Performance Comparison of Different Convolutional Neural Network Models for the Detection of COVID-19

ABSTRACT

Coronavirus, a new virus, has emerged as a pandemic in recent years. Humans are becoming infected with the virus. In the year 2019, the city of Wuhan reported the first ever incidence of coronavirus. Coronavirus-infected people have symptoms that are related to pneumonia, and the virus affects the body’s respiratory organs, making breathing difficult. A real-time reverse transcriptase polymerase chain reaction (RT-PCR) kit is used for diagnosis of disease. Due to a shortage of kits, suspected patients are unable to be treated in a timely manner, which results in spreading of disease. To come up with an alternative, radiologists looked at the changes in radiological imaging like CT scans. The suspected patient’s computed tomography (CT) scan is used to distinguish between a healthy individual and a coronavirus patient using deep learning algorithms. For COVID-19, a lot of deep learning methods have been proposed. The proposed work utilizes CNN architectures like ResNet101, InceptionV3, VGG19, NASNet, and VGG16. Dataset contains 3873 total CT scan images with the class labels as COVID-19 and non-COVID-19. The dataset is divided into train, test, and validation. Accuracy obtained for ResNet101 is 77.42%, InceptionV3 is 78%, VGG19 is 82%, NASNet is 89.51%, and VGG16 is 97.68%, respectively. From the obtained analysis, the results show that the VGG16 architecture gives better accuracy compared to other architectures. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.