Detection of COVID-19 virus using deep learning. (15th March 2022)

• Record Type: Journal Article
• Authors: Mehta, Kewal; Patel, Hritik; Patel, Vraj; Sharma, Ankit K.
• Is Part Of: International journal of computational biology and drug design. Volume 14:Number 6(2021)
• Page Start: 429
• Page End: 446
• Publication Date: 2022-03-15
• Subjects: COVID-19 -- X-ray -- CT-scan -- deep learning -- neural networks -- CNN -- convolutional neural network -- transfer learning; Computational biology -- Periodicals; Drugs -- Design -- Periodicals; 570.285
• Languages: English
• ISSNs: 1756-0756

Abstract

Corona Virus Disease of 2019 (COVID-19) is currently the most threatening and major medical challenge in the world. COVID-19 can be detected using X-ray and CT-scan images of the patient's lungs. With the use of deep learning and neural networks, the process of classifying the patient's CT-scan and X-ray images can be expedited. In this paper, we implemented convolutional neural networks (CNN) for detection of COVID-19 in X-ray and CT-scan images of lungs. Several CNN architectures like VGG16, ResNet-50, Inception-v3, DenseNet 201, Xception, and InceptionResnet-v2 have been implemented and comparative analysis is presented. DenseNet 201 CNN architecture is found to be most accurate in detecting COVID-19 for both X-ray and CT-scan images. The quantitative results suggest promising results for the COVID-19 detection task.