ABSTRACT
The CORONA Virus Disease (COVID-19) is a respiratory disease caused by Severe Acute Respiratory Syndrome Corona Virus 2 (SARS-CoV-2). Although the RT-PCR is the standard testing method, the use of X-rays can be a beneficial alternative COVID-19 testing method, as they can be used to identify abnormalities in the lungs which are suggestive of COVID-19 and can be used to monitor the disease progression.The goal of this paper is to design and implement a GAN enhanced deep learning based COVID-19 detection framework for automatic Covid-19 detection by classifying Chest Radiographs into three classes (Covid-19, pneumonia and Normal). To achieve this, the application of Generative Adversarial Networks and classical data augmentation techniques to a modified VGG-19 convolutional neural network to produce a framework that provides accurate and precise detection of COVID-19 through chest X-rays is proposed. The proposed framework achieved a 91.3% accuracy, 91.3% precision, 90.4% F1 score and 91% recall. © 2021 IEEE.