Received 27 December 2023, accepted 2 March 2024, date of publication 4 March 2024, date of current version 14 March 2024.

Digital Object Identifier 10.1109/ACCESS.2024.3373646

Heart Disease Detection Using Feature Extraction

and Artificial Neural Networks:
A Sensor-Based Approach
AWAD BIN NAEEM 1 , BISWARANJAN SENAPATI2 , (Senior Member, IEEE), DIPEN BHUVA3 ,
ABDELHAMID ZAIDI 4 , ABHISHEK BHUVA5 , MD. SAKIUL ISLAM SUDMAN6 ,
AND AYMAN E. M. AHMED7
1 Department of Computer Science, National College of Business Administration Economics, Multan Campus, Multan 60000, Pakistan
2 Department of Computer Science, Parker Hannifin Corporation, Cleveland, IL 44124, USA
3 Department of Electrical Engineering and Computer Science, Cleveland State University, Cleveland, IL 44115, USA
4 Department of Mathematics, College of Science, Qassim University, Buraydah 51452, Saudi Arabia
5 Department of Computer Science, University of Massachusetts, Boston, MA 02125, USA
6 Department of Electrical Engineering, The University of Texas at Arlington, Arlington, TX 76019, USA
7 Faculty of Engineering, King Salman International University, El-Tor 8701301, Egypt

Corresponding author: Awad Bin Naeem ([email protected])

ABSTRACT This study presents a novel technique for identifying individuals using feature extraction
methods and signal processing approaches. It uses an artificial neural network (ANN) technique to identify
scent patterns in individuals using ten metal oxide semiconductor sensors. Sensor data is scanned and
extracted before using ANN patterns. Before using ANN patterns to generate patterns from sensor data,
it is important to scan and extract sensory information from that data. Each participant is recognized and
scanned for a totally of 1000 different characteristics during the course of the multiple investigations, which
are conducted across a variety of time periods that include 5, 10, 15, and 20 people. Because of the varying
time periods, signals from sensors are received in analog form, which is then transformed by Arduino into
digital form. It is necessary to train an architecture on the data set that has been created. The benchmarks
that are employed for the assessment of the model that is presented for the identification of human odor
include sensitivity, f-measures, accuracy, and specificity, among other things. Experiments are carried out
using the assessment measures, and the findings demonstrate that this model has an accuracy of greater than
85 % in most cases. The research demonstrates the potential of feature extraction methods in identifying
individuals and enhancing human odor identification.

INDEX TERMS Heart disease, deep learning, ANN, feature extraction, sensor.

I. INTRODUCTION primarily be considered throughout this research. Medical

Individuals and medical practitioners (such as physicians and care providers and patients now face obstacles in diagnosing
surgeons) who are dealing with heart illness must take various heart illness that can only be overcome by developing a sys-
aspects into account when using the phrase ‘‘heart disease’’ tem based on artificial neural networks [3]. The system would
[1]. Finally, we’ll talk about potential treatments. This study’s be specially designed to reduce the mortality rate associated
major objective is to construct an expert system that is espe- with heart disease while simultaneously increasing rates of
cially targeted at producing a better model for detecting heart effective diagnosis [4].
disease effectively [2]. It is vital to note that such features will By using neural networks to detect and eradicate errors in
medical expertise, such as misdiagnosis, a prompt preventa-
The associate editor coordinating the review of this manuscript and tive approach might be implemented to eliminate the illness
approving it for publication was Yongming Li . at its source [5]. The development of such a model, however,
2024 The Authors. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.
VOLUME 12, 2024 For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/ 37349
 A. B. Naeem et al.: Heart Disease Detection Using Feature Extraction and ANNs

would need the use of a specialized data set that would in cardiac computed tomography (CT) scans that use x-ray
include the data of numerous individuals who had previously images [18].
been diagnosed with a suspected cardiac condition [6]. It is The present advancement problematic concerns:
expected that this study would help to determine the pattern • Smooth integration of heterogeneous devices
of the brain network that connects the diagnosed patients • Estimation of relevant to the patient’s activities
with other patients. It is suggested in this study that a new • Communication of the accident to the doctor, who is in
Neural Network expert system be developed for the detection charge of recognizing disease and impairment heart aus-
of heart disease. By their symptoms, they suggested some of cultation is the term used to describe the characteristics
the parameters used in the proposed system as input [7]. of the heart sound.
Thallium, a radioactive tracer that is injected during a
stress test, is also used as an input to the knowledge base It is the most rudimentary and widely used approach for the
of the expert neural network system, which is comprised of early identification of cardiac disorders, and it is based on
a database and a rule base and is critical to the system’s the use of heart sounds (HS). Phonocardiogram is the graph
used to represent the HS recording (PCG). The method of
performance [8]. Chronic diseases are long-term illnesses that
auscultation is dependent on the doctor’s abilities). The pulse
are recorded by the World Health Organization (WHO) and
and blood flow are represented by the HS symbol, which is
include a variety of illnesses that are fundamental and major
closer to the heart valves. The cell is the most fundamental
health problems throughout the world. Chronic diseases are
unit of life. Each cell in the human body is one-trillionth of a
defined as illnesses that last for an extended period [9].
meter square. They absorb vitamins from food, convert those
Factors such as other diseases, and all have the same risk fac-
vitamins into energy, and perform unique functions in addi-
tors. Dietary characteristics associated with chronic illnesses
tion to providing body composition. The cells also carry the
include a high intake of meat and oil/fats, as well as a low
body’s fundamental substance and can multiply themselves.
intake of grains and vegetables, among other things [10]. This Cells are made up of several sections, each of which performs
sort of diet promoted an inactive lifestyle, which resulted in a specific purpose. Some of these components are referred to
greater exposure to cigarettes and alcohol [11]. as organelles, and there are specialized systems that perform
In the past 10 years, it has been noted that the number of beneficial functions inside the cell. Normally, they contain
older persons is fast increasing, with many of them suffering just the right amount of each kind of cell in the right place.
from chronic conditions, making healthcare services neces- Because the cell produces a signal that may be utilized to
sary. There is a pressing need to transform healthcare and control how often and how regularly it is propagated. If any
give patients an efficient, creative, and economical solution of these indicators is missing or faulty, cells grow and take
that can be delivered anywhere and at any time in a friendly too many blows, resulting in the formation of a mass known
and cost-effective way [12]. A paradigm of illness preven- as a tumor.
tion that may postpone the need for treatment and effective There are certain classic methods of diagnosis, such as a
intervention is required by us as a society at this time [13]. physical examination and a review of the patient’s medical
Healthcare and medicine must make use of self-tracking history. The tests listed below may provide findings for the
technologies, which have shown to be effective in motivat- risk factor for heart disease diagnosis and can be used to
ing and encouraging such excellent healthcare habits [14]. determine whether a person is at risk. Cardiovascular disease
Over the last year, there has been increased interest in the is a term used to describe disorders of the heart and blood
application of self-tracking to the idea of self-identification. arteries. The correct functioning of the heart is very essential
Self-tracking devices are primarily wireless biosensors that for the existence of all living things. If you or a loved one
may be utilized in the home, automobile, office, and other has heart problems and is not aware that he or she is suffer-
environments, as well as implanted in the body. Some health- ing from the ailment, it might result in death. Diagnosis of
care applications and services are digitally accessible via a coronary artery disease in its early stages is critical for the
mobile device. Self-monitoring has the potential to improve preservation of human life. The number of people dying from
health daily [15]. The purpose of this evaluation is to establish heart disease is increasing dramatically. Cardiovascular dis-
the present demand and market offer, as well as to identify ease is often considered to be the leading cause of mortality
chances for excellent health. The possibility for adult, middle- around the world, according to some estimates. The challenge
aged, and young individuals to live independently and with is to determine the most accurate method of diagnosing car-
a high standard of living, with large-scale societal and eco- diac disease. To increase the accuracy of detection, several
nomic factors advantages, exists [16]. machine-learning algorithms have been applied to a set of
The main cause of death in the world is heart disease, and chosen characteristics.
these types of disorders may be classified in a variety of In this research paper, we aim to investigate the influ-
different ways. Cardiovascular disease is the leading cause ence of the actual use of a Neural Network System on the
of mortality in every country around the globe [17]. In 2019, accuracy of heart disease detection. By exploring various
heart-related disorders claimed the lives of 616,000 individ- aspects of the Neural Network System’s implementation and
uals. Employed that is based on radio waves and magnetic. performance, we hope to gain insights into its practical impli-
It is also possible to assess the detailed architecture of the cations in reducing misdiagnosis caused by human error.
heart using echo and other technologies that may be employed Moreover, we will examine whether a well-designed neural

37350 VOLUME 12, 2024

A. B. Naeem et al.: Heart Disease Detection Using Feature Extraction and ANNs

network system has the potential to lower the likelihood of and different classifiers, concluding that artificial neural
heart disease in humans. By analyzing relevant studies and networks (ANNs) have a positive influence on the early
data, we seek to determine the effectiveness of such sys- diagnosis of cardiac illness in both humans and animals. The
tems in detecting early signs of heart disease and facilitating application of machine learning methods is becoming more
timely interventions. Additionally, we aim to provide valu- popular in medical decision support systems. Medical diag-
able advice to medical professionals based on our findings. nosis aids in the identification of many characteristics that
By identifying key areas for improvement and highlighting indicate the various versions of the illness. It is conceivable
best practices, we intend to support healthcare providers in that a disease will present with signs that are relevant, irrele-
their efforts to diagnose heart disease accurately and effi- vant, and redundant, depending on the diagnostic processes
ciently. By addressing these research questions, we hope used to identify [19]. The presence of redundant charac-
to contribute to the ongoing advancements in the field of teristics contributes to the incorrect categorization of the
medical diagnostics, particularly in relation to heart disease condition. As a result, deleting unnecessary characteristics
detection, ultimately improving patient outcomes and reduc- decreases the quantity of the data as well as the difficulty of
ing the burden of cardiovascular diseases in the population. the calculation. Identifying an appropriate feature subset for
The novelty of this study is that it has a broad application in efficient classification is a difficult issue that requires much
various settings such as homes, offices, markets, and public effort. A thorough search of the data set’s sample space is
places, in addition to trauma centers, hospitals, medical units, required to get this result [20]. According to, increasingly
THQ hospitals, primary health care units, medical, DHQ contribute to the advancement of some developing results for
hospitals, national-level hospitals, and other locations. In the a potentially successful framework for the detection of heart
disciplines of medicine and pattern recognition, the develop- disease. The use of an ensemble-based strategy, which was
ment of smart medical devices for the monitoring of cardiac based on a mix of previous data models and current cardiac
disease is an exciting research problem. The ultimate goal of problems, may have been the basis for their findings. Wissler
this research is to develop technologies that can automatically and colleagues have argued that the use of a data-driven
detect illnesses linked to disasters before they occur. These framework will continue to have a significant impact on
biomarkers, which are derived from perspiration and skin just how medical practitioners detect cardiac chronic condi-
breath and emitted radiations, may one day be utilized to tions in the future [21]. Because medical practitioners have
diagnose diseases of the heart as the primary goal of this only been able to detect heart illness with a limited num-
research. In addition, the following are research contribution: ber of signs and with a limited number of instruments in
1) The goal of this paper is to develop a neural net- the past, the authors were very concerned about the possi-
work system that is successful in diagnosing cardiac bility of a mistake occurring. Has been argued repeatedly,
problems. with the result that the possibility of misdiagnosis has been
2) To determine if a neural network system can be used reduced to a great extent. However, no such model has been
to remove the possibility of misdiagnosis in the case of developed yet. The authors have used data mining classi-
cardiac illness. fication approaches to the cardiovascular disease dataset,
3) To compare the efficacy of a neural network system for which has been evaluated by them. As part of their investi-
heart disease diagnosis to that of previous frameworks gation and construction of a multistage-multivariate quality
for heart disease diagnosis. management system leveraging artificial neural networks,
4) For using current frameworks and neural network sys- the investigation concluded that the network that had been
tems, determine the amount of misdiagnosis likelihood created in this manner was capable of properly detecting the
of cardiac illness at the cellular level. signals [22].
This manuscript is structured as follows. State-of-the-art Using dynamic recognition patterns of the ECG, inves-
is discussed in Section II. In Section III, the context of tigated another major use of (electrocardiogram) ECG
materials and methods is presented. In which we will go approaches, in which the ECG was utilized to classify car-
through the research methodological phases and the biomet- diovascular illness. In this study, the use and demonstration
rical model. Then we describe the identification procedure, of technological tools in the classification and identification
which includes sensing systems, testing and testing prepro- of CV D are carried out via the analysis of the FuWai and
cessing, and feature extraction of data. And then we discuss PTB databases, respectively. It was noted that the use of ANN
architecture sensing system for sensing data. Section IV con- for the categorization of cardiovascular and diabetes diseases
sists of extensive experimental results and their discussions. should be considered. An ANN approach such as is used
The conclusion and future work of this study is described in to test the hypothesis of greater accuracy in data gathering
Section V. when identifying diabetes and heart-related disorders and
diagnoses, which is supported by the results. This is also the
II. LITERATURE REVIEW sort of ANN that has been employed the most often in the
When it comes to the early identification of cardiac illness, articles that have been evaluated from 2008 to 2017 [23].
emphasized the relevance of categorization of the ECG using A report published was the starting point for an investigation
methodologies combined with Artificial Neural Networks. into a suggested method for general practitioners to consult
The researchers experiment with many components of the on the treatment provided to patients suffering from chest
ECG, such as various databases, extraction methodologies, discomfort. The researchers concluded that 8.4 percent of

VOLUME 12, 2024 37351

 A. B. Naeem et al.: Heart Disease Detection Using Feature Extraction and ANNs

patients who experienced chest discomfort were at risk of in the prior literature. For a robust and effective disease
developing a serious type of heart disease by the completion risk prediction, the suggested NFR model incorporates two
of their trial. The data set for the research included 22,294 methodologies that make use of the AUC and accuracy to
patients, with 28 individuals having chest pain. The applica- attain a high degree of precision. The application of machine
tion of machine learning methods is becoming more popular learning methods is becoming more popular in medical deci-
in medical decision support systems. Medical diagnosis aids sion support systems [30]. Medical diagnosis aids in the
in the identification of many characteristics that indicate identification of many characteristics that indicate the various
the various versions of the illness. It is conceivable that a versions of the illness. It is conceivable that a disease will
disease will present with signs that are relevant, irrelevant, present with signs that are relevant, irrelevant, and redundant,
and redundant, depending on the diagnostic processes used depending on the diagnostic processes used to identify it.
to identify [24]. The presence of redundant characteristics The presence of redundant characteristics contributes to the
contributes to the incorrect categorization of the condition. incorrect categorization of the condition [31]. As a result,
As a result, deleting unnecessary characteristics decreases the unnecessary characteristics decrease the quantity of the data
quantity of the data as well as the difficulty of the calcula- as well as the difficulty of the calculation. Identifying an
tion. Identifying an appropriate feature subset for efficient appropriate feature subset for efficient classification is a dif-
classification is a difficult issue that requires much effort. ficult issue that requires much effort. A thorough search of
A thorough search of the data set’s sample space is required the data set’s sample space is required to get this result [32].
to get this result [25]. One of the leading causes of death
across the world is heart failure, particularly in developing III. MATERIAL AND METHOD
countries. The diagnosis of heart failure is difficult, particu- This section contains the methodology of the paper.
larly in undeveloped and underdeveloped nations where there
is a scarcity of human specialists and specialized technology.
As a result, the models that were created outperformed the
data that was tested. We provide a unique diagnostic system
and a new pruning strategy to construct an intelligent system
FIGURE 1. Research method.
that performs well on both training and testing data. that
combines pre-pruning and post-pruning to boost classifica-
tion accuracy and tree size in this work. We can generate
a decision tree using this strategy [26]. The experimental A. RESEARCH METHODOLOGY
findings are derived from 18 benchmark data sets obtained Figure 1 shows the process of the research methodology.
from the University of California, Irvine A repository for
machine learning. Based on the data, we can see that our 1) SENSING
tree pruning technique reduces tree size significantly while Sensors are used to detect the stench of people in the first
simultaneously improving algorithm accuracy [27]. place. There are three distinct kinds of glands in human skin,
A novel approach based on machine learning techniques and it is these glands that are responsible for excreting volatile
is proposed in this study to predict the likelihood of coro- organic chemicals, as well as the volatile compounds them-
nary artery atherosclerosis developing. To estimate missing selves, which do not vary over time. During the production of
values in the atherosclerosis data sets, a ridge expectation glands, bacterial activity occurs on the human skin, causing
maximization imputation approach has been developed. The an odor. Volatile organic compounds (VOCs) are excreted by
conditional likelihood maximization approach is used to people in large amounts, but only a small amount of these
eliminate unimportant properties from the feature space and main volatile chemicals is required for this study, and the
lower the size of the feature space, allowing for faster learn- amount excreted has no relation to age or the surroundings.
ing. The performance of two classification algorithms for There are many glands in the human body that produce
the prediction of heart disease is examined and compared to volatile organic molecules, making it difficult to conduct a
earlier research. The impact of missing value imputation on comprehensive study of human odor. According to research,
prediction performance is also investigated, and it is discov- the body excretes various VOCs in distinct locations. Some of
ered that the suggested REMI method outperforms standard the VOCs, such as aldehydes, ketones, alcohols, carboxylic
approaches by a large margin [28]. Modern machine learning acids, esters, hydrocarbons, and others, are targeted for this
and data mining methods are essential in healthcare systems procedure. Several scent sensors and a series of MQ sensors
since they efficiently translate all available data into useful are employed in this procedure. For this procedure, there is a
information and improve patient outcomes. According to total of nine MQ-series sensors are employed; these sensors
literary works, there is a potential of a 12 percent inaccuracy are numbered 2, 3, 4, 5, 6, 7, 8, 9, 135. Each sensor has a
in the diagnosis of illnesses by medical practitioners, which unique set of abilities, such as being able to handle a variety
may be confirmed. Furthermore, when it comes to effective of volatile substances. Other scent sensors are used to assess
disease risk prediction in medical analysis, the area under the quality of the air. The TGS-2602 smell sensor is also
the curve with accuracy as an assessment measure is given used to measure air quality and detect pollutants such as
more weight than the accuracy of the forecast [29]. The ammonia, VOC, and hydrochloric acid in the air, whereas
AUC’s function, on the other hand, has not been well defined the QS-01 sensor is primarily used to monitor water quality.

37352 VOLUME 12, 2024

A. B. Naeem et al.: Heart Disease Detection Using Feature Extraction and ANNs

These 11 sensors detect the excreted targeted computation 5) DECISION MODULE

from the human body or skin in this procedure. As a last step, a decision module is used to make a match
that is deemed acceptable. This module’s primary goal is to
2) TRANSMISSION establish a personal threshold value for each individual and
to emphasize privacy.
Arduino microcontroller has been used to transmit features
with 14 pins.
C. BIOMETRICAL MODEL
Figure 2 model is a design, which is biometric. This session
3) RECEIVING discusses the research questions and methodology used to
The Arduino is used to convert analog impulses to digital address the hypothesis in the introduction. The ANN model
ones, and the sensors are used to gather the data. The infor- is briefly explained, and real-time data sets or self-generated
mation obtained is in the form of digitally stored raw data data sets are described to achieve the highest accuracy. A vari-
derived from the smell of human skin. The Arduino is used ety of research methodologies are proposed based on rele-
to capture and transmit the raw data to the computer for vant factors and the study’s objectives. Secondary research
additional processing and authentication purposes. is emphasized to identify previous theoretical approaches
and factors relevant to the study’s goals. The study uses
4) PROCESSING 300 patients’ medical records from various hospitals and
There is a requirement for data binning at this stage since the derived heart disease diagnosis system input variables. The
length of digital data is not determined. Binning, also known output is divided into two categories: ‘‘absence of illness’’
as bucketing, is a pre-processing method used to collect data. and ‘‘presence of illness.’’ Heart disease can be predicted
Data binning is necessary for the storage of data since data is using the Back Propagation Algorithm, a learning method
constantly being collected and there is no limit to the length commonly employed in Artificial Neural Networks (ANNs).
of data that may be obtained. The data set includes patients aged 29 to 79, with sexual
orientation indicating sex for men and women in a patient’s
territory unit. Cardiopathy can be demonstrated by four types
B. ACQUISITION ANALYSIS of pain: type one angina, which is caused by constrained coro-
Chemicals excreted from different areas of the body have nary veins, non-angina torment caused by various factors, and
varying levels of similarity. A variety of sensors are employed the fourth type is unlikely to be a sign of cardiovascular dis-
for the detection of common compounds, depending on their ease. The results are analyzed using treetops, the examination
capacity to detect them. The biometric system uses a unique of resting authority per unit area, and Chol, the steroid’s alco-
set of data strings for the chemicals acquired from the body hol problem counterpart. The quick glucose level is measured
parts of people. using Fbs, while resting clinical instrument results, thalach,
and exacting are indicators of angina triggered by activity.
1) QUALITY CHECK MODULE
Old pinnacles include ST discouragement, incline, and ca.
The activity check’s duration is referred to as the numb and
Data quality and the ability to re-sense it is both verified
its classification is numb. A method called ‘‘order’’ is used
by this module form. To determine whether the sample
to predict future outcomes based on existing data. A group of
resolution is primarily for high resolution, it is also the
classifiers is used to improve order exactness and distinguish
responsibility of the same module or system to make this
proof of heart condition. The preparation data set is divided
determination. A module should be used to construct it after
into a training and validation set, and the test data set is used
it has been determined that it is for high resolution.
to test the classifiers’ intensity.

2) FEATURE GENERATOR
The biometric samples collect the biometric characteristics
produced by this module. The feature generator generates and
extracts biometric traits in the form of digital data. By com-
bining all of these traits, a biometric template is created.

3) BIO METRIC TEMPLATE

A biometric template is a collection of biometric characteris-
tics generated from a set of samples using a feature generator.

4) MATCHER MODULE
When using the matcher module, you may compare newly FIGURE 2. Biometric model.
stored templates against those that have previously been
saved. Using one or more previously recorded templates, Figure 3 is a block diagram illustrating the training and
we may do this comparison. testing methodology used to identify an individual.

VOLUME 12, 2024 37353

 A. B. Naeem et al.: Heart Disease Detection Using Feature Extraction and ANNs

5) TESTING
The testing data set is used to create the data set for the testing
phase. It is at this point that the system’s accuracy is deter-
mined to determine whether or not it is working correctly,
and whether or not it is generating the expected results.

E. PRE-PROCESSING
Sensor array response is compressed and sample-to-sample
discrepancies are reduced during the pre-processing step.
FIGURE 3. Training and testing.
Compression and Normalization are two common techniques
for dealing with this issue.

D. IDENTIFICATION PROCESS 1) FEATURE EXTRACTION

Using a sensor system and a pattern recognition system sep- Using a feature extraction approach, it is possible to glean
arates this process into two halves. useful and important information from raw data. After that,
discrete clusters of significant data were created from the
retrieved data. Since there are a lot of variables in a rough data
1) SENSING SYSTEM
set, this is done to save computational resources. Because raw
Is capable of distinguishing between several scents in its data contains so many resources, additional computational
immediate surroundings. To collect samples, this system is power was necessary. Feature extraction techniques integrate
used. An array of MQ sensors and other scent sensors forms relevant variables into features. Fewer computer resources are
the basis of the whole concept. This array of sensors is needed to analyze this data and also accurately reflect the real
capable of detecting various VOCs. In this operation, a metal wide data
oxide array is employed since each sensor is responsible for
a particular duty and detects a distinct gas. These sensors
F. GENERAL MODEL OF SENSING SYSTEM
have a tiny heater. The heaters in sensors are responsible
The sensor array detects odors, and an artificial neural net-
for producing the electrochemical when the VOCs that are
work algorithm determines their order, as shown in fig 4.
evaporating off human skin come into touch with the heaters
In the recommended approach, the gathering of data is illus-
present in the sensors. The data on human odor is gathered
trated in the following graphic portrayal of the sensors.
after an electric current has passed through the circuit. This
model can be utilized at ambient temperature because of the
technology employed in it. As an odor identification system,
a database is developed to store and organize information
about the many types of smells.

2) PATTERN RECOGNITION SYSTEM

For this method of pattern recognition, it’s important to know
that a pattern in the digital world may be mathematically
expressed as well as seen in real life. Pattern recognition
uses a variety of machine learning methods. Analysis of data
FIGURE 4. Biometric model.
stored in databases, as well as pattern recognition of data
stored in databases, is used to classify the data.
The graphic above depicts the VOCs that are exhaled by the
hands of a human being. Volatile organic compounds (VOCs)
3) PATTERN RECOGNITION LEARNING AND TESTING are detected when a person’s hands come into contact with
Giving the system some unique inputs allow it to learn. The an array of sensors. The heaters within the sensors interact
training system is becoming a lot more complex. A system’s with the human body’s volatile organic compounds when
performance determines the algorithm used to analyze the this happens. Analog signals are received from sensors and
data. Afterward, the data set is split into two categories: converted into digital signals by a microcontroller. Only the
training and testing. last five seconds’ worth of data have been collected and are
shown in the table1 underneath.
4) TRAINING
To create a working model of a training system, this method G. ANALYZER OF ECG
is used. At this point, a model is being trained by exposing it Figure 5 explains the ECG analysis process.
to a collection of data in the form of human odor. This data
set is used to run the algorithms, which establish connections 1) ARCHITECTURE
between the input and output data to create a model and The architecture of the training model is described in the
extract the obtained data. figure 6.

37354 VOLUME 12, 2024

A. B. Naeem et al.: Heart Disease Detection Using Feature Extraction and ANNs

TABLE 1. Sample data set.

FIGURE 7. Atrial fibrillation.

FIGURE 8. Placement of electrodes.

H. SIGNALS GENERATING
FIGURE 5. ECG analyzer. 1) STEP I
The data is saved in an Excel file shown in Figure 9.

FIGURE 9. Signals generating step I.

2) STEP II
MATLAB is used to create a new Simulink model shown in
Figure 10, which is then used to build the signals.
FIGURE 6. Architecture.
3) STEP III
2) ATRIAL FIBRILLATION MATLAB is capable of reading data from an Excel workbook
Figure 7 explains atrial fibrillation which explains the ECG as shown in Figure 11.
graph to detect a first-degree heart attack.
4) STEP IV
3) POSITION OF ECG The ECG data provided in the accompanying graphic is cre-
Figure 8 explains the placement of electrodes. ated in signal-building form shown in Figure 12.

VOLUME 12, 2024 37355

 A. B. Naeem et al.: Heart Disease Detection Using Feature Extraction and ANNs

FIGURE 13. Analyzer algorithm.

FIGURE 10. Simulink model for signals generating step 2.

FIGURE 14. Normal ECG data.

FIGURE 11. Excel workbook signals generating step III.

FIGURE 15. Atrial fibrillation ECG data.

3) DATA SET GENERATED FOR FIRST-DEGREE HEART

ATTACK ECG DATA
Figure 16 shows generated first-degree heart attack ECG data.
FIGURE 12. Accompanying graphic for ecg signals generating step IV.

IV. RESULTS AND EXPERIMENTS

I. ANALYZER ALGORITHM OF ECG
An electronic nose has been built for the tests and outcomes.
Analyzing the process of ECG is shown in Figure 13. In this e-nose, gas sensors from the MQ2 through MQ9 series,
as well as TGS2602, mq135, and Qs01, are used. These
1) DATA SET IS GENERATED FOR NORMAL ECG DATA gas sensors assess organic volatile chemicals in human body
Figure 14 shows normal ECG data. odor with the use of these gas sensors. The TGS 2602 sen-
sor can detect isovaleric acid in human body odor, unlike
2) DATA SET IS GENERATED FOR ATRIAL FIBRILLATION ECG the MQ series sensors that incorporate heaters. The term
DATA ‘‘electrochemical sensor’’ refers to a device whose heaters
Figure 15 shows generated atrial fibrillation normal ECG interact with the human body to produce an electrochemical
data. response.

37356 VOLUME 12, 2024

A. B. Naeem et al.: Heart Disease Detection Using Feature Extraction and ANNs

FIGURE 16. First-degree heart attack ECG data.

At room temperature, sensors can detect the odor of the

human body, allowing them to collect information about
a person’s health. The smell may be used to identify a
person’s odor, and sensors act appropriately. Sensors for
detecting various gases are addressed in the first chapter.
The odor of a human body is utilized to collect data on
an Arduino UNO microcontroller. The data is sent using a
module from Arduino. The Arduino UNO module is basic
and straightforward to use, and the hardware may be pro-
grammed with ease. Sensors on this development board allow
it to read inputs and convert them to outputs. These are the FIGURE 17. Experiment No 1 result of ANN of 25 Sample Data.
features of Arduino UNO, a microcontroller based on the
Atmega328p. A typical Arduino board has 14 input/output
pins, six of which are dedicated to analog signals while
the other six are utilized for output. The clock speed of
Arduino is 16 MHZ, and it has two buttons: a start and a
reset button. Also, include a USB cord and a power jack.
So that the Arduino can communicate the data from sensors
to the PC. Using Arduino’s six analog input/output pins, data
is converted to a digital format. After the raw data in the
form of human odor is obtained, the binning procedure is
carried out, and then a neural network code is worked on
in MATLAB for authentication purposes. Twenty people’s
records with five odor samples each are maintained in the
database.

A. FIRST EXPERIMENT
Sensors are used to collect odor samples from the hands of
five people in the first experiment, and analog signals from
human hands are picked up by the 10 sensors. The sensors
are utilized one at a time, and each one takes around five
seconds to process. Data from sensors are analog, and it is
this analog data that Arduino transforms into digital data. For
example, the first five seconds of a sensor’s continuous output
data are collected. The 25 samples of odor were collected FIGURE 18. Experiment No 2 result of ANN of 50 Sample Data
from five people in five seconds, and this data was entered for 10 Sensor.
into a database. A neural network approach known as the
feedforward technique is used to classify patterns in sensory Figure 17 shows the outcomes. In the NN algorithm,
data. MATLAB’s neural network recognition software uses a person’s hand-created fragrance patterns are identified.
the NN (neural network) method on samples of human body Accuracy parameters are specified for standard measurement
odor it receives. assessments that are 73±2.

VOLUME 12, 2024 37357

 A. B. Naeem et al.: Heart Disease Detection Using Feature Extraction and ANNs

data is being gathered from a variety of sensors, including

the 10 sensors that detect analog impulses transmitted by
human hands. To apply the array approach, two arrays are
built, and data is gathered by arranging sensors sequentially.
The first six sensors are gathered, and then the remaining four
sensors are collected. Because the data is in analog form, it is
transformed into digital with the aid of Arduino. The data
from sensors is continuous, thus only the first five seconds
of data are acquired. The data from five people are used to
collect 50 samples of odor in five seconds, and this data is
saved in a database. A neural network approach known as
the feed-forward technique is utilized for pattern recognition
from sensory data. The recognition program of the neural
network of MATLAB performs the NN (neural network)
method on samples of odor received from the human body,
Figure 18 shows the obtained findings. The NN algorithm
recognizes the created patterns of scents obtained by human
hands. A parameter of 80±2 is specified for the precision of
standard measure assessment.

C. THIRD EXPERIMENT
In the third experiment, data from ten people is collected for
tests by removing the odor from each person’s hands. At this
moment, data is collected using various sensors, and analog
FIGURE 19. Experiment No 3 result of ANN of 50 Sample Data signals sent to the ten sensors can detect the presence of
for 5 Sensor.
human hands. To apply the array approach, two arrays are
built, and data is gathered by arranging sensors sequentially.
In the beginning, data from the first six sensors are gathered.
And then data is collected from the last four sensors. Because
the data is in analog form, it is transformed into digital with
the aid of Arduino. The data from sensors is continuous, thus
only the first five seconds of data are acquired. The data from
five people are used to collect 50 samples of odor in five
seconds, and this data is saved in a database. A neural network
approach known as the feed-forward technique is utilized
for pattern recognition from sensory data. The recognition
program of the neural network of MATLAB performs the
NN (neural network) method on samples of odor received
from the human body. The obtained findings are shown in
Figure 19. The NN algorithm recognizes the created patterns
of scents that are collected by human hands. A parameter
of 86±2 is specified for the precision of standard measure
assessment.

D. FORTH EXPERIMENT
To gather data for the third experiment, odors from the hands
of 20 participants were removed and analyzed. Currently,
data is being gathered from a variety of sensors, including
the 10 sensors that detect analog impulses transmitted by
human hands. To apply the array approach, two arrays are
built, and data is gathered by arranging sensors sequentially.
In the beginning, data from the first six sensors are gathered.
FIGURE 20. Experiment No 4 result of ANN of 100 Sample Data And then data is collected from the last four sensors. Because
for 10 Sensor. the data is in analog form, it is transformed into digital with
the aid of Arduino. The data from sensors is continuous, thus
B. SECOND EXPERIMENT only the first five seconds of data are acquired. According to
The fragrance of each person’s hands is gathered for testing in the data from five people, 100 samples of odor are collected
the second experiment, which includes 10 persons. Currently, in five seconds and stored in a database. A neural network

37358 VOLUME 12, 2024

A. B. Naeem et al.: Heart Disease Detection Using Feature Extraction and ANNs

approach known as the feed-forward technique is utilized TABLE 2. Comparison of five experimental results.
for pattern recognition from sensory data. The recognition
program of the neural network of MATLAB performs the
NN (neural network) method on samples of odor received
from the human body. The obtained findings are shown in
Figure 20. The NN algorithm recognizes the created patterns
of scents that are collected by human hands. A parameter
of 90±4 is specified for the accuracy of standard measure
assessment.

E. EXPERIMENT NO.5
To collect data for the third experiment, odors from the
hands of each of the 20 participants were removed. Analog
signals from human hands are picked up by the 10 sensors
at this point. Sensors are placed in a series using the array
approach. This necessitates the creation of two arrays and the
acquisition of data. Initially, six sensors are used, and then the
three remaining sensors are used. Because the information is
TABLE 3. Comparison of experimental results.
analog, Arduino is used to convert it to electronic. There is a
constant stream of data from sensors. Thus, only the first five
seconds of data are collected. According to the data of five
people, 100 samples of odor are acquired in five seconds and
stored in a database. A neural network approach known as
the feed-forward technique is used to classify patterns from
sensory input. The recognition program of the neural net-
work of MATLAB performs NN (neural network) algorithm
on samples of odor received from the human body. Fig 21
depicts the findings obtained. The ANN algorithm is used
to identify the patterns of scents that are created by human
hands. Standard measurements are evaluated using a 95 %
accuracy parameter.

FIGURE 21. Experiment No 5 result of ANN of 100 Sample Data

for 5 Sensor.

Five different trials are carried out to train the system.

In the initial experiment, five people’s hands were used to
gather smell samples, and data were acquired using a variety
of sensors, including 10 sensors that sensed analog impulses
from the hands. Table 2 shows that 71% of the time, the experiment is carried out to improve accuracy by increasing
results are correct. The average accuracy is attained by repeat- the number of data sets. For this reason, the number of people
ing tests several times on the same data set. The second investigated is raised. The data from 10 people is acquired

VOLUME 12, 2024 37359

 A. B. Naeem et al.: Heart Disease Detection Using Feature Extraction and ANNs

TABLE 4. Results obtained through weka. TABLE 4. (Continued.) Results obtained through weka.

for tests in the second experiment by collecting the odor of

each person’s hands. The ten sensors detect analog impulses
given by human hands at this same time utilizing a variety
of sensors. To apply the array approach, two arrays are built,
and data is gathered by arranging sensors sequentially. Each
individual collects ten samples for the experiment.
The acquired findings are displayed in Fig 21, which
demonstrates that increasing the number of subjects increases
accuracy to 85 %. Using the same data set, the number of
trials is expanded to get an average accuracy of 80.2 %.
In the third experiment, a six-sensor array is employed to
gather data. The odor of ten participants is collected at various
intervals of time for studies, and the findings reveal that
the accuracy level has grown to 84.40 %. Experiments are
repeated several times with the same data sets to reach the
average accuracy, which is 86.2 %. Another experiment is
being conducted to boost the accuracy level. For this aim, the
number of individuals has been raised to twenty for the odor
samples. Using scent samples from 20 people, the accuracy
rate is 86.30 %. No matter how many times you do the same
experiment, you only ever get a 95 % accuracy rate once.
As a result, your overall success rate is 86 %. Table 3 shows
a sample data set.
Furthermore, to verify the robustness of the proposed
model, table 4 results using several machine learning tech-
niques (classifiers) on the data set acquired locally were
produced.

V. CONCLUSION
Human body odor is thought to be a great identifier and is
employed in wearable devices for biometric identification.
The odor samples are subjected to a neural network
algorithm for the identification of olfactory patterns. Many
researchers employed various ways, but in one study, an array

37360 VOLUME 12, 2024

A. B. Naeem et al.: Heart Disease Detection Using Feature Extraction and ANNs

of diverse sensors utilized as an electronic nose can detect [6] I. M. El-Hasnony, O. M. Elzeki, A. Alshehri, and H. Salem, ‘‘Multi-label
human odor proved to be the cheapest way between all sys- active learning-based machine learning model for heart disease predic-
tion,’’ Sensors, vol. 22, no. 3, p. 1184, Feb. 2022, doi: 10.3390/s22031184.
tems. The analog singles are transformed into digital signals [7] G. G. N. Geweid and M. A. Abdallah, ‘‘A new automatic identification
using an Arduino microcontroller for authentication and pat- method of heart failure using improved support vector machine based on
tern classification using a neural network. The information on duality optimization technique,’’ IEEE Access, vol. 7, pp. 149595–149611,
samples acquired from various sensors is shown below. In the 2019.
[8] A. Haldorai and A. Ramu, ‘‘Security and channel noise management
first experiment, 10 sensors are utilized to collect data from in cognitive radio networks,’’ Comput. Electr. Eng., vol. 87, Oct. 2020,
five people. For data collecting, each sensor is utilized inde- Art. no. 106784, doi: 10.1016/j.compeleceng.2020.106784.
pendently. According to the statistics in Table 4, the accuracy [9] M. I. Hossain, M. H. Maruf, M. A. R. Khan, F. S. Prity, S. Fatema,
M. S. Ejaz, and M. A. S. Khan, ‘‘Heart disease prediction using distinct
is 71%. The average accuracy is 73.2 % as a result of repeated
artificial intelligence techniques: Performance analysis and comparison,’’
tests on the same data set, with the number of datasets increas- Iran. J. Comput. Sci., vol. 6, pp. 397–417, Jun. 2023, doi: 10.1007/s42044-
ing as the accuracy increases. There are indeed 10 data sets 023-00148-7.
throughout the second experiment, and sensor arrays are used [10] A. B. Naeem, A. M. Soomro, H. M. Saim, and H. Malik, ‘‘Smart road
management system for prioritized autonomous vehicles under vehicle-
to capture the data. Two sensor arrays are developed in the to-everything (V2X) communication,’’ Multimedia Tools Appl., pp. 1–18,
second experiment. Data from 10 person’s samples were used Oct. 2023, doi: 10.1007/s11042-023-16950-1.
as trials for each array. Table 4 shows the results of the study, [11] A. Ishaq, S. Sadiq, M. Umer, S. Ullah, S. Mirjalili, V. Rupapara, and
which show that the accuracy of the results is 80.5 %. The M. Nappi, ‘‘Improving the prediction of heart failure patients’ survival
using SMOTE and effective data mining techniques,’’ IEEE Access, vol. 9,
average is obtained by repeating tests on the same datasets pp. 39707–39716, 2021, doi: 10.1109/ACCESS.2021.3064084.
80.2 % of the time. The third experiment employs an array of [12] N. Karthikeyan, P. Padmanaban, A. Prasanth, and D. Ragunath, ‘‘Machine
sis sensors. The results reveal that the accuracy is enhanced learning based classification models for heart disease prediction,’’ J. Phys.,
Conf., vol. 1916, no. 1, May 2021, Art. no. 012092, doi: 10.1088/1742-
in comparison to the previously achieved accuracy using odor
6596/1916/1/012092.
samples gathered at various periods of 10 people. At this [13] M. A. Naser, A. A. Majeed, M. Alsabah, T. R. Al-Shaikhli, and K. M. Kaky,
moment, the accuracy is 84.40 %. The average accuracy is ‘‘A review of machine learning’s role in cardiovascular disease prediction:
86.2 % as a result of repeated repetitions of tests done on the Recent advances and future challenges,’’ Algorithms, vol. 17, no. 2, p. 78,
2024, doi: 10.3390/a17020078.
same data sets. The accuracy of the earlier collected samples [14] A. B. Naeem, F. Khalid, A. M. Soomro, A. D. Del Mundo, A. Zaidi,
is raised by increasing the number of samples, which is set B. Senapati, and O. P. Doshi, ‘‘Early gender identification of date
to twenty. Using the same materials, additional studies are palm using machine learning,’’ J. Comput. Biomed. Inform., vol. 4,
carried out on the odors of twenty persons. The acquired no. 2, pp. 128–141, 2023. [Online]. Available: https://www.jcbi.org/index.
php/Main/article/view/147
accuracy is 86 % most of the time, and only once does it reach [15] N. Louridi, S. Douzi, and B. El Ouahidi, ‘‘Machine learning-based identi-
95 %, therefore ultimate accuracy is 86 %. It is concluded that fication of patients with a cardiovascular defect,’’ J. Big Data, vol. 8, no. 1,
certain scents are not correctly detected, but other samples p. 133, Dec. 2021, doi: 10.1186/s40537-021-00524-9.
[16] S. Abbas, S. Ojo, A. Al Hejaili, G. A. Sampedro, A. Almadhor,
are correctly identified, resulting in an average of 86 %.
M. M. Zaidi, and N. Kryvinska, ‘‘Artificial intelligence framework for
In the future, effort must be done to improve accuracy by heart disease classification from audio signals,’’ Sci. Rep., vol. 14, no. 1,
improving sensor performance and sensitivity. The sensitivity p. 3123, Feb. 2024, doi: 10.1038/s41598-024-53778-7.
of the sensors must be increased to enhance accuracy. When [17] E. Maini, B. Venkateswarlu, B. Maini, and D. Marwaha, ‘‘Machine
learning–based heart disease prediction system for Indian population:
a person uses deodorants, there are issues with accuracy that An exploratory study done in South India,’’ Med. J. Armed Forces
must be addressed. Detecting human scent became a pressing India, vol. 77, no. 3, pp. 302–311, Jul. 2021, doi: 10.1016/j.mjafi.2020.
need in the marketplace. Regardless of whether a person 10.013.
applies deodorant or not, enhanced technology should be able [18] A. M. Soomro, A. B. Naeem, K. Shahzad, A. M. Madni, A. D. D. Mundo,
M. Sajid, and M. A. Baloch, ‘‘Forecasting cotton whitefly population using
to distinguish between human aroma and other odors. deep learning,’’ J. Comput. Biomed. Informat., vol. 4, no. 1, pp. 64–76,
Dec. 2022, doi: 10.56979/401/2022/67.
REFERENCES [19] A. B. Naeem, A. M. Soomro, A. Bhuva, K. Bashir, D. Bhuva,
[1] M. M. R. Khan Mamun and T. Elfouly, ‘‘Detection of cardiovascular R. R. Maaliw, and W. M. F. Abdel-Rehim, ‘‘Intelligent four-way crossroad
disease from clinical parameters using a one-dimensional convolutional safety management for autonomous, non-autonomous and VIP vehicles,’’
neural network,’’ Bioengineering, vol. 10, no. 7, p. 796, Jul. 2023, doi: in Proc. IEEE Int. Conf. Emerg. Trends Eng., Sci. Technol., Jan. 2023,
10.3390/bioengineering10070796. pp. 1–6, doi: 10.1109/icest56843.2023.10138829.
[2] M. M. Ahsan and Z. Siddique, ‘‘Machine learning-based heart disease [20] K. M. Mohi Uddin, R. Ripa, N. Yeasmin, N. Biswas, and S. K. Dey,
diagnosis: A systematic literature review,’’ Artif. Intell. Med., vol. 128, ‘‘Machine learning-based approach to the diagnosis of cardiovascular vas-
Jun. 2022, Art. no. 102289, doi: 10.1016/j.artmed.2022.102289. cular disease using a combined dataset,’’ Intell.-Based Med., vol. 7, 2023,
[3] A. B. Naeem, B. Senapati, A. S. Chauhan, S. Kumar, J. C. O. Gavilan, Art. no. 100100, doi: 10.1016/j.ibmed.2023.100100.
and W. M. F. Abdel-Rehim, ‘‘Deep learning models for cotton [21] A. B. Naeem, B. Senapati, M. S. Islam Sudman, K. Bashir, and
leaf disease detection with VGG-16,’’ Int. J. Intell. Syst. Appl. A. E. M. Ahmed, ‘‘Intelligent road management system for autonomous,
Eng., vol. 11, no. 2, pp. 550–556, 2023. [Online]. Available: non-autonomous, and VIP vehicles,’’ World Electric Vehicle J., vol. 14,
https://www.ijisae.org/index.php/IJISAE/article/view/2710 no. 9, p. 238, Sep. 2023, doi: 10.3390/wevj14090238.
[4] A. Albahr, M. Albahar, M. Thanoon, and M. Binsawad, ‘‘Computational [22] R. G. Nadakinamani, A. Reyana, S. Kautish, A. S. Vibith, Y. Gupta,
learning model for prediction of heart disease using machine learning S. F. Abdelwahab, and A. W. Mohamed, ‘‘Clinical data analysis
based on a new regularizer,’’ Comput. Intell. Neurosci., vol. 2021, pp. 1–10, for prediction of cardiovascular disease using machine learning tech-
Nov. 2021, doi: 10.1155/2021/8628335. niques,’’ Comput. Intell. Neurosci., vol. 2022, pp. 1–13, Jan. 2022, doi:
[5] A. B. Naeem, B. Senapati, A. S. Chauhan, M. Makhija, A. Singh, 10.1155/2022/2973324.
M. Gupta, P. K. Tiwari, and W. M. F. Abdel-Rehim, ‘‘Hypothyroidism [23] U. Nagavelli, D. Samanta, and P. Chakraborty, ‘‘Machine
disease diagnosis by using machine learning algorithms,’’ Int. J. Intell. learning technology-based heart disease detection models,’’ J.
Syst. Appl. Eng., vol. 11, no. 3, pp. 368–373, 2023. [Online]. Available: Healthcare Eng., vol. 2022, pp. 1–9, Feb. 2022, doi: 10.1155/
https://www.ijisae.org/index.php/IJISAE/article/view/3178 2022/7351061.

VOLUME 12, 2024 37361

 A. B. Naeem et al.: Heart Disease Detection Using Feature Extraction and ANNs

[24] A. M. Soomro, A. B. Naeem, B. Senapati, K. Bashir, S. Pradhan, BISWARANJAN SENAPATI (Senior Member,
R. R. Maaliw, and H. A. Sakr, ‘‘Constructor development: Predicting IEEE) received the first Ph.D. degree in computer
object communication errors,’’ in Proc. IEEE Int. Conf. Emerg. Trends science from the University of Arkansas at Lit-
Eng., Sci. Technol., Bahawalpur, Pakistan, Jan. 2023, pp. 1–7. tle Rock, Little Rock, AR, USA. He is currently
[25] A. Rahim, Y. Rasheed, F. Azam, M. W. Anwar, M. A. Rahim, pursuing the second Ph.D. degree in quantum
and A. W. Muzaffar, ‘‘An integrated machine learning framework computing and quantum communication theory in
for effective prediction of cardiovascular diseases,’’ IEEE Access, Industry 5.0. He has achieved ABD status. He has
vol. 9, pp. 106575–106588, 2021, doi: 10.1109/ACCESS.2021. more than 20 years of industry experience as a
3098688. solution architect, a designer, and an implementer,
[26] A. B. Naeem, B. Senapati, G. A. L. Mahadin, V. Ghulaxe, F. Almeida, as well as a strong passion for research and innova-
S. I. Sudman, and M. I. Ghafoor, ‘‘Determine the prevalence of hepatitis tion in information science, information technologies, quantum computing,
B and C during pregnancy by using machine learning algorithm,’’ Int.
business, and IT management, and a strong focus and interest in IT/IS
J. Intell. Syst. Appl. Eng., vol. 12, no. 13, pp. 744–751, 2024. [Online].
research areas within machine learning, data sciences, AI, deep learning,
Available: https://www.ijisae.org/index.php/IJISAE/article/view/4704
and quantum computing applications to solve real-world business problems.
[27] A. Saboor, M. Usman, S. Ali, A. Samad, M. F. Abrar, and N. Ullah,
A technology innovator and yogic personality who prefers to focus on aca-
‘‘A method for improving prediction of human heart disease using machine
demic research to assist students, researchers, and associates at universities,
learning algorithms,’’ Mobile Inf. Syst., vol. 2022, pp. 1–9, Mar. 2022, doi:
10.1155/2022/1410169. research institutions, and academic institutions to pursue a challenging career
in a progressive environment, where learning, innovation, and creativity
[28] J. Talukdar and T. P. Singh, ‘‘Early prediction of cardiovascular
disease using artificial neural network,’’ J. Paladyn Behav. Robot., are encouraged and where his skills and knowledge can be enhanced to
vol. 14, no. 1, Feb. 2023, Art. no. 20220107, doi: 10.1515/ their maximum potential to contribute to the overall success and growth
pjbr-2022-0107. of the organization, as well as its impact on society and socio-economic
[29] R. Tr, U. K. Lilhore, S. Simaiya, A. Kaur, and M. Hamdi, ‘‘Pre- benefits. He is a Senior Member of Region-4 IEEE, an Active Student Chair
dictive analysis of heart diseases with machine learning approaches,’’ Activity (SAC Chair) in IEEE-Chicago 4, and the Chair of SPX members for
Malaysian J. Comput. Sci., pp. 132–148, Mar. 2022, doi: 10.22452/mjcs. better reach in research and innovations in science, technology, and quantum
sp2022no1.10. computing.
[30] S. Tuli, N. Basumatary, S. S. Gill, M. Kahani, R. C. Arya, G. S. Wander, and
R. Buyya, ‘‘HealthFog: An ensemble deep learning based smart healthcare
system for automatic diagnosis of heart diseases in integrated IoT and
DIPEN BHUVA, photograph and biography not available at the time of
fog computing environments,’’ Future Gener. Comput. Syst., vol. 104,
publication.
pp. 187–200, Mar. 2020, doi: 10.1016/j.future.2019.10.043.
[31] L. Zhu, J. Shen, L. Xie, and Z. Cheng, ‘‘Unsupervised topic hypergraph
hashing for efficient mobile image retrieval,’’ IEEE Trans. Cybern., vol. 47,
no. 11, pp. 3941–3954, Nov. 2017. ABDELHAMID ZAIDI received the Ph.D. degree
[32] A. Sharma and P. K. Mishra, ‘‘Performance analysis of machine learning in statistics from Université Grenoble Alpes,
based optimized feature selection approaches for breast cancer diagno- France, and the Engineering degree in computer
sis,’’ Int. J. Inf. Technol., vol. 14, no. 4, pp. 1949–1960, Jun. 2022, doi: science and applied mathematics from ENSIMAG
10.1007/s41870-021-00671-5. Grenoble, France. He is currently an Associate
Professor with the College of Science, Qassim
University, Saudi Arabia. He works mainly on the
development of computational methods applied to
various subjects of signal and image processing.
He is also the author of three books, such as
Numerical Analysis, Algorithms, and Probability and Statistics. His research
work was published in many top-ranked journals.

ABHISHEK BHUVA, photograph and biography not available at the time of

publication.

MD. SAKIUL ISLAM SUDMAN received the

AWAD BIN NAEEM received the master’s degree
B.Sc. degree in electrical and electronic engineer-
in computer science from the National College
ing from the Ahsanullah University of Science and
of Business Administration Economics, Multan
Technology, Bangladesh, in 2018. He is currently
Campus, in September 2022, with particularly in
pursuing the M.Sc. degree in electrical engineer-
the fields of smart transportation systems, machine
ing with The University of Texas at Arlington.
learning, and deep learning, which sparked his
He is a Graduate Research Assistant with the Shi-
interest in the subject. The master’s studies
madzu Institute Nanotechnology Research Center,
focused on vehicle-to-everything (V2X) commu-
The University of Texas at Arlington. He is also
nication and the development of an intelligent
researching low-frequency noise characteristics of
transportation system for different types of vehi-
MOS devices. This includes device physics of MOS components and mini-
cles. He also proposed an intelligent transportation system framework for
mizing low-frequency noise of diodes in low currents. His research interests
autonomous, non-autonomous, and prioritized vehicles. He is passionate
include data science and engineering and noise reliability in MOS devices.
about intelligent transportation systems and data science problems. He has an
impressive research background, having submitted 17 research publications,
eight of which have been published. He has presented at 12 international
conferences in various countries. His research interests include detection, AYMAN E. M. AHMED, photograph and biography not available at the time
gender identification, population forecasting, deep learning algorithms, and of publication.
machine learning algorithms.

37362 VOLUME 12, 2024