A Comprehensive Survey on the Data-Driven Approaches used for
Tackling the COVID-19 Pandemic
WALID SALAMEH1,*, OLA M. SURAKHI2, MOHAMMAD Y. KHANAFSEH3
1Computer Science Department,
Princess Sumaya University for Technology,
Amman 11941,
JORDAN
2Computer Science Department,
American University of Madaba,
Madaba 11821,
JORDAN
3Computer Science Department,
Birzeit University,
West Bank PO Box 14,
PALESTINE
Abstract: - The current evolution of Artificial Intelligence (AI) is fueled by the massive data sources generated by
the Internet of Things (IoT), social media, and a diverse range of mobile and web applications. Machine learning
(ML) and deep learning become the key to analyzing these data intelligently and developing complementary
intelligent data-driven services in the healthcare sector. The world witnessed many AI-enabled tools that
contributed to fighting against the COVID-19 pandemic and accelerated with unprecedented accuracy the
development and the deployment of many countermeasures. The main objective of this study is to provide a
comprehensive survey on the role of AI and ML methods in the healthcare sector. The study offers cases on how
AI/ML can arm the world against future pandemics. Specifically, the study presents all available datasets, the
main research problems related to COVID-19, and the solutions that AI and ML technologies offer. Finally, based
on the analysis of the current literature, the limitations and open research challenges are highlighted. Our findings
show that AI and ML technologies can play an essential role in COVID-19 forecasting, prediction, diagnosis, and
analysis. In comparison, most of the previous works did not deploy a comprehensive framework that integrates
the ML and DL with network security. This work emphasizes the mandate of including network security in all
COVID-19 applications and providing complete and secure healthcare services.
Key-Words: - Artificial Intelligence, Artificial Neural Network, COVID-19, Data-driven, Diagnosis, Internet of
Things, Machine Learning, Treatment.
Received: August 22, 2023. Revised: February 9, 2024. Accepted: March 6, 2024. Published: April 25, 2024.
1 Introduction
The first appearance of COVID-19 was in Wuhan
City in China at the end of December 2019.
COVID-19 was declared by the World Health
Organization (WHO) to be widespread worldwide
in March 2020, [1]. COVID-19 exponentially
spread worldwide and exceedingly influenced the
healthcare framework in many countries, increasing
the confirmed positive cases and death rate, [2], [3].
Several countries started to force restrictions on the
citizens to overcome the pandemic and stop the
virus from spreading, such as lockdown and
gathering restrictions, school and airport closures,
and more, [4], [5].
Until now, a new generation of COVID-19 virus
appears with no specific medications that can deal
with it precisely. Some of these medications have
been confirmed by the World Health Organization
(WHO), [6], more efforts are still needed to provide
a reliable solution to the COVID-19 Methods. An
orderable solution is required to estimate future
cases, analyze the effect of different features that
help increase/decrease the spreading, and help
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
200
Volume 21, 2024
develop a medical treatment to slow down the
spreading.
Recently, researchers started to utilize Artificial
Intelligence (AI) and Machine Learning (ML)
algorithms to diagnose COVID-19 and investigate
the effect of various health policies on COVID-19
spreading, [7], [8], [9]. The raw historical data for
COVID-19 can be classified into three main groups:
textual information, speech data, and image data.
These data are used widely to diagnose COVID-19,
predict its spreading, vaccine discovery, sentiment
analysis of false news about COVID-19, etc. AI
provides many methods that can help diagnose,
trace, and forecast the spread of the COVID-19
virus based on these data types.
This paper focuses on analyzing how AI and ML
technologies can be employed to provide solutions
and assist in developing public health policies to
mitigate the war on the COVID-19 pandemic from
different perspectives. Studying the COVID-19
pandemic and utilizing AI and ML methods with
the available datasets can bring forth numerous
benefits: 1) Early Detection and Diagnosis. 2) Drug
Discovery and Vaccine Development. 3)
Optimizing Healthcare Resources. 4) Improving
Public Health Policies. And 5) Continuous Learning
and Improvement.
There are other survey and review papers on the
same topic; we analyzed them and differentiated
them from our article. This paper provides a
taxonomy to identify four COVID-19 main research
problems to which ML algorithms and methods are
applied. Based on this taxonomy, we present a
review of the related published papers. Then, we
offer the limitations and challenges that impact this
research area. Finally, we give some suggestions for
improving the performance of using ML methods in
the COVID-19 application management.
The main contributions of this paper can be
summarized as follows:
First, we overview the COVID-19
pandemic and the available datasets. The
dataset is divided into three main categories:
Images, Sound, and Textual dataset.
Then, we listed the main research problems
of the COVID-19 domain where AI and
ML methods can be applied.
A summary of the state-of-the-art works
that utilize ML methods to solve the
COVID-19 pandemic is given.
Finally, we explore the services and
solutions that AI and ML technologies offer.
Along with the challenges and limitations
in the same domain.
2 Related Survey Papers
A few research survey papers have been published
that discussed the application of ML methods in
COVID-19 applications. In this section, we
summarized the recent works and differentiate them
from our paper in three different aspects: 1) the
COVID-19 applications that have been used on the
ML methods, 2) the COVID-19 datasets used for a
published reviewed paper that has been reviewed in
the survey paper and 3) determine whether the
authors highlight the type of COVID-19 features
used for the task of modeling and forecasting in the
published papers as shown in Table 1.
The authors in [10], surveyed the role of AI and
ML in fighting against the COVID-19 pandemic.
The authors discussed five main applications for the
COVID-10 area and analyzed primary datasets in
the same place. However, the effect of health
policies and features for each work has not been
studied. [11], reviewed the research papers that
have been published in Science Direct, Springer,
Hindawi, and MDPI in the area of COVID-19 and
ML methods. The authors highlighted the findings
of each work. In [12], the authors identified the role
of different technologies in tackling COVID-19.
Other authors reviewed Pubmed, Scopus, and
Google databases for the research that applied AI
technology to COVID-19. They suggested that AI
techniques can predict the number of positive cases,
[13]. In some COVID-19 applications, no related
papers were surveyed on that domain. [14], studied
some papers applying AI and ML to predict
COVID-19 spread. [15], presented a review of the
research papers that applied ML to one application
of COVID-19, indicating the number of confirmed
cases. [16], overviewed published articles that used
ML and DL mechanisms for COVID-19 diagnosis.
A similar overview is performed, [17].
The current study stands out from the available
literature in several ways. Firstly, unlike existing
papers, it introduces a taxonomy for systematically
reviewing research papers concerning COVID-19
applications, datasets, and features utilized in
machine learning (ML) mechanisms for modeling
COVID-19 data. While prior literature may touch
upon one or more of these aspects, none
comprehensively categorize and review papers
based on all three dimensions. Moreover, this study
conducts a literature search specifically targeting
relevant studies published in 2020 and 2021, using
key terms such as Machine learning, Deep learning,
Artificial Intelligence, COVID-19, dataset, and
forecasting, and diagnosis. By focusing on recent
research, it ensures relevance and timeliness in its
review. Furthermore, the study goes beyond mere
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
201
Volume 21, 2024
categorization by applying different ML and DL
algorithms and methods to COVID-19 management
across various domain applications. It meticulously
classifies these works based on application types,
providing detailed descriptions of the datasets used
for modeling. Additionally, the study analyzes the
performance results of these papers, emphasizing
the importance of dataset features in influencing
ML efficacy. By synthesizing insights from
application types, datasets, and ML methods, the
study aims to offer recommendations and
suggestions for health sector officials to devise
effective strategies in combatting the pandemic.
This holistic approach distinguishes the current
study as a valuable contribution to the field of
COVID-19 research and ML applications.
Table 1. Related Survey Papers
Survey
Paper
COVID-19 Research Area
COVID-19
Datasets
COVID-19
Modelling
Features
[10]
Diagnosis and Identification,
Forecasting the spread,
Association between
COVID-19 infection and
patient characteristics,
Treatment development,
Supporting applications
√
-
[11]
Diagnosis and Identification
-
-
[12]
Diagnosis and Identification,
Forecasting the spread
-
-
[13]
Early detection and
diagnosis of the infection,
Monitoring the treatment,
Contact tracing, Projection
of cases and mortality,
Development of drugs and
vaccines, Reducing the
workload of healthcare
workers, Prevention of the
disease
-
-
[14]
Outbreak prediction, Virus
spreading, Diagnosis and
treatment, vaccine discovery
-
-
[15]
Forecasting the spread
-
-
[16]
Diagnosis and Identification
√
-
[17]
Forecasting the spread
√
-
[18]
Diagnosis and Identification,
virology and pathogenesis,
drug and vaccine
development, Forecasting
the spread
√
-
[19]
Diagnosis and Identification,
COVID-19 emotional and
sentiment analysis from
social media,
knowledge-based discovery
and semantic analysis from
the collection of scholarly
articles covering COVID-19,
Forecasting the spread
-
-
3 COVID-19 Pandemic
3.1 Taxonomy of COVID-19 Research
Problems of This Research Paper
The COVID-19 pandemic put the world and public
health in a severe and critical challenge. As positive
cases increased exponentially, the COVID-19
pandemic changed human life worldwide and
influenced the economy and citizens’ social life. A
reliable estimation of the number of instances and
feasible strategies to control the pandemic is
urgently needed to provide a potential solution for
the outbreak. AI and ML have been applied recently
in many application areas of COVID-19 to help
make informed decisions by policymakers to
control the pandemic. The research investigation in
this domain can be done under three primary
methodologies as shown in Figure 1.
Tackling any research problem employing
machine learning methodologies follows a
systematic approach. Firstly, researchers delineate
clear research questions pertinent to the domain,
aligning with the objectives of understanding,
diagnosing, treating, or preventing the spread of
COVID-19. Next, they meticulously gather relevant
datasets encompassing images, sound recordings,
textual information, or other pertinent sources.
Subsequently, employing sophisticated analytical
techniques, researchers dissect the data, identifying
patterns, trends, and correlations crucial for
addressing the research questions at hand. Through
feature extraction, key insights are distilled, laying
the groundwork for the subsequent modeling phase.
Finally, employing machine learning algorithms,
these insights are encapsulated into robust models
capable of providing predictive, diagnostic, or
prescriptive solutions, thereby contributing to the
ongoing efforts to combat the COVID-19 pandemic
effectively.
Fig. 1: Taxonomy of COVID-19 Research
Problems of This Paper
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
202
Volume 21, 2024
3.2 COVID-19 Datasets
No country worldwide succeeded in providing
reliable data about COVID-19 information. Most of
the numerous deaths and people with COVID-19
are left unreported. Thus, the lack of accurate data
has become a severe problem in the COVID-19
domain research. The research in this area integrates
the data provided from different resources to create
a valuable dataset that can be used by ML and DL
methods to be applied in the COVID-19 domain
system. In general, three types of COVID-19
datasets are collected from various resources:
images, speech, and text.
Images Dataset
The image dataset comes in three forms: CT scan,
X-ray, and Ultrasound, collected by the screening
tools and devices for the patient with COVID-19.
Most medical images need to be pre-processed
before using them in the modeling process. This
involves two main steps: segmentation and
augmentation. The segmentation process highlights
the infected area (the region of interest) to
differentiate it from the other normal areas. The
augmentation filters the image and transforms it
into another format to increase the size of the
dataset, [20].
The most common way to diagnose patients with
respiratory diseases is by taking a medical image of
the chest. In the case of COVID-19 infection, the
chest area will appear to have a tissue or fluid that
prevents gas exchange or the appearance of a
shadow area in the X-ray image, [21]. Using this
type of dataset can help perform the classification
task on the patient's medical image to determine if
he is infected with COVID-19 or not, or to evaluate
the progress and changes during his infection time.
Examples of the medical images dataset which is
available online are a CT-scan dataset with 125
CT-chest images, [22], an X-ray dataset with 13,800
images for 13,000 patients collected from several
online repositories, [23] and an Ultrasound dataset
with 64 videos, [24].
Speech Dataset
This type of dataset is the least available online due
to the lack of collected data, with a significant
challenge in managing it. This dataset type comes in
three forms: 1) the patient’s cough sound, 2) the
patient's breathing sound, and 3) the patient’s voice.
These data can be used to classify if the patient is
infected with COVID-19 or not, determine his
health status, and trace the patient’s medical
progress, [25]. An online example of the sound
dataset is Coswara, which contains a collection of
recordings of breath and deep coughs of the patients
with more information about the health status and
some geographical information for each one, [26].
Another example of the online sound dataset for
COVID-19 is SACRO, developed from South
Africa through smartphones, [27]. Several studies
showed that a sound dataset helps diagnose and
identify the COVID-19 virus, [28], [29], [30]. A
unique number of features can be extracted from the
sound dataset. These features do not overlap with
the other respiratory infection features. Thus, it can
train complex ML models for accurate diagnosis
and prediction. The WHO reported that coughing is
a symptom for the COVID-19 patient and is
considered the primary reason for spreading the
virus, [31].
Textual Dataset
The textual datasets collected during the pandemic
and found in the literature can be categorized into
different groups. 1) Data about the patients and their
symptoms (statistical data), 2) data about several
reported cases which comes in a time-series format,
3) mobility data related to the transmission of the
citizens, 4) policy data related to the government
policies and rules that have been issued during the
pandemic, and 5) data from the social media which
reflect the semantic of the humans. The most
popular textual dataset reported the number of
people with COVID-19, the number of deaths, the
number of recovered, the positive rate, etc. Almost
every country worldwide has developed a statistical
dataset to report these numbers. For example, John
Hopkins University developed a real-time dataset to
aggregate data about COVID-19 for the researcher
for analyzing and modelling. The dataset is
available online and can be accessed by anyone,
[32]. Research in [33], investigated the effect of
human mobility in China on the spatial distribution
of COVID-19 using the mobility data. Another
dataset developed by Oxford University contains
data from various countries about the rules and
restrictions that the government has applied to
control the virus spreading, [34]. This data includes
essential features such as travel limitation, social
distancing, face masking, cancellation of public
events, etc., which are necessary to show its
influence on the virus transmission. Another dataset
is developed from a tweet of 2500 participants for
the semantic analysis to understand the emotions
and worries toward the COVID-19 pandemic, [35].
A similar dataset was created from Arabic tweets,
where the authors collected about 2,433,660 tweets
from Arab countries to analyze the public response
and behavior toward the pandemic, [36].
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
203
Volume 21, 2024
3.3 The Main Research Problems of
COVID-19
Artificial Intelligence (AI) technology and machine
learning science have been applied in many areas.
The healthcare system is one of the most important
fields where AI technology can provide many
services to enhance its performance. The
COVID-19 pandemic opened horizons for
researchers to apply state-of-the-art technologies to
improve health services and find a solution to stop
the spreading of this pandemic. This section
summarizes the various areas where AI technology
can be applied in fighting against the COVID-19
pandemic.
Diagnosis
The primary diagnosis of patients with COVID-19
methods is 1) molecular diagnostic, which includes
RT-PCR test, and 2) medical images, [37]. The
diagnosis can be performed either in the clinic or
laboratory. These diagnoses result in different kinds
of data such as text, speech, and image that can be
utilized in the AI and ML methods to provide
solutions and services for the healthcare system in
the COVID-19 domain. For example, ML methods
like Convolutional Neural Networks (CNN) can
analyze medical images (CT scan and X-ray) to
extract features and differentiate between
COVID-19 and other diseases. On the other hand,
the cough sound and the patient's breathing data can
detect the COVID-19 symptoms, [38].
These speech datasets include a unique feature
that confirms whether a patient is infected with
COVID-19 or not. The ML methods with the deep
architecture can model speech datasets and provide
an accurate classification for COVID-19-positive
cases. The use of such techniques is urgent to fast
and automate the diagnosis of the COVID-19 virus.
Treatments
The exponential growth in the number of positive
cases of COVID-19 and deaths put countries around
the world in a critical situation. Clinical and
scientists worldwide have been urged to work hard
to search for a vaccine or drugs with efficient
operations. The traditional way of drug
development is a complex process that may need a
lot of time. This contradicts the virus’s rapid spread,
which requires a solution to curb it. AI technology
and ML methods can speed up drug and vaccine
discovery. Some of the operations where AI and
ML technologies can be applied in drug
development are: representing the relationship
between entities such as pair of genes and the
interactions between molecules, discovering new
chemical compounds to identify COVID-19,
predicting the best protein that could serve as an
effective vaccine and more.
Forecasting
Since its appearance in 2019, the number of
confirmed cases of COVID-19 has been increasing
exponentially. It reached 290 million in Jan 2022.
Determining the number of future positive cases is
the primary key to planning against the pandemic.
The statistical methods and ML techniques can
analyze the pandemic status and forecast the growth
of the virus spreading. Some of the exciting
measurements indicated in the COVID-19 domain
include the total number of people with COVID-19,
the infection rate, and the number of confirmed
deaths. This data can help provide a general insight
about the quality of the healthcare system in a
specific country and track the infection-spreading
parameters that cause the increase/decreasing of the
transmission.
Tracing
The AI tools and ML techniques could provide
solutions to avoid virus spread by tracking and
screening its transmission. The use of smart devices
such as mobile phones and sensors can accelerate
the development of a monitor system that tracks
patients through AI applications, [39], [40]. A large
amount of data can be aggregated from these
devices. These data can be modeled by ML methods
and algorithms and classified into different
categories such as mild, urgent, etc. A
decision-making process can be adopted to help
decide whether the patient needs intensive care or
respiratory support, etc. The continuous monitoring
of the patients can help reduce the number of
patients visiting the hospital (for mild cases), which
prevents the healthcare system from failing, [41]. A
recent improvement in the AI application for the
COVID-19 domain is the development of medical
chatbots based on ML and Deep Learning (DL)
models, [42], [43], [44]. The chatbots can assist
patients by continuous answering and guiding in
dealing with the disease’s potential problems.
3.4 Types of Features Included in the
Datasets
Many studies have been proposed to analyze the
association and correlation between the spreading of
COVID-19 disease and other features. The
determination of the future severity of the pandemic
depends on the spreading speed relating to a set of
factors that may accelerate the spreading or not.
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
204
Volume 21, 2024
In this paper, we classify the features into three
groups) The patient characteristics and health status,
2) the environment and meteorological data, and 3)
the mobility and country policy. Analyzing the
correlation between these features and the risks of
the COVID-19 outbreak may help to point out the
main reason for the virus spreading. In this way, an
effective solution could be provided to the
government authorities to control the outbreak and
minimize the load on the healthcare sector.
Patient Characteristics and Health Status
Several studies investigate the correlation between
human characteristics and the risk of being infected
with COVID-19 or not. This includes many features
such as blood type, age, gender, smoking or not,
obesity, health status and historical diseases, and
more, [45], [46], [47], [48], [49]. For example, the
authors in [46], studied the relation between the
blood type and the risk of COVID-19 infection and
found that the blood type of group A has a higher
risk of getting an infection than another blood type.
A higher risk of disease with the patient age is
found by many researchers based on the analysis of
the association between patient age and the number
of positive cases and mortality, [50], [51]. The
results showed that patients above 75 years old are
at a higher risk of fatality. Other studies analyzed
the correlation between COVID-19 infection rate,
death rate, and patient gender. In [52], a higher
death rate in males than females are reported. Also,
the immune response and the infection rate are
different in males and females due to the biological
differences in features, [53]. COVID-19 is a
respiratory disease, and as smoking is a bad habit
that destroys the lungs and weakens the immune
system, the smoking patient becomes at a higher
risk than the non-smoker patient, [54], [55]. The
same results are conducted for the patient's health
status. It was found that patients with other severe
diseases such as diabetes are at a higher risk of
having a respiratory illness like COVID-19, [56].
Environment and meteorological data
Previous studies have proved that meteorological
information is vital in spreading the COVID-19
virus. This includes addressing the relationship
between the weather conditions and the infection
rate. Previous studies showed a correlation between
temperature and the infection rate, [57], [58]. Other
features that influence the spread of the disease are
population size and the country’s location. It has
been shown by researchers in [59] that the
population size is a crucial transmissibility feature
that causes an increase or decrease in the spreading
speed of COVID-19. The higher the population size,
the faster spreading of the virus, [57]. Many other
features related to the environmental country
conditions can influence the spreading of the virus,
such as sea level air pollution rate, [60], [61], [62].
In [61], the authors applied a study on 65 different
countries to investigate the effect of many
environmental features (wind speed, sea pressure,
rainfall, etc.) on spreading the virus. The results
showed a strong correlation between environmental
features and virus spreading.
Country Policy and Mobility
In the pandemic policy, the governments play a
vital role in preventing the virus from spreading
quickly by managing the right policy decisions. It
became a challenge for the authorities to balance the
need to control the quarantine and other aspects like
economic and social. Therefore, most countries
force policies and strategies on several levels in
response to the spreading of the COVID-19
pandemic and to tackle the emerging situations,[63].
For example, some countries force travel
restrictions, school closures, lockdowns (complete
or partial), and more. The effect of these features on
controlling the virus spreading will lead the
authorities in the governments to determine the
most effective interventions in containing the
outbreak, [64].
Many organizations have started to record these
data and provide it publicly on the websites such as
the World Health Organization (WHO), [65], the
World Meter Corona Virus Statistics website, [66],
the Centers for Disease Control, and Prevention
(CDC), [67] and more.
These data can help model the number of
infected cases (People with COVID-19) based on
their correlation with these features. The analysis of
this kind of work will provide a strategy for the
country to be followed to give a practical solution
that can help control the spreading of the virus. The
authors in [64] applied a regression analysis to
identify the main features that affect the cases
(infected and recovered) of COVID-19 in Europe,
the United States, and China. They found that
government actions such as border closures, full
lockdowns, and a high rate of COVID-19 testing
were not associated with statistically significant
reductions in the number of critical cases or overall
mortality. Another study investigated the effect of
social distancing, border restrictions, quarantine,
and isolation on the local transmission of the
COVID-19 virus in Hong Kong. They found that
viral transmission is reduced when forcing such
policies, [68].
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
205
Volume 21, 2024
4 Modelling COVID-19 Pandemic
4.1 Taxonomy of Machine Learning and
Deep Learning in COVID-19 Pandemic
Management
In this section, we summarized the role of the ML
and DL technologies in facing the COVID-19
outbreak. The summary is tabular with a taxonomy
for the recent works that applied ML and DL
methods in COVID-19 management. The taxonomy
divided these works based on the COVID-19
application domain that has been used. It includes
four main sections: 1) Diagnosis and Identification,
2) Treatments Development (Vaccine and Drugs), 3)
Forecasting the Spread, and 4) Patient Tracing.
Table 2 (Appendix) listed the works that applied
ML an DL methods in the Diagnosis and
Identification of COVID-19.
In the Diagnosis section, the ML and DL
methods have been applied to classify the dataset
used (medical images, sound, or textual) for the
disease diagnoses or identify the relationship
between the dataset and COVID-19 symptoms. The
performance of each method is then evaluated based
on statistical measures such as precision, recall,
F-measure, and accuracy that assess the results.
These models and techniques could provide a
solution concerning the decision-making process to
control the outbreak done by government officials.
In the treatment development section, AI
technology and ML algorithms have been applied
widely to develop a drug or vaccine against
COVID-19. Using such methods can reduce the
time and the cost needed to design a sophisticated
development drug pipeline, making them more
effective in identifying a new antiviral drug.
In a recent publication, the authors proposed a
deep-learning model to predict a drug that targets
the SARS-CoV-2-related proteins, [79]. Pham et al.,
used the DeepCE algorithm to predict a treatment
for COVID-19 by repurposing the drug compound,
[80]. In another study, an ML model is used to
predict a new indication about the possible herbal
and drug combinations based on several positive
drug-disease associations, [81]. In general, the AL
technology can provide solutions for tracing the
drug development and proving its effectiveness
against the COVID-19 virus, [82], [83]. A deep
learning-based pipeline model has been developed
to screen the small molecules against the virus, [84],
while other authors predicted antiviral peptides
using ML algorithms, [85].
On the other hand, the vaccine becomes the
best solution to impact the pandemic. While many
companies developed different vaccine components,
the AL technology can be utilized to analyze the
issues related to the efficiency of these vaccine
candidates. This will include the analysis of virus
immunity, the side effects based on the historical
health information, manufacturing and storage, and
more. All of this development will help the
improvement of the vaccine to be more effective
and safer. Some researchers used ML methods to
predict the best protein that could be the most
suitable for an effective vaccine, [86], [87]. In [86],
the authors used the XGBoost model, and in [87],
the authors utilized generative deep learning
models.
However, the development companies
published little information about the
methodologies pipeline of the vaccine process with
minimal information about how they integrate the
ML in the development process.
Forecasting is the most popular application
where ML methods have been applied. Since its
appearance in 2019, people with COVID-19 have
increased exponentially worldwide. So, it becomes
essential to determine the future severity of the
outbreak. This includes the analysis of the pandemic
status using ML and DL methods to extract features
that may lead to the virus spreading. Table 3
(Appendix) summarized some recent works that
applied ML or DL methods and algorithms in the
COVID-19 forecasting spread modeling.
ML methods are applied on mobile devices,
like mobiles, in the tracing application to track the
patients, [39], [40]. These devices can provide many
services, such as patient monitoring, diagnosing,
and screening. Based on the data aggregated by the
smart devices, the AI technologies and ML methods
can be applied to analyze data and provide a helpful
decision-making service such as deciding if the
patient is in urgent need of an Intensive Care Unit
(ICU) admission, [101]. Some of the AI
applications where ML is applied to provide
decision-making services include:
1) Classify the patient status into light,
medium, or severe, [102], [103], [104].
2) Monitoring the patient’s symptoms, [105],
[106], [107].
3) Chatbots provide useful information and
guidance for the patients, [42], [43], [44].
4) Social media sentiment analysis of
population realization towards COVID-19, [108],
[109], [110].
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
206
Volume 21, 2024
4.2 Challenges and Limitations
This section highlights the critical challenges that
can be concluded from analyzing the literary works.
1) Most literary works evaluate their results
based on statistical measurements like accuracy,
F-score, etc. According to this evaluation, the
proposed model is authenticated, and the results are
accepted. Few works show the level of uncertainty.
2) Most of the works explicitly explained the
proposed model’s methodology. Thus, the model is
transparent.
3) Few works used a pre-trained model. The
use of previous models increases generalization and
supports the idea of transfer learning that produces a
robust forecasting model.
4) The type of dataset used in the proposed
model affects the complexity level. As more
features are included in the training model, the
complexity increases. Most Gene, image, and sound
datasets include complicated features that need
more analysis and pre-processing steps.
5) The dataset size used in the model training
influences the forecasting accuracy very much. A
more extensive dataset will result in more accurate
and satisfactory results, [111]. A collaboration
between all medical sectors worldwide is highly
recommended to integrate all data sources and
expand the existing dataset.
6) Using the traditional ML methods results
in a simpler model with satisfactory results. In
contrast, the DL methods are more complicated,
which results in a complex model with multi-hidden
layers that increase training complexity, [112]
7) Different ML methods have different
prediction performances and, therefore, can be used
in other classes of COVID-19 applications.
8) The size of the dataset greatly influences
the ML method’s performance. As the dataset size
increases, the model performance will be improved.
Most of the COVID-19 existing datasets are limited,
and the performance of the ML model cannot be
generalized to include all the pandemic aspects.
9) The hybridization of the ML model
outperforms single forecasting models, [113], [114].
The behavior of the trained data can be learned by
many models in ensemble learning (hybrid
approach), which results in a compatible forecasting
value with the observed ones.
10) Few studies focused on modeling the role
of the vaccine and drug in combating the pandemic.
11) While every region has its specific
weather conditions, a few works include
geographical location and weather conditions in the
virus spreading.
12) Different features influenced the
spreading of the virus. Few of these works analyzed
the correlation between the COVID-19 spread and
infection and the other features. Most of the
correlation analysis included the confirmed, dead,
and recovered cases.
13) The number of works conducted in the
European countries was found to be the most than
that in the other countries.
14) Security and privacy issues are critical
when dealing with the healthcare system, [115],
[116]. Integrating AL technologies and security
applications such as Blockchain technology is
recommended to maintain patient privacy while
providing a high-level service using AI and ML
methods.
15) DL methods and algorithms are complex
and need a high computational resource to model,
process, and work with big data. Integrating Fog
computing and Edge technology is recommended to
handle this challenge.
16) In general, ML and DL methods are
black-box models, [117]. The interpretation of
model behavior regarding the choice of features and
generated accuracy is essential to explain it to
medical experts and decision-makers.
17) Most ML and DL proposed in the
literature show promising performance and good
results in the forecasting, prediction, diagnosis, and
analysis. In reality, most of these models do not
deploy. A comprehensive framework that integrates
the ML and DL models with network security is
required to include all the COVID-19 applications
and provide a complete healthcare service.
5 Conclusions and Future Works
In this paper, we have conducted a comprehensive
survey about the application of AI technology and
ML methods for intelligent data analysis and
applications to tackle the COVID-19 pandemic. We
have discussed and analyzed how ML methods can
be used to provide a solution to mitigate the impact
of the pandemic. It was found that AI
solutions-based ML algorithms can be used for
diagnosis, treatment, forecasting, and monitoring
applications. While the ML algorithm needs to be
trained through solid knowledge related to
application data, this paper also surveyed the
real-time datasets that are provided as open-source
for the COVID-19 domain. ML algorithms require
training on relevant data to learn patterns and make
predictions. In the context of the COVID-19
pandemic, this means that algorithms need to be
trained on datasets containing information about the
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
207
Volume 21, 2024
virus, its transmission, symptoms, diagnosis,
treatment, and other related factors. Solid
knowledge about these aspects is crucial for
effective algorithm training. In addition to
discussing the training of ML algorithms, the paper
also explores real-time datasets that are openly
accessible in the COVID-19 domain. These datasets
provide up-to-date information about various
aspects of the pandemic, such as infection rates,
hospitalization numbers, testing data, genomic
sequences of the virus, and vaccination statistics. By
surveying real-time datasets, the paper ensures that
the ML algorithms discussed in the study are not
only trained on relevant historical data but also have
access to the latest information about the pandemic.
This allows for more accurate and timely
predictions, analyses, and decision-making
processes. Moreover, open-access datasets promote
transparency, collaboration, and reproducibility in
research, enabling other researchers and
stakeholders to verify findings and build upon them.
Finally, we have discussed the limitations and
challenges of the current AI solutions-based ML
algorithms. These challenges create a research path
for the future applications of ML methods in real
COVID-19 environments.
In envisioning future research directions, the
integration of emerging technologies such as
blockchain holds promise in devising
comprehensive frameworks to address the
COVID-19 pandemic while safeguarding privacy
and security concerns. Blockchain, renowned for its
decentralized and immutable nature, offers a
potential solution to the challenge of securely
managing and sharing sensitive health data amidst
the pandemic. By leveraging blockchain technology,
researchers can design frameworks that ensure the
integrity and confidentiality of COVID-19-related
data, while facilitating seamless data exchange
among healthcare providers, researchers, and public
health authorities. This can enable more effective
contact tracing, monitoring of infection spread, and
vaccine distribution efforts, all while preserving
individual privacy rights.
Furthermore, to enhance the performance of
prediction models, future research endeavors could
focus on the inclusion of additional features from
diverse domains, such as meteorological data and
medical records. Incorporating meteorological data,
including temperature, humidity, and air quality
indices, into predictive models can provide valuable
insights into the environmental factors influencing
virus transmission dynamics. Similarly, integrating
comprehensive medical data, including patient
demographics, comorbidities, and treatment
histories, can enrich predictive models by capturing
a more nuanced understanding of individual
susceptibility to COVID-19 and disease progression
patterns. By harnessing these multidimensional
datasets, researchers can refine prediction models to
yield more precise and accurate results, ultimately
empowering healthcare professionals and
policymakers with actionable insights for mitigating
the impact of the pandemic.
Acknowledgement:
This work has been conducted during the Sabbatical
year of Prof. Walid A. Salameh, thus, we would like
to thank and acknowledge PSUT for their
continuous support.
References:
[1] Arden, Madelynne A., and Joseph Chilcot.
“Health Psychology and the Coronavirus
(COVID-19) Global Pandemic: A Call for
Research.” British Journal of Health
Psychology, vol. 25, no. 2, Mar. 2020, pp.
231–32.
[2] Manderson, Lenore, and Susan Levine.
“COVID-19, Risk, Fear, and Fall-Out.”
Medical Anthropology, vol. 39, no. 5, Mar.
2020, pp. 367–70.
[3] Sornette, Didier, et al. “Interpreting,
Analysing and Modelling COVID-19
Mortality Data.” Nonlinear Dynamics, vol.
101, no. 3, Aug. 2020, pp. 1751–76.
[4] A. Sandford, “Coronavirus: Half of
humanity now on lockdown as 90 countries
call for confinement,” Euronews, vol. 3, no.
04, p. 2020, 2020.
[5] Shakil, M. H., Munim, Z. H., Tasnia, M., &
Sarowar, S. (2020). COVID-19 and the
environment: A critical review and research
agenda. Science of the Total Environment,
vol. 745, pp.141022.
[6] Ndwandwe, Duduzile, and Charles S.
Wiysonge. “COVID-19 Vaccines.” Current
Opinion in Immunology, vol. 71, Aug. 2021,
pp. 111–16
[7] Alimadadi, Ahmad, et al. “Artificial
Intelligence and Machine Learning to Fight
COVID-19.” Physiological Genomics, vol.
52, no. 4, Apr. 2020, pp. 200–02.
[8] S. Lalmuanawma, J. Hussain, and L.
Chhakchhuak, “Applications of machine
learning and artificial intelligence for
covid-19 (sars-cov-2) pandemic: A review,”
Chaos, Solitons & Fractals, vol. 139, p.
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
208
Volume 21, 2024
110059, 2020
[9] M. van der Schaar, A. M. Alaa, A. Floto, A.
Gimson, S. Scholtes, Wood, E. McKinney,
D. Jarrett, P. Lio, and A. Ercole, “How
artificial intelligence and machine learning
can help healthcare systems respond to
covid-19,” Machine Learning, vol. 110, no.
1, pp. 1–14, 2021
[10] N. El-Rashidy, S. Abdelrazik, T. Abuhmed,
E. Amer, F. Ali, J.-W. Hu, and S.
El-Sappagh, “Comprehensive survey of
using machine learning in the covid-19
pandemic,” Diagnostics, vol. 11, no. 7, p.
1155, 2021.
[11] Kwekha-Rashid, Ameer Sardar, et al.
“Coronavirus Disease (COVID-19) Cases
Analysis Using Machine-Learning
Applications.” Applied Nanoscience, vol.
13, no. 3, May 2021, pp. 2013–25
[12] O. E. Mbunge, B. Akinnuwesi, S. G.
Fashoto, A. S. Metfula, and P. Mash-wama,
“A critical review of emerging technologies
for tackling covid-19 pandemic,” Human
behavior and emerging technologies, vol. 3,
no. 1, pp. 25–39, 2021
[13] R. Vaishya, M. Javaid, I. H. Khan, and A.
Haleem, “Artificial intelli-gence (ai)
applications for covid-19 pandemic,”
Diabetes & Metabolic Syndrome: Clinical
Research & Reviews, vol. 14, no. 4, pp.
337–339, 2020
[14] Q.-V. Pham, D. C. Nguyen, T. Huynh-The,
W.-J. Hwang, and P. N. Pathirana,
“Artificial intelligence (ai) and big data for
coronavirus (covid-19) pandemic: A survey
on the state-of-the-arts,” IEEE access, vol. 8,
p. 130820, 2020
[15] A. Ahmad, S. Garhwal, S. K. Ray, G.
Kumar, S. J. Malebary, and M. Barukab,
“The number of confirmed cases of
covid-19 by using machine learning:
Methods and challenges,” Archives of
Computational Methods in Engineering, vol.
28, no. 4, pp. 2645–2653, 2021
[16] Z. A. A. Alyasseri, M. A. Al-Betar, I. A.
Doush, M. A. Awadallah, K. Abasi, S. N.
Makhadmeh, O. A. Alomari, K. H.
Abdulkareem, Adam, R. Damasevicius et
al., “Review on covid-19 diagnosis models
based on machine learning and deep
learning approaches,” Expert systems, vol.
39, no. 3, p. e12759, 2022
[17] M. R. H. Mondal, S. Bharati, and P. Podder,
“Diagnosis of covid-19 using machine
learning and deep learning: A review,”
Current Medical Imaging, vol. 17, no. 12,
pp. 1403–1418, 2021.
[18] J. Chen, K. Li, Z. Zhang, K. Li, and P. S. Yu,
“A survey on applications of artificial
intelligence in fighting against covid-19,”
ACM Computing Surveys (CSUR), vol. 54,
no. 8, pp. 1–32, 2021
[19] J. Shuja, E. Alanazi, W. Alasmary, and A.
Alashaikh, “Covid-19 open source data sets:
a comprehensive survey,” Applied
Intelligence, vol. 51, no. 3, pp. 1296–1325,
2021.
[20] J. Zhao, Y. Zhang, X. He, and P. Xie,
“Covid-ct-dataset: a ct scan dataset about
covid-19,” arXiv preprint
arXiv:2003.13865, vol. 490, 2020, [Online].
https://github.com/UCSD-AI4H/COVID-C
T (Accessed Date: June 17, 2020).
[21] X. Li, W. Zeng, X. Li, H. Chen, L. Shi, X. Li,
H. Xiang, Y. Cao, Chen, C. Liu et al., “Ct
imaging changes of corona virus disease
2019 (covid-19): a multi-center study in
southwest china,” Journal of translational
medicine, vol. 18, no. 1, pp. 1–8, 2020.
[22] J. P. Cohen, P. Morrison, and L. Dao,
“Covid-19 image data collection,” arXiv
preprint arXiv:2003.11597, 2020, [Online].
https://github.com/ieee8023/covid-chestxra
y-dataset (Accessed Date: March 25, 2020).
[23] L. Wang, Z. Q. Lin, and A. Wong,
“Covid-net: A tailored deep convolutional
neural network design for detection of
covid-19 cases from chest x-ray images,”
Scientific Reports, vol. 10, no. 1, pp. 1–12,
2020.
[24] C. McDermott, M. Łacki, B. Sainsbury, J.
Henry, M. Filippov, and C. Rossa,
Sonographic diagnosis of covid-19: A
review of image processing for lung
ultrasound, Frontiers in big Data, vol. 4,
Mar. p. 2, 2021
[25] J. Han, K. Qian, M. Song, Z. Yang, Z. Ren,
S. Liu, J. Liu, H. Zheng, Ji, T. Koike et al.,
“An early study on intelligent analysis of
speech under covid-19: Severity, sleep
quality, fatigue, and anxiety,” arXiv preprint
arXiv:2005.00096, 2020, [Online].
https://arxiv.org/abs/2005.00096 (Accessed
Date: May 14, 2020).
[26] N. Sharma, P. Krishnan, R. Kumar, S.
Ramoji, S. R. Chetupalli, K. Ghosh, S.
Ganapathy et al., “Coswara–a database of
breathing cough, and voice sounds for
covid-19 diagnosis,” arXiv preprint
arXiv:2005.10548, 2020, [Online].
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
209
Volume 21, 2024
https://arxiv.org/abs/2005.10548 (Accessed
Date: August 11, 2020).
[27] M. Pahar, M. Klopper, R. Warren, and T.
Niesler, “Covid-19 cough classification
using machine learning and global
smartphone record-ings,” Computers in
Biology and Medicine, vol. 135, p. 104572,
2021.
[28] A. A. Farid, G. I. Selim, and H. A. A. Khater,
“A novel approach of ct images feature
analysis and prediction to screen for corona
virus disease (covid-19)”, International
Journal of Scientific & Engineering
Research, 11(3), 1141–49 2020, 2020.
[29] A. Bernheim, X. Mei, M. Huang, Y. Yang, Z.
A. Fayad, N. Zhang, Diao, B. Lin, X. Zhu,
K. Li et al., “Chest ct findings in
coronavirus disease-19 (covid-19):
relationship to duration of infection,”
Radiology, vol. 295, no. 3, June 2020, p.
200463
[30] X. Mei, H.-C. Lee, K.-y. Diao, M. Huang, B.
Lin, C. Liu, Z. Xie, Ma, P. M. Robson, M.
Chung et al., “Artificial
intelligence–enabled rapid diagnosis of
patients with covid-19,” Nature medicine,
vol. 26, no. 8, pp. 1224–1228, 2020.
[31] World Health Organization, “Coronavirus
disease 2019 (covid-19): situation report,
73,” World Health Organization, 2020,
[Online].
https://www.who.int/publications/m/item/si
tuation-report---73 (Accessed Date: April
24, 2024).
[32] E. Dong, H. Du, and L. Gardner, “An
interactive web-based dashboard to track
covid-19 in real time,” The Lancet
infectious diseases, vol. 20, no. 5, pp.
533–534, 2020.
[33] M. U. Kraemer, C.-H. Yang, B. Gutierrez,
C.-H. Wu, B. Klein, D. M. Pigott, O. C.-. D.
W. Group†, L. du Plessis, N. R. Faria, R. Li
et al., “The effect of human mobility and
control measures on the covid-19 epidemic
in china,” Science, vol. 368, no. 6490, pp.
493–497, 2020.
[34] T. Hale, N. Angrist, B. Kira, A. Petherick, T.
Phillips, and S. Webster, “Variation in
government responses to covid-19,” 2020,
medRxiv 2020.07.04.20145334,
https://doi.org/10.1101/2020.07.04.201453
34.
[35] B. Kleinberg, I. van der Vegt, and M. Mozes,
“Measuring emo-tions in the covid-19 real
world worry dataset,” arXiv preprint
arXiv:2004.04225, 2020, [Online].
https://arxiv.org/abs/2004.04225 (Accessed
Date: May 14, 2020).
[36] S. Alqurashi, A. Alhindi, and E. Alanazi,
“Large arabic twitter dataset on covid-19,”
arXiv preprint arXiv:2004.04315, 2020,
[Online]. https://arxiv.org/abs/2004.04315
(Accessed Date: April 22, 2020).
[37] S. Alpdagtas, E. Ilhan, E. Uysal, M. Sengor,
C. B. Ustundag, and Gunduz, “Evaluation
of current diagnostic methods for covid-19,”
APL bioengineering, vol. 4, no. 4, p. 041506,
2020
[38] K. Zhang, X. Liu, J. Shen, Z. Li, Y. Sang, X.
Wu, Y. Zha, W. Liang, Wang, K. Wang et al.,
“Clinically applicable ai system for
accu-rate diagnosis, quantitative
measurements, and prognosis of covid-19
pneumonia using computed tomography,”
Cell, vol. 181, no. 6, pp. 1423–1433, 2020.
[39] N. El-Rashidy, S. El-Sappagh, S. Islam, H.
M El-Bakry, and S. Abdel-razek, “Mobile
health in remote patient monitoring for
chronic diseases: Principles, trends, and
challenges,” Diagnostics, vol. 11, no. 4, p.
607, 2021
[40] A. Rahman, M. S. Hossain, N. A. Alrajeh,
and F. Alsolami, “Adver-sarial
examples—security threats to covid-19
deep learning systems in medical iot
devices,” IEEE Internet of Things Journal,
vol. 8, no. 12, pp. 9603–9610, 2020.
[41] G. Yang, H. Zhang, D. Firmin, and S. Li,
“Recent advances in artificial intelligence
for cardiac imaging,” Computerized
Medical Imaging and Graphics vol. 90,
June 2021, p. 101928
[42] A. Gagneux-Brunon, M. Detoc, S.
Bruel, B. Tardy, O. Rozaire, Frappe,
and E. Botelho-Nevers, “Intention to get
vaccinations against covid-19 in french
healthcare workers during the first
pandemic wave: a cross-sectional survey,”
Journal of Hospital Infection, vol. 108, pp.
168–173, 2021.
[43] M. Alhasan and M. Hasaneen, “Digital
imaging, technologies and artifi-cial
intelligence applications during covid-19
pandemic,” Computerized Medical Imaging
and Graphics, vol. 91, p. 101933, 2021.
[44] Z. Zeng, P.-J. Chen, and A. A. Lew, “From
high-touch to high-tech: Covid-19 drives
robotics adoption,” Tourism geographies,
vol. 22, no. 3, pp. 724–734, 2020.
[45] C. A. Latz, C. DeCarlo, L. Boitano, C. Png,
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
210
Volume 21, 2024
R. Patell, M. F. Conrad, Eagleton, and A.
Dua, “Blood type and outcomes in patients
with covid-19,” Annals of hematology, vol.
99, no. 9, pp. 2113–2118, 2020
[46] J. Zhao, Y. Yang, H. Huang, D. Li, D. Gu, X.
Lu, Z. Zhang, L. Liu, Liu, Y. Liu et al.,
“Relationship between the abo blood group
and the coronavirus disease 2019 (covid-19)
susceptibility,” Clinical Infectious Diseases,
vol. 73, no. 2, pp. 328–331, 2021
[47] A. Algunmeeyn, F. El-Dahiyat, M. M.
Altakhineh, M. Azab, and Z.-U.- Babar,
“Understanding the factors influencing
healthcare providers’ burnout during the
outbreak of covid-19 in jordanian hospitals,”
Journal of pharmaceutical policy and
practice, vol. 13, no. 1, pp. 1–8, 2020.
[48] A. S. Abdelhafiz, A. Ali, H. H. Ziady, A. M.
Maaly, M. Alorabi, and E. A. Sultan,
“Prevalence, associated factors, and
consequences of burnout among egyptian
physicians during covid-19 pandemic,”
Frontiers in public health, vol. 8:590190.
Dec. 2020
[49] F. Bekele, T. Sheleme, G. Fekadu, and K.
Bekele, “Patterns and associated factors of
covid-19 knowledge, attitude, and practice
among general population and health care
workers: A systematic review,” SAGE open
medicine, vol. 8, p. 2050312120970721,
2020
[50] C. Hoffmann and E. Wolf, “Older age
groups and country-specific case fatality
rates of covid-19 in europe, usa and canada,”
Infection, vol. 49, no. 1, pp. 111–116, 2021.
[51] N. Sudharsanan, O. Didzun, T.
Barnighausen,¨ and P. Geldsetzer, “The
contribution of the age distribution of cases
to covid-19 case fatality across countries: a
nine-country demographic study,” Annals of
internal medicine, vol. 173, no. 9, pp.
714–720, 2020.
[52] Capuano, Annalisa, et al. “Covid-19 Kills
More Men Than Women: An Overview of
Possible Reasons.” Frontiers in
Cardiovascular Medicine, vol. 7, July 2020.
Crossref,
https://doi.org/10.3389/fcvm.2020.00131.
[53] C. P. Tadiri, T. Gisinger, A. Kautzky-Willer,
K. Kublickiene, M. T. Herrero, V. Raparelli,
L. Pilote, and C. M. Norris, “The influence
of sex and gender domains on covid-19
cases and mortality,” Cmaj, vol. 192, no. 36,
pp. E1041–E1045, 2020.
[54] R. N. van Zyl-Smit, G. Richards, and F. T.
Leone, “Tobacco smoking and covid-19
infection,” The Lancet Respiratory
Medicine, vol. 8, no. 7, 664–665, 2020
[55] A. Gulsen,¨ B. A. Yigitbas, B. Uslu, D.
Dromann,¨ and O. Kilinc, “The effect of
smoking on covid-19 symptom severity:
systematic review and meta-analysis,”
Pulmonary medicine, vol. 2020, Sept. 2020,
pp. 1–11.
[56] P. Wang, J. Sha, M. Meng, C. Wang, Q. Yao,
Z. Zhang, W. Sun, X. Wang, G. Qie, X. Bai
et al., “Risk factors for severe covid-19 in
middle-aged patients without comorbidities:
a multicentre retrospective study,” Journal
of Translational Medicine, vol. 18, no. 1, pp.
1–12, 2020.
[57] K. Aabed and M. M. Lashin, “An analytical
study of the factors that influence covid-19
spread,” Saudi Journal of Biological
Sciences, vol. 28, no. 2, pp. 1177–1195,
2021
[58] L. Y. K. Nakada and R. C. Urban,
“Covid-19 pandemic: environmental and
social factors influencing the spread of
sars-cov-2 in sao˜ paulo, brazil,”
Environmental Science and Pollution
Research, vol. 28, no. 30, 40 322–40 328,
2021.
[59] X. Zhang, H. Liu, H. Tang, M. Zhang, X.
Yuan, and X. Shen, “The effect of
population size for pathogen transmission
on prediction of covid-19 spread,” Scientific
reports, vol. 11, no. 1, pp. 1–12, 2021
[60] X. Xing, Y. Xiong, R. Yang, R. Wang, W.
Wang, H. Kan, T. Lu, D. Li, J. Cao, J.
Penuelas˜ et al., “Predicting the effect of
confinement on the covid-19 spread using
machine learning enriched with satellite air
pollution observations,” Proceedings of the
National Academy of Sciences, vol. 118, no.
33, 2021.
[61] S. Lin, Y. Fu, X. Jia, S. Ding, Y. Wu, and Z.
Huang, “Discovering correlations between
the covid-19 epidemic spread and climate,”
Inter-national Journal of Environmental
Research and Public Health, vol. 17, no. 21,
p. 7958, 2020.
[62] S. Wang, Y. Zhang, J. Ma, S. Zhu, J. Shen, P.
Wang, and H. Zhang, “Responses of decline
in air pollution and recovery associated with
covid-19 lockdown in the pearl river delta,”
Science of the Total Environment, vol. 756,
p. 143868, 2021
[63] A. Dewi, A. Nurmandi, E. Rochmawati, E.
P. Purnomo, M. D. Rizqi, A. Azzahra, S.
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
211
Volume 21, 2024
Benedictos, W. Suardi, and D. T. K. Dewi,
“Global pol-icy responses to the covid-19
pandemic: proportionate adaptation and
policy experimentation: a study of country
policy response variation to the covid-19
pandemic,” Health Promotion Perspectives,
vol. 10, no. 4,359, 2020
[64] R. Chaudhry, G. Dranitsaris, T. Mubashir, J.
Bartoszko, and S. Riazi, “A country level
analysis measuring the impact of
government actions, country preparedness
and socioeconomic factors on covid-19
mortality and related health outcomes,”
EClinicalMedicine, vol. 25, p. 100464,
2020
[65] WHO, “Coronavirus disease (COVID-19)
pandemic ,” Geneva: WHO, 2020, [Online].
https://www.who.int/emergencies/diseases/
novel-coronavirus-2019 (Accessed Date:
April 22, 2024).
[66] D. Worldometer, “Covid-19 coronavirus
pandemic,” World Health Organization,
2020, [Online].
https://www.worldometers.info/coronavirus
(Accessed Date: April 13, 2024).
[67] A. W. CDC, “Centers for disease control
and prevention,” 2020, [Online].
https://www.cdc.gov/coronavirus/2019-nco
v/ (Accessed Date: April 22, 2024).
[68] B. J. Cowling, S. T. Ali, T. W. Ng, T. K.
Tsang, J. C. Li, M. W. Fong, Q. Liao, M. Y.
Kwan, S. L. Lee, S. S. Chiu et al., “Impact
assessment of non-pharmaceutical
interventions against coronavirus disease
2019 and influenza in hong kong: an
observational study,” The Lancet Public
Health, vol. 5, no. 5, pp. e279–e288, 2020
[69] F. Gabbay, S. Bar-Lev, O. Montano, and N.
Hadad, “A lime-based explainable machine
learning model for predicting the severity
level of covid-19 diagnosed patients,”
Applied Sciences, vol. 11, no. 21, p. 10417,
2021
[70] S. S. Aljameel, I. U. Khan, N. Aslam, M.
Aljabri, and E. S. Alsulmi, “Machine
learning-based model to predict the disease
severity and outcome in covid-19 patients,”
Scientific programming, vol. 2021, Apr.
2021, pp. 1–10 .
[71] L. Li, L. Qin, Z. Xu, Y. Yin, X. Wang, B.
Kong, J. Bai, Y. Lu, Z. Fang, Song et al.,
“Using artificial intelligence to detect
covid-19 and community-acquired
pneumonia based on pulmonary ct:
evaluation of the diagnostic accuracy,”
Radiology, vol. 296, no. 2, pp. E65–E71,
2020
[72] J. Pu, J. Leader, A. Bandos, J. Shi, P. Du, J.
Yu, B. Yang, S. Ke,Guo, J. B. Field et al.,
“Any unique image biomarkers associated
with covid-19?” European radiology, vol.
30, no. 11, pp. 6221–6227, 2020.
[73] X. Deng, H. Shao, L. Shi, X. Wang, and T.
Xie, “A classification– detection approach
of covid-19 based on chest x-ray and ct by
using keras pre-trained deep learning
models,” Computer Modeling in
Engineering & Sciences, vol. 125, no. 2, pp.
579–596, 2020
[74] M. Aly, K. H. Rahouma, and S. M. Ramzy,
“Pay attention to the speech: Covid-19
diagnosis using machine learning and
crowdsourced respiratory and speech
recordings,” Alexandria Engineering
Journal, vol. 61, no. 5, pp. 3487–3500,
2022
[75] K. K. Lella and A. Pja, “Automatic
diagnosis of covid-19 disease using deep
convolutional neural network with
multi-feature channel from respiratory
sound data: cough, voice, and breath,”
Alexandria Engineering Journal, vol. 61,
no. 2, pp. 1319–1334, 2022
[76] D. Arias-Garzon,´ J. A. Alzate-Grisales, S.
Orozco-Arias, H. B. Arteaga-Arteaga, M. A.
Bravo-Ortiz, A. Mora-Rubio, J. M.
Saborit-Torres, J. A. M. Serrano, M. de la
Iglesia Vaya,´ O. Cardona-Morales et al.,
“Covid-19 detection in x-ray images using
convolutional neural net-works,” Machine
Learning with Applications, vol. 6, p.
100138, 2021
[77] L. A. Amar, A. A. Taha, and M. Y.
Mohamed, “Prediction of the final size for
covid-19 epidemic using machine learning:
A case study of egypt,” Infectious Disease
Modelling, vol. 5, pp. 622–634, 2020.
[78] H. Burdick, C. Lam, S. Mataraso, A.
Siefkas, G. Braden, R. P. Dellinger, A.
McCoy, J.-L. Vincent, A. Green-Saxena, G.
Barnes et al., “Prediction of respiratory
decompensation in covid-19 patients using
machine learning: The ready trial,”
Computers in biology and medicine, vol.
124, p. 103949, 2020
[79] B. R. Beck, B. Shin, Y. Choi, S. Park, and K.
Kang, “Predicting commercially available
antiviral drugs that may act on the novel
coronavirus (sars-cov-2) through a
drug-target interaction deep learning model,”
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
212
Volume 21, 2024
Computational and Structural
Biotechnology Journal, vol. 18, pp.
784–790, 2020.
[80] T.-H. Pham, Y. Qiu, J. Zeng, L. Xie, and P.
Zhang, “A deep learning framework for
high-throughput mechanism-driven
phenotype com-pound screening and its
application to covid-19 drug repurposing,”
Nature machine intelligence, vol. 3, no. 3,
pp. 247–257, 2021
[81] E. Kim, A.-s. Choi, and H. Nam, “Drug
repositioning of herbal compounds via a
machine-learning approach,” BMC
Bioinformatics, vol. 20, no. 10, pp. 33–43,
2019
[82] R. Chakraborty and S. Parvez, “Covid-19:
An overview of the current pharmacological
interventions, vaccines, and clinical trials,”
Biochemical Pharmacology, vol. 180, p.
114184, 2020
[83] Y. Zhu, J. Li, and Z. Pang, “Recent insights
for the emerging covid-19: Drug discovery,
therapeutic options and vaccine
development,” Asian Journal of
Pharmaceutical Sciences, vol. 16, no. 1, pp.
4–23, 2021.
[84] H. Zhang, K. M. Saravanan, Y. Yang, M.
Hossain, J. Li, X. Ren, Y. Pan, Y. Wei et al.,
“Deep learning based drug screening for
novel coronavirus 2019-ncov,”
Interdisciplinary Sciences: Computational
Life Sciences, vol. 12, no. 3, pp. 368–376,
2020.
[85] Z. Xu, L. Yang, X. Zhang, Q. Zhang, Z.
Yang, Y. Liu, S. Wei, and W. Liu,
“Discovery of potential flavonoid inhibitors
against covid-19 3cl proteinase based on
virtual screening strategy,” Frontiers in
molecular biosciences, p. 247, vol. 7, Sept.
2020.
[86] A. Kumar, P. K. Gupta, and A. Srivastava,
“A review of modern technologies for
tackling covid-19 pandemic,” Diabetes &
Metabolic Syndrome: Clinical Research &
Reviews, vol. 14, no. 4, pp. 569–573, 2020.
[87] Amilpur, S., & Bhukya, R. (2022).
Predicting novel drug candidates against
Covid-19 using generative deep neural
networks. Journal of Molecular Graphics
and Modelling, vol. 110, pp. 108045.
[88] F. Rustam, A. A. Reshi, A. Mehmood, S.
Ullah, B.-W. On, W. Aslam, and G. S. Choi,
“Covid-19 future forecasting using
supervised machine learning models,” IEEE
access, vol. 8, pp. 101 489–101 499, 2020
[89] J. CSSE, “Covid-19 data repository by the
center for systems science and engineering
(csse) at johns hopkins university,” Github
repository, 2020, [Online].
https://covidbehaviors.org/ (Accessed Date:
April 22, 2024).
[90] E. A. Rashed and A. Hirata, “One-year
lesson: Machine learning prediction of
covid-19 positive cases with meteorological
data and mobility estimate in Japan,”
International Journal of Environmental
Research and Public Health, vol. 18, no. 11,
p. 5736, 2021.
[91] A. I. Shahin and S. Almotairi, “A deep
learning bilstm encoding-decoding model
for covid-19 pandemic spread forecasting,”
Fractal and Fractional, vol. 5, no. 4, p. 175,
2021.
[92] R. G. da Silva, M. H. D. M. Ribeiro, V. C.
Mariani, and L. dos San-tos Coelho,
“Forecasting brazilian and american
covid-19 cases based on artificial
intelligence coupled with climatic
exogenous variables,” Chaos, Solitons &
Fractals, vol. 139, p. 110027, 2020
[93] S. A. Alanazi, M. Kamruzzaman, M.
Alruwaili, N. Alshammari, S. A. Alqahtani,
and A. Karime, “Measuring and preventing
covid-19 using the sir model and machine
learning in smart health care,” Journal of
healthcare engineering, vol. 2020, Nov.
2020, pp. 1–12.
[94] A. Roy and B. Kar, “Characterizing the
spread of covid-19 from human mobility
patterns and sociodemographic indicators,”
in Proceedings of the 3rd ACM
SIGSPATIAL international workshop on
advances in resilient and intelligent cities,
2020, pp. 39–48.
[95] G. Pinter, I. Felde, A. Mosavi, P. Ghamisi,
and R. Gloaguen, “Covid-19 pandemic
prediction for hungary; a hybrid machine
learning approach,” Mathematics, vol. 8, no.
6, p. 890, 2020
[96] G. Pandey, P. Chaudhary, R. Gupta, and S.
Pal, “Seir and regression model based
covid-19 outbreak predictions in india,”
arXiv preprint arXiv:2004.00958, 2020,
[Online]. https://arxiv.org/abs/2004.00958.
(Accessed Date: April 1, 2020).
[97] R. Sujath, J. M. Chatterjee, and A. E.
Hassanien, “A machine learning forecasting
model for covid-19 pandemic in india,”
Stochastic Envi-ronmental Research and
Risk Assessment, vol. 34, no. 7, pp. 959–972,
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
213
Volume 21, 2024
2020.
[98] P. Wang, X. Zheng, J. Li, and B. Zhu,
“Prediction of epidemic trends in covid-19
with logistic model and machine learning
technics,” Chaos, Solitons & Fractals, vol.
139, p. 110058, 2020
[99] S. Dutta and S. K. Bandyopadhyay,
“Machine learning approach for
confirmation of covid-19 cases: Positive,
negative, death and release,” Iberoamerican
Journal of Medicine, vol. 2, no. 3, pp.
172–177, 2020
[100] A. Tob´ıas, “Evaluation of the lockdowns
for the sars-cov-2 epidemic in italy and
spain after one month follow up,” Science of
the Total Environment, vol. 725, p. 138539,
2020
[101] N. El-Rashidy, S. El-Sappagh, S. Islam, H.
M. El-Bakry, and S. Abdel-razek,
“End-to-end deep learning framework for
coronavirus (covid-19) detection and
monitoring,” Electronics, vol. 9, no. 9, p.
1439, 2020.
[102] A. Shelke, M. Inamdar, V. Shah, A. Tiwari,
A. Hussain, T. Chafekar, and N. Mehendale,
“Chest x-ray classification using deep
learning for automated covid-19 screening,”
SN computer science, vol. 2, no. 4, pp. 1–9,
2021
[103] T. A. Soomro, L. Zheng, A. J. Afifi, A. Ali,
M. Yin, and J. Gao, “Artificial intelligence
(ai) for medical imaging to combat
coronavirus disease (covid-19): A detailed
review with direction for future re-search,”
Artificial Intelligence Review, pp. 1–31,
2021
[104] S. Rahmatizadeh, S. Valizadeh-Haghi, and
A. Dabbagh, “The role of artificial
intelligence in management of critical
covid-19 patients,” Journal of Cellular and
Molecular Anesthesia, vol. 5, no. 1, pp.
16–22, 2020
[105] M. Z. Alom, M. Rahman, M. S. Nasrin, T.
M. Taha, and V. K. Asari, “Covid mtnet:
Covid-19 detection with multi-task deep
learning approaches,” arXiv preprint
arXiv:2004.03747, 2020, [Online].
https://arxiv.org/abs/2004.03747 (Accessed
Date: April 18, 2020).
[106] M. A. Ibrahim and A. Al-Najafi, “Modeling,
control, and prediction of the spread of
covid-19 using compartmental, logistic, and
gauss models: a case study in iraq and
egypt,” Processes, vol. 8, no. 11, p. 1400,
2020.
[107] M. Tavakoli, J. Carriere, and A. Torabi,
“Robotics, smart wearable technologies,
and autonomous intelligent systems for
healthcare during the covid-19 pandemic:
An analysis of the state of the art and future
vision,” Advanced Intelligent Systems, vol.
2, no. 7, p. 2000071, 2020
[108] J. Torales, M. O’Higgins, J. M.
Castaldelli-Maia, and A. Ventriglio, “The
outbreak of covid-19 coronavirus and its
impact on global mental health,”
International journal of social psychiatry,
vol. 66, no. 4, pp. 317–320, 2020.
[109] S. Li, Y. Wang, J. Xue, N. Zhao, and T. Zhu,
“The impact of covid-19 epidemic
declaration on psychological consequences:
a study on active weibo users,”
International Journal of Environmental
Research and Public Health, vol. 17, no. 6,
p. 2032, 2020.
[110] T. T. Nguyen, Q. V. H. Nguyen, D. T.
Nguyen, E. B. Hsu, S. Yang, and P. Eklund,
“Artificial intelligence in the battle against
coronavirus (covid-19): a survey and future
research directions,” arXiv preprint
arXiv:2008.07343, 2020, [Online].
https://arxiv.org/abs/2008.07343 (Accessed
Date: March 17, 2020).
[111] P. L. Fung, M. A. Zaidan, O. Surakhi, S.
Tarkoma, T. Petaj¨a,¨ and Hussein, “Data
imputation in in situ-measured particle size
dis-tributions by means of neural networks,”
Atmospheric Measurement Techniques, vol.
14, no. 8, pp. 5535–5554, 2021.
[112] A. Salameh and O. Surakhi, “An optimized
convolutional neural network for
handwritten digital recognition
classification,” Journal of Theoretical and
Applied Information Technology, vol. 98, no.
21, pp. 3494–3503, 2020
[113] O. M. Surakhi, M. A. Zaidan, S. Serhan, I.
Salah, and T. Hussein, “An optimal stacked
ensemble deep learning model for
predicting time-series data using a genetic
algorithm—an application for aerosol
particle number concentrations,”
Computers, vol. 9, no. 4, p. 89, 2020.
[114] O. Surakhi, S. Serhan, and I. Salah, “On the
ensemble of recurrent neural network for air
pollution forecasting: Issues and challenges,”
Adv. Sci. Technol. Eng. Syst. J, vol. 5, pp.
512–526, 2020
[115] Yeng, P. K., Nweke, L. O., Woldaregay, A.
Z., Yang, B., & Snekkenes, E. A. (2021).
Data-driven and artificial intelligence (AI)
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
214
Volume 21, 2024
approach for modelling and analyzing
healthcare security practice: a systematic
review. In Intelligent Systems and
Applications: Proceedings of the 2020
Intelligent Systems Conference (IntelliSys),
Vol. 1 (pp. 1-18). Springer International
Publishing.
[116] O Tariq, N., Qamar, A., Asim, M., & Khan,
F. A. (2020). Blockchain and smart
healthcare security: a survey. Procedia
Computer Science, vol. 175, 2020, pp.
615–20.
[117] M. A. Zaidan, O. Surakhi, P. L. Fung, and T.
Hussein, “Sensitivity analysis for predicting
sub-micron aerosol concentrations based on
meteorological parameters,” Sensors, vol.
20, no. 10, p. 2876, 2020.
Contribution of Individual Authors to the
Creation of a Scientific Article (Ghostwriting
Policy)
- Walid Salameh, Ola Surakhi and Mohammad
Khanafseh carried out the formal analysis,
investigation, and methodology.
- Walid Salameh leads the planning and execution
of research activities
- The original draft is created and prepared by Ola
Surakhi
- Walid Salameh, Ola Surakhi and Mohammad
Khanafseh reviewed and edited the final version
Sources of Funding for Research Presented in a
Scientific Article or Scientific Article Itself
This research was funded by the Scientific Research
Deanship at Princess Sumaya University for
Technology.
Conflict of Interest
The authors have no conflicts of interest to declare.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.e
n_US
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
215
Volume 21, 2024
APPENDIX
Table 2. ML and DL technologies used in the Diagnosis and Identification of COVID-19.
Related Works
ML/DL Methods
Features
Datasets
Country
Results
[69]
Multilayer perceptron
artificial neural
networks and decision
trees
Patient’s medical
history and symptoms
and Patient’s clinical
outcome
Mexican Federal
Health Secretary,
General Director of
Epidemiology
Mexican
The model achieved
up to 80% prediction
accuracy for the
dataset used
[70]
logistic regression
(LR), random forest
(RF), and extreme
gradient boosting
(XGB)
Clinical features (20)
287 COVID-19
samples of patients
from the King Fahad
University Hospital
Saudi Arabia
The RF outperformed
the other classifiers
with an accuracy of
0.95
[71]
Convolutional Neural
Network (CNN)
Chest CT scans
4563
three-dimensional (3D)
volumetric chest CT
scans from 3506
patients acquired at six
medical centers
between August 16,
2016, and February 17
China
A deep learning
method was able to
identify coronavirus
disease 2019 on chest
CT scans
[72]
3D CNN model
Chest CT scans
498 CT scans from
151 positive
COVID-19 subjects
and 497 CT scans from
different subjects with
various types of
pneumonia
China
Achieved a moderate
diagnostic ability
overall (Area under
the curve (AUC) of
0.70 with 99%
Confidential interval
(0.56–0.85)
[73]
Using SVM (Support
Vector Machine),
CNN (Conventional
Neural Networks),
ResNet50,
InceptionResNetV2,
Xception, VGGNet16
X-ray images
5857 Chest X-rays and
767 Chest CTs for
COVID-19-positive
cases
-
Achieve 84% and 75%
classification accuracy
[74]
deep learning model
cough, breathing, and
speech
Coswara dataset
-
AUC of 96.4% and an
accuracy of 96%
[75]
convolutional neural
network
Cough, voice, and
breath
sounds dataset from
Cambridge University
UK
The proposed
approach improves
system performance to
diagnose COVID-19
disease and provides
better results on the
COVID-19 respiratory
sound dataset.
[76]
VGG19 and U-Net
X-ray images
BIMCV-COVID19, ,
BIMCV-COVID, and
Spain Pre-COVID era
dataset
Spain
Achieved 97% of
accuracy
[77]
Regression Linear
Regression model
COVID-19 confirmed,
and mortal cases
Egyptian Ministry of
health and population
Egypt
The proposed model
was beneficial for the
Egyptian government
in managing the
COVID-19 outbreak
for the following
months.
[78]
Classification Logistic
Regression
Patient health record
United States health
systems 197 patients
USA
The proposed
algorithm can
accurately identify
16% more patients
than a widely used
scoring system while
minimizing
false-positive results
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
216
Volume 21, 2024
Table 3. ML and DL technologies used in the Forecasting of COVID-19
Related
Works
ML/DL Methods
Features
Datasets
Country
Results
[88]
linear regression (LR), least
absolute shrinkage, selection
operator (LASSO), support vector
machine (SVM), and exponential
smoothing (ES)
The number of new
positive cases, the
number of deaths, and
the number of
recoveries.
GitHub repository
provided by the
Center for Systems
Science and
Engineering, Johns
Hopkins University
[89]
International
The ES performs best among all
the used models.
[90]
Long short-term memory (LSTM)
network
Meteorological and
mobility data
Google Cloud online
COVID-19
Japan
The proposed framework
provided more accurate and
consistent estimations than that
offered by Google Cloud
[91]
encoding–decoding LSTM
Confirmed positive
cases, death cases, and
recovery cases
Saudi Ministry of
Health website, the
existing interactive
dashboards, and the
available application
program interface
(API)
International
The proposed model generated
high accuracy with less error rate
[92]
Bayesian regression neural
network, cubist regression,
k-nearest neighbors, quantile
random forest, variational mode
decomposition, and support vector
regression
cumulative
COVID-19 cases and
exogenous variables
such as daily
temperature and
precipitation
COVID-19 Data
Repository [2]
And API (Application
Program Interface) in
27 Brazilian State
Health Offices
Brazilian and
American
states
It was observed that climatic
variables, such as temperature and
precipitation, indeed influence
increasing the accuracy when
predicting COVID-19 cases, and
the adopted models can be
recommended as a promising
model for forecasting
[93]
Machine learning algorithms along
with SIR and SIR-F models
Patient information,
mobility, number of
cases
John Hopkins
University dataset
Saudi Arabia
The results show that government
lockdowns and isolation of
individuals are not enough to stop
the pandemic
[94]
supervised machine learning
algorithms
human mobility data,
number of cases
Collected
USA
tree-based classifiers performed
best on the forecasting task.
Gradient Boosting had the highest
classification accuracy.
[95]
A hybrid machine learning method
of adaptive network-based fuzzy
inference system (ANFIS) and
multi-layered
perceptron-imperialist competitive
algorithm (MLP-ICA)
Number of cases and
deaths
The statistical reports
of COVID-19 cases
and mortality rate of
Hungary
Hungary
suggests machine learning as a
potential technology to be
considered to model the outbreak
[96]
SEIR model and Regression model
confirmed cases
John Hopkins
University dataset
India
The model gave a short-term
prediction (two weeks) to help the
Government and doctors prepare
their plans
[97]
linear regression, Multilayer
perceptron, and Vector
autoregression method
confirmed, death, and
recovered cases
COVID-19 Kaggle
data
India
The MLP method gave better
prediction results than that of the
LR and VAR method
[98]
Logistic model and FbProphet
model
confirmed cases,
recovered cases, and
death cases
John Hopkins
University dataset
International
The model can significantly
improve estimates of the number
of infections.
[99]
Long short-term memory (LSTM)
and Gated Recurrent Unit (GRU)
confirmed, negative,
released, deceased
cases
kaggle
-
The proposed approach helps
generate suitable results based on
the critical disease outbreak
[100]
quasi-Poisson regression
the trends of the daily
incident diagnosed
cases, deaths, and
intensive care unit
admissions
The websites of the
Italian and Spanish
Ministries of Health
Italy and
Spain
the positive signs already shown
by the decreasing trend slopes
after a more restrictive lockdown
in Italy and Spain
WSEAS TRANSACTIONS on BIOLOGY and BIOMEDICINE
DOI: 10.37394/23208.2024.21.21
Walid Salameh, Ola M. Surakhi,
Mohammad Y. Khanafseh
E-ISSN: 2224-2902
217
Volume 21, 2024
