Classification of COVID 19 in Chest CT Images using Convolutional Neural Network

Abstract

The single-celled organisms called a virus are the root cause of many harmful diseases that affects both animals and humans. Coronavirus is one type of virus that is usually found in animals but if transmitted to humans can cause a wide range of respiratory diseases. One such disease caused is the COVID-19 diseases which caused a worldwide epidemic since its origin in 2019. The disease which is said to have originated in China had caused more than 252000 deaths worldwide in a few months. The test for the COVID-19 involves analysing the throat swab sample which may take days if not a week and by the time the results come the infection would have spread. Hence there is a need to improve the testing procedure for COVID-19. In this paper, we have come up with an automated Image Analysis technique to diagnose COVID-19 using the chest Computed Tomography images of the chest that uses a Convolutional Neural Network. This developed method has shown very good accuracy and efficiency in recognizing the COVID-19 infected CT images.