“FINAL REPORT FORMAT EXAMPLE FOR GP (1)”

Classifying Chest X-ray Images as Normal, Corona and Viral

Pneumonia Diseases using Multi-class SVM

201710138 ‫محمود محمد الخطيب‬

201710946 ‫عبدهللا عصام سليمان‬
201710308 ‫عبدهللا لؤي احمد‬
201710031 ‫احمد محمد عوده‬

Under the supervision of

Dr. Amani Al-Ghraibah

Submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in
Medical Engineering

Faculty of Engineering
Al-Ahliyya Amman University
Amman – Jordan

[January 2022]

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
1-22
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

Table of Contents

Table of Contents........................................................................................................................................................ I
List of Figures ............................................................................................................................................................ II
List of Acronyms ..................................................................................................................................................... III
Chapter 1: Introduction.......................................................................................................................................... 1
Chapter 2: Literature Review...............................................................................................................................3
Chapter 3: Methodology......................................................................................................................................... 6
References................................................................................................................................................................... 11

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
1-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

List of Figures

Figure 1: Examples of chest X-ray images ..…………......…………………………………2

Figure 2: A flowchart of proposed method …………………………………….……..……6

Figure 3 example of ROI process ……………………………………………………….…7
Figure 4 Histogram equalization processing……………………………….………………8
Figure 5 SVM hyperplane………………………………………………….………………10

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
2-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

List of Acronyms

PCR: Polymerase Chain Reaction

CT: Computed Tomography
SARS-CoV-2: Severe Acute Respiratory Syndrome Coronavirus 2
GLCM : Gray Level Co-Occurrence Matrix
SVM : Support Vector Machine
CAD : Computer-aided detection
DL : deep learning

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
3-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

Chapter 1: Introduction

A coronavirus is a kind of common virus that causes an infection in your nose, sinuses, or
upper throat. In early 2020, after a December 2019 outbreak in China, the World Health
Organization identified SARS-CoV-2 as a new type of coronavirus. The outbreak quickly
spread around the world. COVID-19 is a disease caused by SARS-CoV-2 that can trigger
what doctors call a respiratory tract infection. It can affect your upper respiratory tract
(sinuses, nose, and throat) or lower respiratory tract (windpipe and lungs).[1]
It spreads the same way other coronaviruses do, mainly through person-to-person contact.
And its infections range are from mild to deadly.

SARS-CoV-2 is one of seven types of coronavirus, including the ones that cause severe
diseases like Middle East respiratory syndrome (MERS) and sudden acute respiratory
syndrome (SARS)[4]. The other coronaviruses cause most of the colds that affect us during
the year but aren't a serious threat for other healthy people. COVID-19 may develop to
pneumonia, pulmonary failure and death . Pneumonia seems to be the most common severe
manifestation of COVID-19, distinguished mainly by fever, dry cough and dyspnea. COVID-
19 may share some common radiographic features with other pneumonias(like bacterial-
pneumonia and viral pneumonia ), which making its discriminability difficult for radiologists.
[2]

Pneumonia is a common lung infection where the lungs’ air sacks become inflamed. These
sacs may also filled with fluid, pus, and cellular debris. It can be caused by viruses, fungi, or
bacteria .Bacterial pneumonia may involve just one small section of your lung, or it may
encompass your entire lung. Pneumonia can make it difficult for your body to get enough
oxygen to your blood, which can cause cells to not work properly. While viral pneumonia is a
complication of the viruses that causes colds and the flu.[22]

Since the beginning of this pandemic until today, Corona cases in Jordan have reached more
than one million cases, and deaths have reached more than 12000 deaths. A critical step in the

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
4-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

fight against COVID-19 is effective screening of infected patients, such that those infected
can receive immediate treatment and care, as well as be isolated to mitigate the spread of the
virus.[3] Current diagnostic tests for COVID-19 include reverse-transcription polymerase
chain reaction (RT-PCR), real-time RT-PCR (rRT-PCR), and reverse-transcription loop-
mediated isothermal amplification (RT-LAMP) .[5]
Patients who have been exposed to the virus and present severe symptoms could still get a
negative result in the RT-PCR test [6]. Therefore, in these cases, COVID-19 should be
diagnosed with medical imaging techniques, such as chest X-ray or CT scan.

Although CT has proved to be one of the most precise diagnostic methods for COVID-19 , it
has some important limitations, including around 70× higher ionizing radiation than X-ray ,
its high cost, and the fact that it cannot be performed as a bedside test .[7]

Figure 1 Examples of chest X-ray images [7]

Moreover, CT is not suitable for monitoring the evolution of specific cases, particularly in
critical ill patients.On the other hand, X-ray is a less sensitive modality in the detection of
COVID-19 as compared to CT [8]. However, X-ray is a cheaper and faster alternative, and it
is also available in most hospitals. Therefore, X-ray will likely be the primary imaging
modality used for COVID-19 diagnosis and management.[8]

With high clinical suspicion for COVID-19 infection, positive X-ray findings can obviate the
need for CT scanning . However, it is important to consider that these techniques may present
limitations to particular patients such as pregnant women, since they could cause harm to
EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
5-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

unborn children .[9]

This work, proposes a method to discriminate between healthy, COVID-19 and viral
pneumonia diseases based on new features and machine learning . The novelty of the project
is the new features that will be extracted from the chest X-ray images and used to build an
automated model using multi-class Support Vector Machine (SVM). The model will
distinguish between the three diseases and classify any new input X-ray image as Normal,
COVID-19 or Viral pneumonia disease.

Chapter 2: Literature Review

Wang and his colleagues [10] designed a method to classify chest x-ray images as Covid-19
and pneumonia. Their method focused on using a computer-aided detection CAD scheme
based on deep learning (DL) for detection and localization of COVID-19.The proposed CAD
framework is composed of two steps with DLs: the Discrimination-DL and the Localization-
DL. The result was very good and it reached 98% for Discrimination-DL. and 93% accuracy
for the other method Localization-DL.[10]

Abu Owidah et al [11]proposed method to to classify chest x-ray images as normal images or
images with COVID-19 disease using advanced deep learning method. Effective features
were extracted using wavelet analysis and Mel Frequency Cepstral Coefficients (MFCC)
method, and they used in the classification process using the Support Vector Machine (SVM)
classifier .The results obtained an overall accuracy of 98.8% using five wavelet features,
where the classification using MFCC features, MFCC-delta, and MFCC-delta-delta features
reached accuracy around 97% on average[11]

Abbas et al. [12] validated and a deep CNN, called Decompose, Transfer, and Compose
(DeTraC), for the classification of COVID-19 chest X-ray images. they trained a
DeTraCResNet18-based binary model with 196 images (COVID-19 = 105, normal = 80,

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
6-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

SARS = 11) to detect COVID and the model achieved 95.12% accuracy, 97.91% sensitivity,
91.87% specificity, and 93.36% precision[12]

Wang and his colleagues [13] designed COVID-Net based on deep convolutional neural
network to detect of COVID-19 cases from chest X-ray (CXR) images. They used 3 different
groups of X-ray images, i.e. COVID 19, pneumonia, and normal/healthy. They got a
classification accuracy of 92.6% with images from all groups.[13]

Apostolopoulos and Mpesiana[14] evaluated the performance of different CNNs (VGG19,

Inception, MobileNetV2, Xception, and InceptionResNetV2) in COVID-19 detection. They
used three different groups of Xray images comprising of COVID 19, pneumonia, and
normal . the highest achieved accuracy was 96.78%.[14]

Alquran his colleagues [15] employs various image processing techniques besides extracted
texture features from the radiological images and feeds them to different artificial intelligence
(AI) scenarios to distinguish between normal, pneumonia, and COVID-19 cases. The best
overall accuracy achieved is 93.1% by exploiting Ensemble classifier[15].

Afshar et al .[16] developed an integrated deep learning model focused on a capsule network
and used a four-class dataset of chest X-ray images (Normal, Bacterial, non-COVID-19 viral,
and COVID-19 viral infection) achieving 95.17% accuracy[16].

Chowdhury et al. [17]proposed a parallel-dilated convolutional neural network (CNN) based

COVID-19 detection system from chest X-ray images, named as Parallel-Dilated COVIDNet
(PDCOVIDNet) . Difering convolution dilation rate in a parallel form demonstrates the
proof-of-principle for using PDCOVIDNet to extract radiological features for COVID-19
detection. performance metrics: the accuracy, precision, recall and F1 scores reach 96.58%,
96.58%, 96.59% and 96.58%, respectively [17].

Duran-Lopez et al.[18] proposed a system that performs a set of preprocessing algorithms to

the input images for variability reduction and contrast enhancement, which are then fed to a
custom Convolutional Neural Network in order to extract relevant features and perform the

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
7-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

classification between COVID-19 and normal cases. Their work achieved an average
accuracy of 94.43% and an AUC of 0.988 [18].

Hussain et al.[19] proposed a new CNN model to detect COVID-19 using chest X-ray .This
model called CoroDet and it is capable of providing accurate diagnostics for 2 class
classification (COVID and Normal), 3 class classification (COVID, Normal, and
Pneumonia), and 4 class classification (COVID, Normal, non-COVID viral pneumonia, and
non-COVID bacterial pneumonia). The classification accuracy of the proposed model is
99.1% for 2 classes, 94.2% for 3 class cases, and 91.2% for 4 classes classification, which is
the highest achieved accuracy [19].

A.Alqudah and A.M. Alqudah [20]employed different Artificial Intelligence (AI) techniques
to propose a system for early detection of COVID-19 using chest X-ray images. These
images are classified using different AI algorithms and a combination of them, then their
performance was evaluated to recognize the best of them. These algorithms include a
convolutional neural network (CNN), Softmax, support vector machine (SVM), Random
Forest, and K nearest neighbor (KNN). Here CNN is into two scenarios, the first one is to
classify X-ray images using a softmax classifier, and the second one is to extract automated
features from images and pass these features into other classifiers (SVM, RFF, and KNN).
According to the results, the performance of all classifiers was good and most of them record
accuracy, sensitivity, specificity, and precision more than 98% [20].

Wang et al.[21] proposed an efficient diagnostic method that uses a combination of deep
features and machine learning classification to classify between normal cases and COVID-19
diseases. Results of the diagnostic accuracy of Xception + SVM is as high as 99.33% [21]

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
8-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

Chapter 3: Methodology
The block diagram in Figure 2 shows the proposed steps that will be done in this project.

Image Extract
Input dataset Classification
Processing Features

Figure 2 A flowchart of our proposed method

3.1 Dataset Description

The images has been used in this project are taken from kaggle website [22] ,it is a free
online souce for the dataset. The dataset is organized into 3 classes (covid, pneumonia ,
normal) .we used a 1200 images from each classes. Figure 3 shows the sample images from
three classes

(a) (b) (c)

Figure 3 Sample images of the used dataset [22]

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
9-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

3.2 Image Preprocessing

Complexity, inaccuracy, and inadequacy are some of the issues associated with image data.
Therefor, before building an automated classification system, the data must be preprocessed
(cleaned and formatted to the desired format) in order to achieve the desired results[23].
There are several techniques used to preprocess image data. Examples include ; denoising
image and improves image contrast .

3.2.1 Average filter

Average (or mean) filtering is a technique for 'smoothing' images by reducing the intensity
variance between neighbouring pixels. The average filter replaces each value with the
average value of neighboring pixels, including itself, as it moves through the image pixel by
pixel. It is often used to reduce noise in images. Average filtering is often taken for a
convolution filter. It is built on a kernel, which represents the form and size of the
neighborhoods to be sampled while calculating the mean, just like other convolutions.[25]

3.2.2 Adjust image intensity values

In an image, intensity of a pixel is defined as the value of the pixel. For example in grayscale
image there are 256 gray levels . In grayscale image can have a value from 0 to 255 and that
will be its intensity .Intensity adjustment is an image enhancement technique that adjust an
image's intensity values to improve an image. Figure 4 shows effect of adjust intensity
values[26].

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
10-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

Figure 4 Adjust image intensity values [26]

3.2 Feature Extraction

Feature extraction is a technique for extracting visual content from images for indexing and
retrieval. Primitive or low level image features can be either general features, such as
extraction of color, texture and shape or domain features[27]

3.2.1 Texture Feature

Texture is a repeated pattern of information or structure arrangement with regular intervals.
Texture, in general, refers to the surface characteristics and appearance of an object as
determined by the size, shape, density, arrangement, and proportion of its elementary parts.
Gray Level Co-Occurrence Matrix (GLCM) is employed for textural feature extraction as it is
a technique that allows extraction of statistical information from the image regarding the
pixel pair distributions. GLCM is an effective method for texture analysis, especially in
biomedical images . The textured features consist of Contrast (CON), Correlation (CORR),
Dissimilarity (DISS), Angular Second Moment (ASM), and Entropy (ENT) [27].

3.2.2 Wavelet features

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
11-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

Wavelets are powerful time-frequency analysis tools, which have wide applications in image
processing. Wavelet analysis is used to divide information present on an image into two
discrete components — approximations and details . A image is passed through two filters,
high pass and low pass filters. Then decomposed into high frequency (details) and low
frequency components (approximation). At every level, we get 4 images. The approximation
shows an overall trend of pixel values and the details as the horizontal, vertical and diagonal
components. By using the absolute values for the wavelet coefficients, three image details are
used to determine the energy of any decomposition level (the high pass images). Then the
total energy for the three high pass images is computed. If the number of the decomposition
levels set to be five ,then five energy values corresponding to each of the 5 decomposition
levels will be extracted.[28]

3.3 Classification
The Support Vector Machine classifier is a binary supervised classifier that can be improved
and used for multiclass classification in different scenarios. SVM basically finds a separation
region between the margins of two classes and maximize it by identifying hyperplanes
between the margins of two classes. This can be done by employing kernel functions like
linear, polynomial, and radial basis functions to transfer the employed features into a higher-
dimensional space (RBF). It's called linear SVM if the kernel is linear, and it's utilized when
the data is linearly separable. Other kernels, such as Polynomial, RBF, have been employed
to distinguish between distinct classes [29]. Figure 5 shows the hyperplane between two
classes.

In most basic type of SVM, it doesn’t support multiclass classification. However, for
multiclass classification, the same principle is utilized after breaking down the multi-
classification problem into smaller subproblems, all of which are binary classification
problems. The popular methods which are used to perform multi-classification on the
problem statements using SVM are as follows: One vs One (OVO) approach, One vs All
(OVA) approach, and Directed Acyclic Graph (DAG) approach. In this project, one of the

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
12-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

three methods will be used to classify the chest X-ray images as one of the three proposed
classes.

Figure 5 SVM hyperplane [29]

Chapter 4: Result and discussion

Firstly we divided the dataset to 90% training and 10 % testing. Then we applied average
filter to remove noise .following to improve the appearance of image we adjust image
intensity values. After that we extract features from three classes of image to reduce the
amount of redundant data from the dataset. SVM classifier are applied to classifying and
recognition of images.
4.1 result of image processing
As we shown in figure 6 the effect of average filter is that the image become smooth .

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
13-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

Figure 6 Applied average filter on images

After that to adjust image intensity values of image we used a imadjust . Figure 7 shows how
the image became more brightness

Figure 7 :(a) Original image (b) imadjust image

4.2 Evaluation of SVM Classifier

To evaluate the performance of a classification algorithm ( classification model ), we will
detect a confusion matrix. A confusion matrix is a technique for summarizing the
performance of a classification algorithm. Calculating a confusion matrix can give you a
better idea of what your classification model is getting right and what types of errors it is
making[30].

4.2.1 evaluation of SVM classification model using texture features

Figure shows the confusion matrix of the training and testing image using texture features.

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
14-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

Figure8 The Confusion matrix of 3 classes with a texture features

4.2.2 evaluation of SVM classification model using Wavelet features

Figure9 shows the confusion matrix of the training and testing image using Wavelet features

Figure 9 The Confusion matrix of 3 classes with a Wavelet features

4.2.3 Evaluation of SVM classification model using combine texture and Wavelet

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
15-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

features
Figure10 shows the confusion matrix of the training and testing image using texture and
Wavelet features.

Figure10 The Confusion matrix of 3 classes with texture and Wavelet features

4.3 Discussion
The following equation show the calculation of the accuracy of the classifier
TP+TN
Accuracy = (1)
TP+ TN + FP+ FN

True Positive (TP) indicates the model predicted an outcome of true, and the actual
observation was true. False Positive (FP) indicates the model predicted a true outcome, but
the actual observation was false. False Negative (FN) indicates the model predicted a false
outcome, while the actual observation was true. Lastly, we have the True Negative (TN),
which indicates the model predicted an outcome of false, while the actual outcome was also
false. The results of the classifier model is shown in Tab. 3.
Table 1: Classification accuracy obtained from different features extraction method

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
16-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

texture features Wavelet features Both

Training 90.2% 88.3% 97.0%
Testing 90.8% 85.8% 95.8%
Overall 90.3 88.0% 96.9%

As shown in table 1 we can see clearly that the SVM classifier with texture features is better
than classification with Wavelet features in both training and testing accuracy . But when
using both Wavelet features and texture features to classification of three classes we got high
accuracy.

References
1. World Health Organization. Archived: WHO Timeline—COVID-19. 2021. Available
online: https://www.who.int/news/item/ 27-04-2020-who-timeline---covid-19
(accessed on 1 August 2021)
2. Nguyen, T.T.; Criss, S.; Dwivedi, P.; Huang, D.; Keralis, J.; Hsu, E.; Phan, L.; Nguyen, L.H.;
Yardi, I.; Glymour, M.M. Exploring US shifts in anti-Asian sentiment with the emergence
of COVID-19. Int. J. Environ. Res. Public Health 2020, 17, 7032.
3. Cui, J., Li, F., & Shi, Z. L. (2019). Origin and evolution of pathogenic coronaviruses. Nature
Reviews Microbiology, 17(3), 181-192.
4. Momattin, H.; Al-Ali, A.Y.; Al-Tawfiq, J.A. A Systematic Review of therapeutic agents for
the treatment of the Middle East Respiratory Syndrome Coronavirus (MERS-CoV).
Travel Med. Infect. Dis. 2019, 30, 9–18

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
17-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

5. Zhai, P., Ding, Y., Wu, X., Long, J., Zhong, Y., & Li, Y. (2020). The epidemiology, diagnosis
and treatment of COVID-19. International journal of antimicrobial agents, 55(5), 105955.
6. Li, Y., Yao, L., Li, J., Chen, L., Song, Y., Cai, Z., & Yang, C. (2020). Stability issues of RT‐
PCR testing of SARS‐CoV‐2 for hospitalized patients clinically diagnosed with COVID‐
19. Journal of medical virology, 92(7), 903-908.
7. Duran-Lopez, L., Dominguez-Morales, J. P., Corral-Jaime, J., Vicente-Diaz, S., & Linares-
Barranco, A. (2020). COVID-XNet: A custom deep learning system to diagnose and locate
COVID-19 in chest X-ray images. Applied Sciences, 10(16), 5683.
8. Jacobi, A., Chung, M., Bernheim, A., & Eber, C. (2020). Portable chest X-ray in coronavirus
disease-19 (COVID-19): A pictorial review. Clinical imaging, 64, 35-42.
9. Ratnapalan, S., Bentur, Y., & Koren, G. (2008). Doctor, will that x-ray harm my unborn
child?. Cmaj, 179(12), 1293-1296.
10. Wang, Z., Xiao, Y., Li, Y., Zhang, J., Lu, F., Hou, M., & Liu, X. (2021). Automatically
discriminating and localizing COVID-19 from community-acquired pneumonia on chest X-
rays. Pattern recognition, 110, 107613.
11. Owida, H. A., Al-Ghraibah, A., & Altayeb, M. (2021). Classification of Chest X-Ray Images
using Wavelet and MFCC Features and Support Vector Machine Classifier. Engineering,
Technology & Applied Science Research, 11(4), 7296-7301.
12. Abbas, A.; Abdelsamea, M.M.; Medhat Gaber, M. Classification of COVID-19 in Chest X-ray
Images Using DeTraC Deep Convolutional Neural Network. Appl. Intell. 2021, 51, 854–
864
13. Wang, L., Lin, Z. Q., & Wong, A. (2020). Covid-net: A tailored deep convolutional neural
network design for detection of covid-19 cases from chest x-ray images. Scientific
Reports, 10(1), 1-12.
14. Apostolopoulos, I. D., & Mpesiana, T. A. (2020). Covid-19: automatic detection from x-ray
images utilizing transfer learning with convolutional neural networks. Physical and
Engineering Sciences in Medicine, 43(2), 635-640.
15. Alquran, H., Alsleti, M., Alsharif, R., Qasmieh, I. A., Alqudah, A. M., & Harun, N. H. B. (2021,
June). Employing Texture Features of Chest X-Ray Images and Machine Learning in COVID-
19 Detection and Classification. In MENDEL (Vol. 27, No. 1, pp. 9-17)
16. Afshar, P., Heidarian, S., Naderkhani, F., Oikonomou, A., Plataniotis, K. N., & Mohammadi, A.
(2020). Covid-caps: A capsule network-based framework for identification of covid-19 cases
from x-ray images. Pattern Recognition Letters, 138, 638-643.

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
18-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

17. Chowdhury, N. K., Rahman, M. M., & Kabir, M. A. (2020). PDCOVIDNet: a parallel-dilated
convolutional neural network architecture for detecting COVID-19 from chest X-ray
images. Health information science and systems, 8(1), 1-14.
18. Duran-Lopez, L., Dominguez-Morales, J. P., Corral-Jaime, J., Vicente-Diaz, S., & Linares-
Barranco, A. (2020). COVID-XNet: A custom deep learning system to diagnose and locate
COVID-19 in chest X-ray images. Applied Sciences, 10(16), 5683.
19. Hussain, E., Hasan, M., Rahman, M. A., Lee, I., Tamanna, T., & Parvez, M. Z. (2021).
CoroDet: A deep learning based classification for COVID-19 detection using chest X-ray
images. Chaos, Solitons & Fractals, 142, 110495
20. Alqudah, A. M., Qazan, S., & Alqudah, A. (2020). Automated systems for detection of COVID-
19 using chest X-ray images and lightweight convolutional neural networks
21. Wang, D., Mo, J., Zhou, G., Xu, L., & Liu, Y. (2020). An efficient mixture of deep and machine
learning models for COVID-19 diagnosis in chest X-ray images. PloS one, 15(11), e0242535.
22. A. ASRAF (2020) . COVID19_Pneumonia_Normal_Chest_Xray_PA_Dataset .online :
https://www.kaggle.com/datasets/amanullahasraf/covid19-pneumonia-normal-chest-xray-pa-
dataset .
23. Olukunle, A., & Ehikioya, S. (2002, May). A fast algorithm for mining association rules in
medical image data. In IEEE CCECE2002. Canadian Conference on Electrical and Computer
Engineering. Conference Proceedings (Cat. No. 02CH37373) (Vol. 2, pp. 1181-1187). IEEE.
24. Wakatani, A. (2002, January). Digital watermarking for ROI medical images by using
compressed signature image. In Proceedings of the 35th annual Hawaii international
conference on system sciences (pp. 2043-2048). IEEE.
25. Jayamanmadharao, M., Chinta Someswara Rao, and DR KVVS REDDY.
"IMPLEMENTATION OF OBJECT ORIENTED APPROCH FOR NOISE REDUCTION USING
FUZZY LOGIC." published in M. Jayamanmadharao et al./International Journal of
Engineering Science and Technology (IJEST) ISSN (2011): 0975-5462.
26. Rothe, P. R., and R. V. Kshirsagar. "A study on the method of image preprocessing for
recognition of crop diseases." IJCA Proceedings on International Conference on Benchmarks
in Engineering Science and Technology 2012 ICBEST. No. 3. 2012.
27. Mohanaiah, P., Sathyanarayana, P., & GuruKumar, L. (2013). Image texture feature
extraction using GLCM approach. International journal of scientific and research
publications, 3(5), 1-5.
28. A. Al-Ghraibah, M. Algharibeh, W. Al-Muhtaseb, F. Al-Khateeb, and I. Al-Anis,
"Investigating the Significance of New Features Extracted from Long Bones X-ray

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
19-xxii
 STUDENT
Al-Ahliyya Amman University VERSION)
‫جامعة عمان األهلية‬
Faculty of Engineering

Images," in 2020 IEEE 5th Middle East and Africa Conference on Biomedical Engineering
(MECBME), Amman, Jordan, Oct. 2020,
29. Chamasemani, F. F., & Singh, Y. P. (2011, September). Multi-class support vector machine
(SVM) classifiers--an application in hypothyroid detection and classification. In 2011 Sixth
International Conference on Bio-Inspired Computing: Theories and Applications (pp. 351-
356). IEEE.
30. Xu, Jianfeng, Yuanjian Zhang, and Duoqian Miao. "Three-way confusion matrix for
classification: A measure driven view." Information sciences 507 (2020): 772-794.

EP-F24-02-005, Rev. a
Ref.: 08/30/2020 2021
Date: 19/04/2021
20-xxii