Vol.40, No.2, 2022, pp.765-777, doi:10.32604/csse.2022.019757
Learning Patterns from COVID-19 Instances
  • Rehan Ullah Khan*, Waleed Albattah, Suliman Aladhadh, Shabana Habib
Department of Information Technology, College of Computer, Qassim University, Buraydah, Saudi Arabia
* Corresponding Author: Rehan Ullah Khan. Email:
Received 24 April 2021; Accepted 28 May 2021; Issue published 09 September 2021
Coronavirus disease, which resulted from the SARS-CoV-2 virus, has spread worldwide since early 2020 and has been declared a pandemic by the World Health Organization (WHO). Coronavirus disease is also termed COVID-19. It affects the human respiratory system and thus can be traced and tracked from the Chest X-Ray images. Therefore, Chest X-Ray alone may play a vital role in identifying COVID-19 cases. In this paper, we propose a Machine Learning (ML) approach that utilizes the X-Ray images to classify the healthy and affected patients based on the patterns found in these images. The article also explores traditional, and Deep Learning (DL) approaches for COVID-19 patterns from Chest X-Ray images to predict, analyze, and further understand this virus. The experimental evaluation of the proposed approach achieves 97.5% detection performance using the DL model for COVID-19 versus normal cases. In contrast, for COVID-19 versus Pneumonia Virus scenario, we achieve 94.5% accurate detections. Our extensive evaluation in the experimental section guides and helps in the selection of an appropriate model for similar tasks. Thus, the approach can be used for medical usages and is particularly pertinent in detecting COVID-19 positive patients using X-Ray images alone.
Coronavirus; COVID-19; machine learning; deep learning; convolutional neural network
Cite This Article
R. Ullah Khan, W. Albattah, S. Aladhadh and S. Habib, "Learning patterns from covid-19 instances," Computer Systems Science and Engineering, vol. 40, no.2, pp. 765–777, 2022.
This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.