COVID-19 Diagnosis Using Machine Learning

Over 4 million individuals have already died as a result of the deadly contagious viral COVID-19 worldwide. The infection can seriously harm the lungs, increasing the chance of fatal health effects. The only way to lower the mortality rate due to this deadly illness and to halt its growth is through early detection. Deep learning has recently come to light as one of the most useful methods for computer aided diagnosis for helping clinicians make correct illness diagnoses. However, deep learning models require a lot of processing, so hardware with TPUs and GPUs is required to execute these models. To create machine learning models that can be used on mobile and peripheral devices, experts are currently working. In this context, the goal of this study is to create a concise Convolution Neural Network-based computer-aided diagnostic system that can be used on devices with limited processing capacity, such as mobile phones and iPads, to identify the presence of the Covid-19 virus in x-ray pictures. On the basis of various assessment parameters, the findings plainly show that the suggested model outperforms other transfer learning models such as Resnet50, Inception, and Xception. According to various evaluation parameters, the findings definitely show that the proposed model outperforms other transfer learning models like Resnet50, Inception, and Xception.