Mousa Atiyah Mukheef Alghazali - Department of Computer Engineering, Isfahan (Khorasgan) Branch, Islamic Azad University, Isfahan, Iran,
Negar Majma - Department of Computer Engineering Naghshe jahan higher education Institute, Isfahan, Iran,

Coronavirus or COVID-۱۹ for short is a viral illnessbrought on by SARS-CoV-۲, which means severe acuterespiratory syndrome coronavirus ۲. The worldwide economyand health seem to be negatively impacted by the spread ofCOVID-۱۹. An important stage in the fight against COVID-۱۹ isan infected patient's chest X-ray that is positive. Early findingspoint to anomalies in patients' chest X-rays that are indicativewith COVID-۱۹. Studies have demonstrated that the accuracy ofCOVID-۱۹ patient identification using chest X-rays is extremelyoptimistic, which has led to the development of a range of deeplearning algorithms. Convolutional neural networks (CNNs), onekind of deep learning network, need a large quantity of trainingdata. It is challenging to compile a sizable number ofradiographic pictures in such a short period of time due to therecent nature of the epidemic. Therefore, in this study, weconstruct a model called CovidGAN that uses a GenerativeAdversarial Network (GAN) to produce synthetic chest X-ray(CXR) images. We also suggest a hybrid CNN-LSTM network toidentify COVID-۱۹ in x-ray pictures. With the proposed hybridnetwork, classification accuracy is ۹۸.۷۵%.

Automated Detection, Covi-۱۹, GANNetwork, deep neural network.