 The paper presents a deep learning technique based on convolution neural network, CNN, to detect and diagnose COVID-19 disease using chest x-ray images and proposes a modified AlexNet architecture for different scenarios with varying pooling layers and neurons in the second fully connected layer. The results obtained are comparable or better than the original AlexNet with fewer neurons in the second fully connected layer. This article was authored by Shadman Q. Solly or K. H. Abdullah, Zanier S. H. Ahmed and others.