Page 119 - Control Theory in Biomedical Engineering

CHAPTER 5

Genetic fuzzy logic based system

for arrhythmia classification

Hela Lassoued, Raouf Ketata
National Institute of Applied Sciences and Technology (INSAT), Tunis, Tunisia

1 Introduction

Recently, machine-learning approaches have had an important impact in
many areas of science and technology (Angra and Sachin, 2017). Their quick
evolution is providing chances to improve the quality of decision-making
( Johnson et al., 2016). In fact, the world of medical research has faith in these
approaches, which through experience and self-learning can resolve many
issues, such as the classification and diagnosis of arrhythmias (Rajkomar
et al., 2019). Accordingly, the use of a decision support system (DSS) based
on machine-learning approaches becomes a necessity in medicine, especially
when the diagnosis requires a lot of knowledge and experience. Moreover, a
DSS is mostly used to save time and allow for rapid and efficient action. DSS
systems include reasoning, evaluation, learning and many other skills of
human intelligence. They guarantee not only the neutrality and objectivity
of the expert but also the quality of the decision. They are also largely used
by experts as a guide in order to facilitate communication between them.
Today, some cardiovascular arrhythmias such as ventricular tachycardia
and ventricular fibrillation lead unexpectedly to cardiac arrest and in most
cases lead to sudden death (Huynh et al., 2014). Moreover, statistics claim
that cardiovascular arrhythmias are among the leading causes of death world-
wide (Ettehad et al., 2016). These arrhythmias, resulting from cardiac dys-
function, are explained by the presence of certain factors, such as unhealthy
eating habits, lack of physical activity, high stress, family history, age and
many others (Chen et al., 2015). One way to achieve appropriate care for
these arrhythmias is to rely on the electrical activity of the heart, illustrated
by the electrocardiogram (ECG) signal (Chen et al., 2017). Despite the tech-
nological evolution in the field of medical instrumentation, ECG signal
remains an essential examination in cardiology. However, its manual analysis
requires careful inspection due to its long duration (24–48hours). The DSS

Control Theory in Biomedical Engineering © 2020 Elsevier Inc.
https://doi.org/10.1016/B978-0-12-821350-6.00005-6 All rights reserved. 105