DETECTION AND ANALYSIS OF COVID-19 X-RAY DATA USING TRANSFER LEARNING

Abstract

-.For this research X-ray image dataset have been used from the patients with Covid19, bacterial pneumonia diseases, and Normal incidents to detect the Covid-19 disease automatically. The research focuses to out beat the performance of pioneering architectures like (CNN)Convolutional Neural Network which are made in modern years for image classification. Importantly, Transfer Learning architecture was utilized. Various anomaly detection is an achievable goal in small medical image datasets using transfer learning, often leads to exceptional results. The datasets were gathered from open sources medical repositories which are available online. The outcomes specify that combining Transfer Learning with X-ray imaging can excerpt substantial biomarkers related to the Covid-19 disease. As we used three algorithms which are VGG16, Resnet50, Convolutional Layer which provides best accuracy obtained being 96.78%, 98.66%, and 96.46%, respectively. Meanwhile all the previous analytic tests now have such high failure rates that they cause concern, the possibility of incorporating X- rays into the diagnostic process has increased. Key Words: COVID-19, Pneumonia, X-ray, Transfer Learning, Coronavirus.