Paper Title
A Novel Approach to COVID 19 Diagnosis from X-Ray Images using Comparison of Neural Networks

Abstract - The Corona virus ailment (COVID-19), a contagious agent, is generated by the SARS-CoV-2 pathogen. It was detected in 2019 and has triggered a respiratory sickness epidemic. COVID-19 is a menacing pathogen that has taken incalculable lives globally and left millions more with deep-rooted chronic issues. The bulk of those affected by the disease will undergo light to medium inhaling distress and will recuperate without requiring medical assistance. On the contrary, many people may become critically ill and need to get themselves hospitalized. It is diagnosed using a laboratory testing process. However, due to limited testing kits and false-negative RT PCR testing, X-ray visuals are the most effective method for detecting COVID 19. They have also been seen to be best suited for the detection of COVID 19, which solves time constraint problems and is non-contact testing. The coronavirus is transmittable and can spread widely. There are numerous neural networks that can detect COVID 19. So, there is a dire need to distinguish between several neural networks for the identification of COVID 19 and choose the best method. In pursuit of the same, we have made an approach to draw some significant differences between the different neural networks by considering a common Covid 19 detection problem. As evidenced by the analytical results presented in paper, CNN and VGG-19 achieved 93.6% and 98.7% accuracy respectively. VGG-19 outperformed CNN. Keywords - Corona Virus, VGG-19, CNN, Chronic Respiratory Illness, COVID 19, X-Ray, Deep Learning.