Skip to Main content Skip to Navigation
Other publications

Reduced Order Methods: Faster Solvers For Cardiac Electro-physiology problems

Cesare Corrado 1 Nejib Zemzemi 1 yves Coudière 1, 2 
1 CARMEN - Modélisation et calculs pour l'électrophysiologie cardiaque
IMB - Institut de Mathématiques de Bordeaux, Inria Bordeaux - Sud-Ouest, IHU-LIRYC
Abstract : The numerical simulation of the electrical activity of the heart is highly demanding in terms of computational cost. This is explained by the multiscale behavior characterizing the bi-domain equations governing the propagation of the electrical wave. In fact small space and time scales are required in order to accurately solve the bidomain problem and compute conduction velocities and activation times. Reduced order methods have been recently introduced and successfully used different fields, they allow solving the electrical wave propagation as accurate as finite element method (or any other space discretization method). The idea is to project the linear problem on a reduced basis instead of the full finite element basis. The reduced basis is extracted from a dataset of precomputed solution. In this work we consider the Proper Orthogonal Decomposition as reduced order model. This method allows good performance in terms of computational time. Moreover we show that this method improves the scalability when solving the bidomain equations in parallel.
Document type :
Other publications
Complete list of metadata
Contributor : Cesare Corrado Connect in order to contact the contributor
Submitted on : Wednesday, January 8, 2014 - 3:48:33 PM
Last modification on : Wednesday, April 27, 2022 - 3:46:58 AM


  • HAL Id : hal-00925809, version 1



Cesare Corrado, Nejib Zemzemi, yves Coudière. Reduced Order Methods: Faster Solvers For Cardiac Electro-physiology problems. 2013. ⟨hal-00925809⟩



Record views