Deep Learning for Detection of COVID-19 from Chest X-ray Image

Since December 2019, the world has entered a new phase, where the health sector and lifestyle of people has threatened worldwide due to global pandemic COVID-19. Because of this infectious disease, approximately 6 million people have died worldwide, and still, it is increasing. RT-PCR (Reverse Transcription-Polymerase Chain Reaction) is the prime method of testing covid- 19 now, but its pricey cost, lengthy testing result, and several other factors reduce the quality of prediction accurately within the time. In this paper, a deep learning approach is used to detect the Covid-19 disease using a dataset of chest X-ray images. A total 21165 chest X-ray images of 4-categories are taken and applied on the deep learning model ResNet50. The experiments show that it has better accuracy of 97.17% as compared to the existing models. In addition, the data set is also applied to Convolutional Neural Network model which has given the accuracy of 96.68%.