Detection and Spread Prediction of COVID-19 from Chest X-ray Images using Convolutional Neural Network-Gaussian Mixture Model

Abstract

In this paper, we diagnose the existence of novel coronavirus disease 2019 (COVID-19) using the chest X-ray images of patients. We perform a multi-class classification of the chest X-ray images of COVID-19 infected patients, other patients suffering from bacterial pneumonia, and healthy persons using convolutional neural network (CNN). Further, we enhance our data for the prediction task using Monte-Carlo simulation on the original data distributions, comprising the confirmed and death cases of the COVID-19 patients. Additionally, for the prediction of COVID-19 pandemic, we use linear regression of the components of Gaussian mixture model (GMM). Using the chest X-ray pneumonia dataset from Kaggle and the University of Montreal, we obtain training and testing classification accuracies of 100% and 96.66% respectively using our CNN model. Further, we obtain the linear regression equations that predict the COVID-19 spread from the GMM.