Prediction Of Covid-19 Cases At The Early Stageusing Deep Learning Method Taken Over CT Scan Images

Abstract

A highly transmissible disease called Corona Virus Disease (COVID-19) causeddue to a syndrome found in respiratory action that is Severe Acute Coronavirus 2(SARS-CoV-2). People of old age, low immunity, medical issues and history of respiratory problems mainly related to lungsbecome more exposed to COVID-19. The best way to save a person who is suffering from Covid-19 is early prediction of the virus and early treatment. The normal testing methods for Covid-19 takes time to produce the result. These tests require a doctor’s diagnosis to determine to extent of the virus that has spread. There is a requirement of early detection of the covid-19 to save as many lives as possible. This proposed work proposes a comparative study between the different algorithms to apply the algorithm that fits the best. The comparative study helps to prove which algorithm gives the best accuracy for the particular problem. The proposed work includes aCOVID-19 prediction system which is automatic using the best fitting algorithm is used in places where experienced doctors and practitioners are unavailable, in this case workers in the field of health care can use this proposed work to prevent further loss of life by early prediction. Hence, a working methodology is proposed that predicts Covid-19 at early stages using a best fitting algorithm for the problem by training it with images of CT scan ofnon-Covid patients and Covid patients.