Title: A parallel accelerated adaptive mesh algorithm for the solution of electrical models of the heart

Authors: Rafael Sachetto Oliveira; Bernardo M. Rocha; Denise Burgarelli; Wagner Meira Jr.; Rodrigo Weber Dos Santos

Addresses: Department of Computer Science, Federal University of São João de Rei, Av. Visconde do Rio Preto, s/n, Colônia do Bengo, São João del-Rei, MG 36301-360, Brazil; Department of Computer Science, Federal University of Minas Gerais, Av. Antônio Carlos, 6627, Pampulha, Belo Horizonte, MG 31270-010, Brazil. ' National Laboratory for Scientific Computing, Avenida Getúlio Vargas, 333 Petrópolis, RJ 25651-076, Brazil; Department of Computer Science, Federal University of Juiz de Fora, Rua José Kelmer, S/n, Campus Universitário, Bairro São Pedro, Juiz de Fora, MG 36036-330, Brazil. ' Department of Mathematics, Federal University of Minas Gerais, Av. Antônio Carlos, 6627, Pampulha, Belo Horizonte, MG 31270-010, Brazil. ' Department of Computer Science, Federal University of Minas Gerais, Av. Antônio Carlos, 6627, Pampulha, Belo Horizonte, MG 31270-010, Brazil. ' Department of Computer Science e Computational Modeling Program, Federal University of Juiz de Fora, Rua José Kelmer, S/n, Campus Universitário, Bairro São Pedro, Juiz de Fora, MG 36036-330, Brazil

Abstract: Computer models have become valuable tools for the study and comprehension of the complex phenomena of cardiac electrophysiology. However, the high complexity of the biophysical processes translates into complex mathematical and computational models. In this paper, we evaluate a parallel numerical algorithm based on mesh adaptivity and finite volume method aiming to accelerate these simulations. This is a very attractive approach since the spreading electrical wavefront corresponds only to a small fraction of the cardiac tissue. Usually, the numerical solution of the partial differential equations that model the phenomenon requires very fine spatial discretisation to follow the wavefront, which is approximately 0.2 mm. The use of uniform meshes leads to high computational cost as it requires a large number of mesh points. In this sense, the tests reported in this work show that simulations of two-dimensional models of cardiac tissue have been accelerated by more than 340 times using the adaptive mesh algorithm and parallel computing, with no significant loss in accuracy.

Keywords: adaptive mesh; cardiac electrophysiology; simulation; parallel computing; electrical models; modelling; heart tissue; finite volume method.

DOI: 10.1504/IJHPSA.2012.050988

International Journal of High Performance Systems Architecture, 2012 Vol.4 No.2, pp.89 - 100

Received: 10 Aug 2012
Accepted: 07 Sep 2012
Published online: 02 Sep 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article