Accurate Prediction of COVID-19 (+) Using AI Deep VGG16 Model
Alavikunhu Panthakkan, S. M. Anzar, Saeed Al Mansoori, Hussain Al-Ahmad
Abstract
Current research aims at the efficient prediction of COVID-19 (+) by employing advanced machine intelligence techniques by means of lung X-rays. In this paper, we have presented the promising VGG16 transfer learning model for the accurate and faster diagnosis of COVID-19 (+). The system provides a binary classification of the lung X-ray image into COVID-19 (+) and Normal. The effectiveness of the system being proposed is appraised by means of the performance metrics such as accuracy, precision, recall, and f1 score. Experiments were performed with 2000 X-ray specimens. For the two-class classification of the reported sample size, the proposed VGG16 model provides an outstanding recognition accuracy of 99.5%, which is loftier to all the contemporary methods provided in the literature. The suggested approach is extremely efficient and precise, for that reason, it can be used to aid and support radiologists and healthcare professionals to identify COVID-19 (+) utilizing the lung X-rays.