Mathematical cardiac electrophysiology (Q2449323)

This book aims to present recent mathematical and numerical advances in cardiac electrophysiology, in a comprehensive manner and with a special emphasis on large-scale numerical simulations. This is certainly a quite difficult undertaking for a highly interdisciplinary subject, drawing from mathematical physiology, large-scale computing and bioengineering; the authors employ in their quest about 130 figures and compile a reference list of about 590 items. Although the desired outcome is highly specialized and research-oriented, the presentation is as clear as possible, which makes the book a valuable resource for graduate students and researchers in electrophysiology and cardiology, not only for the amount and quality of the material reviewed, but also for the quality of the presentation. There is also an appendix which lists further useful tools (main cardiac simulation projects, simulation libraries and modelling environments, and related monographs). The book consists of nine chapters, corresponding (apart from the first one) to the major modelling or numerical tools employed throughout the book. The first chapter reviews the basic physiology and anatomy of the heart, most notably the cardiac tissue organization and the functional characteristics of the fundamental cell types, together with the main phases of cardiac action potentials (the variations of the potential difference across the cell membrane caused by the excitation of a cardiac cell). The main features of the electrocardiogram (ECG) are also described and the main cardiac imaging techniques (invasive and non-invasive) are also surveyed. The second chapter reviews the main mathematical models of cellular bioelectric activity: the Nernst-Planck equation, the Goldman-Hodgkin-Katz current-voltage relation and the Nernst equilibrium potential, together with their derivations. The celebrated Hodgkin-Huxley (HH) cardiac action potential model is also derived, together with the ventricular models based on the HH formalism (Beeler-Reuter, Luo-Rudy, FitzHugh-Nagumo and Fenton-Karma). The third chapter moves one step up, from cells to cell arrangements. The authors start with one-dimensional models of cardiac fibers, deriving the cable equation and continuing with a homogenization technique. Several concepts relating to traveling waves are presented, these concepts being then illustrated for the case of the so-called bistable equation. For the higher-dimensional case, further models of cardiac tissue are introduced, leading to the bidomain model, in which the intracellular and extracellular spaces are thought as being separated by an active membrane. From a mathematical viewpoint, well-posedness results in a distributional sense are obtained via a Hilbert space variational formulation. Alternative approaches to well-posedness (semi-discretizations, Galerkin expansions and fixed point arguments) are also outlined. The fourth chapter moves again one step up, from models of cell arrangements to reduced macroscopic models. Also, the focus switches (mostly) from analyzing mathematical formalisms to discussing the numerical specifics of the models under consideration (the linear anisotropic monodomain model, to which the bidomain model reduces when both domains have the same anisotropy ratio, the eikonal models and the relaxed non-linear anisotropic monodomain model). The fifth chapter is dedicated to the modelling of anisotropic cardiac sources. This chapter is again more mathematically-oriented, differential and integral formulations of the potential field being presented. The existence and uniqueness of the (weak) solution is established by using a variational formula. A source-splitting technique and an interpretation of the field components (axial component, jump component and heart surface component) are also presented. This chapter concludes with a numerical simulation of the source-splitting technique. The sixth chapter reviews the inverse problem of electrocardiology, namely the estimation of the epicardial extracellular potential distribution from the body surface map. Formulations in terms of the potential, of the wavefront and of the cardiac sources are presented and discussed. The seventh chapter presents the main numerical techniques used for the space and time discretizations of the monodomain and bidomain cardiac models. While the space discretization is based on the Galerkin procedure, the time discretization may be explicit, fully implicit or semi-implicit and may employ operator-splitting methods and decoupling techniques. Finally, several considerations regarding the numerical approximation of the eikonal-diffusion equation are presented. The eighth chapter is dedicated to the construction and analysis of parallel solvers for the discrete bidomain system, based on overlapping Schwarz methods. Scalable convergence rate bounds for two-level and multilevel additive Schwarz preconditioners are then provided, the scalability of the solver on parallel machines following as a result. The combination between Schwarz preconditioners and block preconditioners is also illustrated. Finally, the ninth chapter illustrates the applications of the solvers mentioned in the previous chapter to the simulation of some of the most important phenomena of cardiac electrophysiology (the genesis of cardiac excitation and virtual electrode polarization, the anisotropic propagation of excitation and recovery fronts, the morphology of electrograms, the effects of cardiac heterogeneities on fronts propagation, the computation of excitation and repolarization time markers, the effects of ischemic regions and the simulation of cardiac reentry phenomena).