COVID-19 Classification based on Chest X-Ray Images Using Machine Learning Techniques

Journal of Computer Science and Technology Studies

View Publication Info
 
 
Field Value
 
Title COVID-19 Classification based on Chest X-Ray Images Using Machine Learning Techniques
 
Creator imad, Muhammad
Khan, Naveed
Ullah, Farhat
Abul Hassan, Muhammad
Hussain, Adnan
Faiza
 
Subject COVID-19, Support Vector Machine, K-Nearest Neighbors, Random Forest, Naïve Bayes algorithm, Decision Tree
 
Description The coronavirus (COVID-19) pandemic rapidly spread from the infected person who has a severe health problem around the world. World Health Organization (WHO) has identified the coronavirus as a global pandemic issue. The infected person has a severe respiratory issue that needs to be treated in an intensive health care unit. The detection of COVID-19 using machine learning techniques will help in healthcare system about fast recovery of patients worldwide. One of the crucial steps is to detect these pandemic diseases by predicting whether COVID19 infects the human body or not. The investigation is carried out by analyzing Chest X-ray images to diagnose the patients. In this study, we have presented a method to efficiently classify the   COVID-19 infected patients and normally based on chest X-ray radiography using Machine Learning techniques. The proposed system involves pre-processing, feature extraction, and classification. The image is pre-processed to improve the contrast enhancement. The Histogram of Oriented Gradients (HOG) is used to extract the discriminant features. Finally, In the classification step, five different Machine Learning algorithms, such as Support Vector Machine (SVM), K-Nearest Neighbors, Random Forest, Naïve Bayes algorithm, and Decision Tree) are used to efficiently classify between COVID-19 and normal chest X-ray images. The different metric measures like accuracy, precision, recall, specificity and F1are used to analyze the results. The result evaluation shows that SVM provides the highest accuracy of 96% among the other four classifiers (K-Nearest Neighbors and Random Forest achieved 92% accuracy, 90% accuracy of Naïve Bayes algorithm and 82% accuracy of Decision Tree).
 
Publisher Al-Kindi Center for Research and Development
 
Date 2020-10-06
 
Type info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
Peer-reviewed Article
 
Format application/pdf
 
Identifier https://al-kindipublisher.com/index.php/jcsts/article/view/531
 
Source Journal of Computer Science and Technology Studies; Vol. 2 No. 2 (2020): Journal of Computer Science and Technology Studies; 01-11
2709-104X
 
Language eng
 
Relation https://al-kindipublisher.com/index.php/jcsts/article/view/531/491
 
Rights Copyright (c) 2020 Journal of Computer Science and Technology Studies
 

Contact Us

The PKP Index is an initiative of the Public Knowledge Project.

For PKP Publishing Services please use the PKP|PS contact form.

For support with PKP software we encourage users to consult our wiki for documentation and search our support forums.

For any other correspondence feel free to contact us using the PKP contact form.

Find Us

Twitter

Copyright © 2015-2018 Simon Fraser University Library