An Efficient Approach For Classifying Covid Patients Using Chest Radiographic Pictures

Abstract

In the year 2019-20, the corona virus that is covid-19 shows the diverse effect on human life. It is very difficult to detect the virus in its initial stage. In this paper we proposed the model which classify the radiographic images of chest into covid-19, normal and pneumonia. For doing so, we have made usage of multi-class classification. This paper demonstrated an approach that make use of Deep CNN effectively for automatic detection of this disease. To specify the worthiness of the constructed model we have also applied SVM classifier and Logistic regression classifiers to show the comparison of different techniques. The neural network gives highest classification accuracy of 97.5%.Thus, the model works well for detection covid-19 images in the dataset. The comparison of these three techniques is also performed to evaluate the performance of the model of CNN.