Identification of parameters describing phenomenological cardiac action potential models using sigma-point methods

Abstract:

Phenomenological action potential (AP) models allow reproducing characteristic features of cardiomyocytes' electrical activity without fully describing the underlying biophysics, thus being very useful for whole-heart elec-trophysiological simulations. Methods to identify the parameter values of phenomenological models commonly attempt to reproduce specific AP properties rather than the whole AP waveform. In this work we propose the use of a sequential estimation approach based on sigma-point filters to adjust such parameters. The proposed methodology has been tested in estimating the parameters of the phenomenological Bueno-Cherry-Fenton model to replicate the APs generated with in silico models as well as experimentally measured APs. With the new method the whole AP waveforms can be reproduced more accurately than with previous parameter fitting methods and the AP duration restitution curves are in better agreement with available experimental data.

Año de publicación:

2017

Keywords:

Fuente:

scopus