Automatic Detection of COVID-19 Using a Stacked Denoising Convolutional Autoencoder

The exponential increase in new coronavirus disease 2019 ({COVID-19}) cases and deaths has made COVID-19 the leading cause of death many countries. Thus, this study, we propose an efficient technique for automatic detection pneumonia based on X-ray images. A stacked denoising convolutional autoencoder (SDCA) model was proposed to classify images into three classes: normal, pneumonia, {COVID-19}. SDCA used obtain a good representation input data extract relevant features from noisy model’s architecture mainly composed eight autoencoders, which were fed two dense layers SoftMax classifiers. evaluated with 6356 datasets different sources. experiments evaluation applied 80/20 training/validation split five cross-validation splitting, respectively. metrics classification accuracy, precision, sensitivity, specificity both schemes. Our results demonstrated superiority classifying high accuracy 96.8%. Therefore, can help physicians accelerate diagnosis.