Classification of COVID-19 using Chest X-ray Images with Deep Learning Techniques-CNN & ResNet-18
Kotra Sai Kirthana
Kotra Sai Kirthana*, Department of Computer Science and Engineering, Gitam University, Visakhapatnam (A.P), India.
Manuscript received on 29 April 2022. | Revised Manuscript received on 05 May 2022. | Manuscript published on 30 May 2022. | PP: 49-52 | Volume-11 Issue-6, May 2022. | Retrieval Number: 100.1/ijitee.F99220511622 | DOI: 10.35940/ijitee.F9922.0511622
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: To classify the covid-19 images as infectious or normal, it has been evident that the chest X-ray is a powerful tool to diagnose due to its crucial characteristics of convenience, inexpensiveness and rapid pace. The work aims to determine covid-19 in the infected patients by training models with the dataset using convolutional neural networks (CNN) and ResNet-18 and to draw comparisons in their performances respectively. To handle the dataset by applying various operations to simplify and to make it ready for training, validation and testing procedures of both the algorithms involved. The accuracies obtained on testing CNN and RESNET-18 models are 96.07% and 96.67% respectively. Hence the objective of the work is achieved and the results are obtained by implementing covid-19 classification using chest x-ray images with CNN and resnet-18 models.
Keywords: Coronavirus (COVID-19) Infection, Deep Learning, Convolutional Neural Network (CNN), Resnet-18, Transfer Learning, Chest X-ray Images.
Scope of the Article: Deep Learning