In recent years, we have globally impacted with the COVID-19 which was very uncommon and highly contagious and dangerous respiratory illness, demanding early detection for effective containment and further spread. In this research we proposed an innovative methodology that utilizes images of X-Ray for early prediction and early detection of COVID9. By employing transfer learning with a pretrained neural network, coupled with the Nearest-Neighbour method for optimization and processing of data, the methodology achieves a remarkable accuracy of 96.37%. This accuracy surpasses other available model based on different deep learning models, like ResNet-50, AlexNet & VGG for COVID-19 identification & detection using X-ray images. Use of X-ray images are now preferably used modality for identification & detection of COVID-19, given its widespread utilization and effectiveness. However, manual treatment & examination using X-ray images is very challenging, specifically in the field which is facing a limitation of skilled medical staff. The application of deep learning models has shown promise and effective results in automating the diagnosis for early identification of COVID-19 with help of images of X-ray. Overall aim of our study is to design a computer aided convolutional neural network for early detection and identification of COVID-19. The suggested architecture, Covid-R, is specifically developed for prediction and analysis for COVID-19 cases using CT images & X-ray images. We trained our model with publicly available and online COVID-19 images dataset and images of X-rays directly collected from hospitals, that includes and are mixed images of patients having COVID & non-COVID symptoms. It firmly believes that this study holds significant potential in alleviating the workload of frontline radiologists, expediting patient diagnosis and treatment, and facilitating pandemic control efforts.

Covid-R, Pneumonia, X-Ray image