Open Access iconOpen Access



COVID-19 Automatic Detection Using Deep Learning

Yousef Sanajalwe1,2,*, Mohammed Anbar1, Salam Al-E’mari1

1 National Advanced IPv6 Centre of Excellence (Nav6), Universiti Sains Malaysia, 11800 USM, Penang, Malaysia
2 Computer Science Department, Northern Border University (NBU), 9280 NBU, Ar’ar, the Kingdom of Saudi Arabia

* Corresponding Author: Yousef Sanajalwe. Email: email

(This article belongs to this Special Issue: Emergent Topics in Intelligent Computing and Communication Engineering)

Computer Systems Science and Engineering 2021, 39(1), 15-35.


The novel coronavirus disease 2019 (COVID-19) is a pandemic disease that is currently affecting over 200 countries around the world and impacting billions of people. The first step to mitigate and control its spread is to identify and isolate the infected people. But, because of the lack of reverse transcription polymerase chain reaction (RT-CPR) tests, it is important to discover suspected COVID-19 cases as early as possible, such as by scan analysis and chest X-ray by radiologists. However, chest X-ray analysis is relatively time-consuming since it requires more than 15 minutes per case. In this paper, an automated novel detection model of COVID-19 cases is proposed to perform real-time detection of COVID-19 cases. The proposed model consists of three main stages: image segmentation using Harris Hawks optimizer, synthetic image augmentation using an enhanced Wasserstein And Auxiliary Classifier Generative Adversarial Network, and image classification using Conventional Neural Network. Raw chest X-ray images datasets are used to train and test the proposed model. Experiments demonstrate that the proposed model is very efficient in the automatic detection of COVID-19 positive cases. It achieved 99.4% accuracy, 99.15% precision, 99.35% recall, 99.25% F-measure, and 98.5% specificity.


Cite This Article

Y. Sanajalwe, M. Anbar and S. Al-E’mari, "Covid-19 automatic detection using deep learning," Computer Systems Science and Engineering, vol. 39, no.1, pp. 15–35, 2021.

cc 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.
  • 2259


  • 1497


  • 0


Share Link