Deep Learning Classification Schemes for the Identification of COVID-19 Infected Patients using Large Chest X-ray Image Dataset

Abstract

The fast global spread of the COVID-19 pandemic highlights the immediate need to develop reliable automated strategies for a robust diagnosis of the illness. Image processing of the chest X-ray and CT-scan images can be considered as a reliable strategy for the identification of the covid-infected patients along with the severity of the infection in their lungs. Such a scheme could ultimately help to classify covid-infected individuals from the normal and/or other similar respiratory syndromes (i.e. pneumonia). This paper analyses the ability of various deep learning strategies such as VGG19, Resnet101, VGG16, and InceptionResNetV2 in the classification of covid-infected patients from pneumonia and normal individuals using an exceptionally large dataset of 15,163 X-ray images, reporting 94% accuracy for the best model. Clinical Relevance—Results indicate the reliable capabilities of the deep learning techniques, such as the InceptionResnetV2 structure, for a robust identification and classification of covid-19-infected individuals using chest X-ray images.