Abstract
COVID-19 is novel virus from CoronaViruses (Coy) family that target the respiratory system causing range of diseases from simple cold to sever diseases such as Mers-Cov and SARS-Cov. Building an automated diagnosis model to detect infection of this virus from CTX is demanded to reduce its complication. In this research, seven image classification models, three variants of SVM (Linear, Polynomial and RBF) and four pre-trained deep learning models (VGG16, InceptionV3, Xception and ResNet50) are evaluated to detect infection by COVID-19. The effect of data augmentation using CCGAN is also analysed in this research. The performance of denominated models generally improved when evaluated with augmented data then when they evaluated with not augmented one.