Activation Propagation in Cardiac Ventricles Using the Model with the Conducting System

The characteristics of the activation propagation in a geometrical model of cardiac ventricles formed by several ellipsoids, with or without a fast conducting endocardial layer representing the Purkinje fibers and with activation started in one or several endocardial locations were compared in this study. The activation propagation was simulated by two approaches. In the first one, temporal and spatial changes of the membrane potential were numerically modeled by a reaction-diffusion (RD) equation of the propagation with the transmembrane ionic current density defined by modified FitzHugh-Nagumo equations. The propagation was numerically solved in Comsol Multiphysics environment. In the second approach, the electrical excitation of the working ventricular myocardium was simulated by a cellular automaton (CA) model that was implemented in Matlab environment. Local activation times in both ventricles were computed by both approaches in ventricular models with and without the fast conducting layer representing Purkinje fibers (heterogeneous and homogeneous model). In both models, the activation was initiated either in a single starting position or gradually in nine starting positions imitating more physiological conditions. Despite some differences in the activation sequences, by both approaches acceptable activation times of the whole ventricles were obtained for the homogeneous model with nine starting points and for the model with conducting layer regardless of number of starting positions.