Identification of nonlinear cardiac cell dynamics using radial basis function regression

We present a novel method for the identification of the dynamics of physiological cardiac cell models. The main aim of the technique is to improve the computational efficiency of large-scale simulations of the electrical activity of the heart. The method identifies the dynamical attractor of a detailed physiological model using statistical learning techniques. In particular, a radial basis function regression method is used to capture the intrinsic dynamical features of the model, thus reducing the computational cost to quantitatively generate cardiac action potentials in a wide range of pacing conditions. The approach permits to recover key properties such as the action potential morphology and duration in a wide range of pacing frequencies.