertainties in the models by data assimilation procedures - variational (4D-var) or sequential (Kalman like). In this context, our objective is to propose and analyze data assimilation methods adapted to the specificity of the biomechanical systems considered and to the available data, in particular image sequences.
|Note:|| Philippe Moireau is a senior researcher at Inria, head of the M3DISIM Inria research team also affiliated with Ecole Polytechnique. He graduated from Ecole Polytechnique, Telecom ParisTech, and Paris 6 University, with a dual background in applied mathematics and computer sciences. His areas of expertise are in mathematical modeling, numerical analysis and data assimilation, with applications to the cardiovascular problems.|