dataset

The rapid outbreak of COVID-19 due to the novel coronavirus SARS-COV-2 is the biggest issue faced by mankind today. It is important to detect the positive cases as early as possible to prevent the further spread of this pandemic. AI-based X-ray screening is a promising approach for COVID-19 testing in both symptomatic and asymptomatic patients. However, a unique challenge for algorithms is to be able to distinguish between COVID-19 versus other lower respiratory diseases which may look similar in X-ray imagery. We evaluate a Convolutional Neural Network (CNN) model which can accurately detect traces of COVID-19 virus in patients using raw Chest X-ray images as well as disambiguate these patients from those with bacterial Pneumonia. This proposed model is used to give accurate diagnostics for multi-class classification using Transfer Learning.