Page 39 - Artificial Intelligence for Computational Modeling of the Heart

Chapter 1 Multi-scale models of the heart for patient-specific simulations 9

along the bundle of His and its terminal fibers with a speed that
is about four times larger than in the myocardial tissue [59]. As a
result, the depolarization of the ventricle is triggered in the endo-
cardium and then reaches the epicardium.
Detailed studies of the electrical conduction system have been
based on histology as well as invasive measurements of electrical
activation [60]. Such studies shed light on the role and function of
the specialized myocardial tissues. For instance, early studies sug-
gested that the network of Purkinje fibers is densely diffused in the
subendocardium, and that it connects to the ventricular muscle in
discrete regions called Purkinje fiber-ventricular muscle junctions
(PVJ) [61].
The coordinated electrical activity of the myocardium induces
a dynamic change of the electrical potential in the tissues and or-
gans surrounding the heart. This produces a measurable electrical
signal on the surface of the body, which can be captured by elec-
trocardiography. This is the most accessible measurement of the
electrical activity of the heart, and has been used as a valuable
tool to understand the healthy electrophysiology as well as detect
pathological alterations [62]. Invasive measurement of electrical
activity on the endocardial surface additionally allows the assess-
ment of the role of local substrate properties and abnormalities
(such as the presence of scarred or fibrotic myocardial tissue) on
the activation pattern. For this reason, it has become a crucial tool
for instance in the planning and delivery of ablation therapy for
cardiac arrhythmias [63].
Physiological modeling of cardiac electrophysiology aims at
identifying and understanding the underlying biophysical phe-
nomena, with the ultimate goals of allowing risk stratification
of patient populations based on estimated model parameters;
and designing interventional strategies for the optimal treatment
of electrical heart diseases such as cardiac arrhythmias. These
have been topics of interest in the research community and have
sparked the development of various modeling approaches. Sem-
inal work by Hodgkin and Huxley introduced a modeling frame-
work to study cellular excitability, paving the way for computa-
tional modeling of electrophysiology at the cellular and tissue
level [12,64–66].
The following sections provide a brief overview of the most
common modeling approaches, including cellular, tissue and
body level models.

1.2.1 Cellular electrophysiology
Historically, phenomenological models were the first models to
be developed, derived from experimental observations on nerves