COVID-19 diagnosis using X-ray images based on convolutional neural networks

Shalaby, W. A.; Saad, W.; Shokair, M.; Dessouky, M. I.; El-Samie, F. E. A.

Coronavirus (COVID-19) is considered as a viral disease that is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Spreading of COVID-19 will continue to affect health and economics. Chest X-ray and CT imaging techniques are crucial for infected patients in the battle with COVID-19. Recently, Convolutional Neural Network has been considered as a type of deep learning tools, and it can be used for detecting diseases such as COVID-19. This paper introduces an efficient architecture for COVID-19 diagnosis from an X-ray dataset. The proposed architecture starts with image pre-processing using lung segmentation and image resizing. Deep feature extraction is performed using the proposed CNN model and different pre-trained models. The classification process is performed using either a Support Vector Machine (SVM) or a Softmax classifier. Simulation results prove that the proposed model can classify COVID-19 images with high accuracies of 98.7% and 98.5% for SVM and Softmax classifiers, respectively. The performance metrics are the processing time, system complexity, accuracy, sensitivity, confusion matrix, F1 score, precision, Receiver Operating Characteristic (ROC) curve, and specificity. © 2021 IEEE.