Prediction of COVID-19 Cases Using CNN with X-rays
D. Haritha, N Swaroop, M. Mounika
Abstract
The Corona Virus Disease popularised as COVID-19 is a highly transmissible viral infection and has severe impact on global health. It impacted the global economy also very badly. Ift positive cases can be detected early, this pandemic disease spread can be curtailed. Prediction of COVID-19 disease is advantageous to identify patients at a risk of health conditions. Applications of Artificial Intelligence (AI) techniques for COVID prediction from X-rays can be very useful, and can help to overcome the shortage of availability of doctors and physicians in remote places. This paper proposes a transfer learning model using Googlenet for COVID-19 prediction from chest X-ray images. For image classification we used GoogleNet which is one of the CNN architecture and is also named as InceptionV1. The positively classified images by our model indicate the presence of COVID-19. The results obtained in COVID prediction using GoogleNet with a training accuracy of 99% and testing accuracy of 98.5% emphasize the use of Transfer Learning models in disease prediction.