Non-contact advanced method of COVID-19 classification using deep learning with chest x-ray images.

The first appearance of the novel coronavirus (COVID-19) was on December 31st 2019, in the Wuhan City of China. This novel coronavirus (COVID-19) spread rapidly around the world, thus causing a pandemic. The most devastating effect was caused on the daily lives, global economy and public health system. In order to treat the affected patients quickly the most critical step is to detect the positive cases in much advance period of time in order to help prevent further spread of this disease. With the help of the recent findings, it has been found that the radiology imagining techniques contain the salient information about this virus. The advanced Artificial Intelligence (AI) technique coupled with this radiological imaging has found to be helpful for the accurate detection of this novel coronavirus (COVID-19) disease. This is an advanced application which is helpful for the study of this paper. In this study, the new model used for the detection of this coronavirus (COVID-19) by using the raw chest X-ray images automatically. This proposed model provides accurate diagnostics for binary classification (COVID-19 vs normal) and also the multi-class classification (COVID-19 vs Normal vs Pneumonia). The classification accuracy from this proposed model is about 98.09% for binary classes classification and 87.03% for multi-class classification. [ABSTRACT FROM AUTHOR]