A Comprehensive Investigation of Machine Learning Feature Extraction and Classification Methods for Automated Diagnosis of COVID-19 Based on X-Ray Images

• DOI: 10.32604/cmc.2021.012874
• Publication Date: 2021-01-04T08:12:18Z
• Authors: Mazin Abed Mohammed; Karrar Hameed Abd...; Begonya Garcia-Za...; Salama A. Mostafa; Mashael S. Maashi; Alaa S. Al-Waisy; Mohammed Ahmed Subhi; Ammar Awad Mutlag; Dac-Nhuong Le
• Keywords: Identification; Kernel (algebra)

The quick spread of the Coronavirus Disease (COVID-19) infection around world considered a real danger for global health. biological structure and symptoms COVID-19 are similar to other viral chest maladies, which makes it challenging big issue improve approaches efficient identification disease. In this study, an automatic prediction is proposed automatically discriminate between healthy infected subjects in X-ray images using two successful moderns traditional machine learning methods (e.g., artificial neural network (ANN), support vector (SVM), linear kernel radial basis function (RBF), k-nearest neighbor (k-NN), Decision Tree (DT), CN 2 rule inducer techniques) deep models MobileNets V2, ResNet50, GoogleNet, DarkNet Xception). A large dataset has been created developed, namely <i>vs.</i> Normal (400 cases, 400 COVID cases). To best our knowledge, currently largest publicly accessible with number confirmed cases. Based on results obtained from experiments, can be concluded that all performed well, had achieved optimum accuracy 98.8% ResNet50 model. comparison, techniques, SVM demonstrated result 95% RBF 94% coronavirus disease 2019.