IEEE Computational Intelligence Magazine - November 2020 - 11

disease. In this survey, we aim to assemble and summarize the
latest developments and insights in transforming computational
intelligence approaches, such as machine learning, evolutionary
computation, soft computing, and big data analytics, into practical applications for fighting COVID-19. We also explore
some potential research issues on computational intelligence
for defeating the pandemic.

I. Introduction

OVID-19 is an infectious disease caused by a novel
coronavirus and has been declared by the World
Health Organization (WHO) as a pandemic in
March 2020. Since this disease was first identified in
December 2019, it has become a global pandemic and has
caused infections in millions of people. The coronavirus
death toll surpassed 687,000 worldwide as of the end of
July 2020, and the number of infections and deaths continues to rise. Such an extremely serious situation has led to
high threat in healthcare systems worldwide and severe
damage in the global economy being. To combat COVID19, many countries are working to develop novel and effective mechanisms to overcome this disaster. Governments,
industry leaders, and academics alike are devoting substantial resources and effort into mitigating the effects of the
pandemic. Over the past few months, various emerging
solutions and systems for combating COVID-19 have been
developed and deployed. For example, fast screening methods utilizing different types of clinical data, including
X-rays, computed tomography (CT) scans, and vital signs,
have enabled timely diagnosis and disease monitoring.
Computer systems are also being designed for risk profiling,
patient surveillance, contact tracing, or propagation modeling by using social media data.
Owing to the advancement of computational intelligence,
numerous integrations of computational intelligence mechanisms with various devices and systems have already achieved
considerable success in dealing with the underlying challenges
of epidemic diseases such as new influenzas [1], SARS [2],
and MERS [3]. As a result, many systems and solutions for
combating COVID-19 have adopted computational intelligence, and the design of proper computational intelligence
mechanisms plays a crucial role in building such solutions.
Since the integration of computational intelligence mechanisms with var ious devices and systems under different
application conditions would require different types of computational intelligence techniques, including data analytics,
computational modeling, high-performance computing, artificial intelligence, and in particular its subfield of machine
learning, many researchers have devoted their efforts to developing systems of computational intelligence specifically for the
fight against COVID-19.
By the end of July 2020, more than 200,000 scholarly articles were published regarding COVID-19, SARS-CoV-2, and
other related coronaviruses [4]. However, these articles did not
address in-depth the key issues in applying computational

intelligence to combating the COVID-19 pandemic. Thus, it
would be exhausting to filter and summarize studies related to
computational intelligence from such a large number of articles. In light of the above observations, now is the time to systematically categorize and review the current progress of
research on computational intelligence. Accordingly, this survey
aims to assemble and summarize the highlights of the latest
developments and insights in applying computational intelligence approaches, such as machine learning, evolutionary computation, soft computing, and big data analytics, to practical
applications used to combat COVID-19.
The remainder of this paper is organized as follows. In Section II, we briefly survey the history of computational intelligence. In Sections III through VII, we categorize computational
intelligence into its five principles and determine the urgent
issues concerning COVID-19, which have been, or can be,
resolved using computational intelligence approaches. We then
review the current computational intelligence studies that have
attempted to address these urgent issues based on these five
principles. Then, in Section VIII, we review some current systems or applications for combating COVID-19 that have
employed principles of computational intelligence. Finally, in
Section IX, we conclude the article and discuss recommendations for future studies.
II. Overview

Computational intelligence techniques have already been successfully integrated into various systems for dealing with the
underlying challenges of epidemic diseases. Before we introduce the specific issues which computational intelligence can
be used to solve to fight COVID-19, we should first understand the history and various categories of this method. Based
on the principles of computational intelligence, we can further
clarify what types of issues can be dealt with when battling
COVID-19 with computational intelligence.
A. Brief Introduction to Computational Intelligence

Computational intelligence was formally defined by Bezdek
in 1994 [5] [6] such that a system is called "computationally
intelligent" if the system deals with data on a basic level
(such as pixels of an image), contains a module of pattern
recognition, and does not utilize prior knowledge in the
sense of artificial intelligence. According to Bezdek's definition, computational intelligence is one branch of artificial
intelligence. Actually, the goals of both artificial intelligence
and computational intelligence are the same, which is to
realize general intelligence. Marks [7] clarified the difference
between artificial intelligence and computational intelligence
by claiming that the former is made from hard computing
technologies, whereas the latter is made from soft computing technologies.
Therefore, we can presume that two types of machine intelligence exist: 1) artificial intelligence, which is developed by the
concept of hard-computing and 2) computational intelligence,
which is developed by the concept of soft-computing.

NOVEMBER 2020 | IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE

IEEE Computational Intelligence Magazine - November 2020

Table of Contents for the Digital Edition of IEEE Computational Intelligence Magazine - November 2020