COVID-19 detection through convolutional neural networks and chest X-ray images
by K. Venkata Subbareddy; L. Nirmala Devi
International Journal of Medical Engineering and Informatics (IJMEI), Vol. 14, No. 4, 2022

Abstract: To break the chain of COVID-19, a powerful and fast screening system is required which identifies the COVID-19 affected cases quickly such that the appropriate measures like quarantine or treatment can be taken. The traditional RT-PCR test is found to have larger misclassification rate. It also consumes more time to get the result. To solve this problem, in this paper, we have introduced a new model for COVID-19 detection based on chest X-ray (CXR) images and convolutional neural networks (CNNs). The proposed model is an automatic detection model, which considers the CXR image as input and performs an in-depth analysis to discover the COVID-19. The proposed CNN model is a very simple and effective, which is composed of five convolutional layers and three pooling layers. Every convolutional layer has different sized filters and different number of filters, which extracts all the possible features from CXR image. Simulation experiments are conducted over a newly constructed dataset based on the publicly available CXR (both COVID-19 and non-COVID-19) images. Simulation is done under two phases; 3-class and 2-class and obtained an average accuracy of 92.22% and 94.44% respectively. Thus, the average accuracy is measured as 93.33%.

Online publication date: Tue, 05-Jul-2022

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Medical Engineering and Informatics (IJMEI):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com