Detection of COVID-19 from Chest X-ray images using Deep learning

COVID-19 is the infectious disease caused by the last corona virus that was discovered. This new virus and disease was unknown before the outbreak in Wuhan, China, in December 2019. COVID-19 is now pandemic and affects many countries around the world. Since COVID-19 is a type of flu, it can be diagnosed using radiology imaging techniques. With rapid development in the field of deep learning, there had been intelligent systems to classify between normal, pneumonia and COVID-19 patients. In this study, we have proposed a new neural network that is a concatenation of VGG16, ResNet-50 and Xception networks. The overall average accuracy of the proposed network for all classes is 98%, and the average accuracy for detecting COVID-19 cases is 100%.