Detection of Coronavirus from Chest X-ray Images Using 2D Convolutional Neural Network

The major outbreak caused due to the coronavirus (COVID-19) pandemic is severely endangering the health and life of numerous people globally. Chest X-ray radiography and computer tomography are the two imaging modalities to assess and detect the severity of COVID-19 infections. The study aims to develop a convolutional neural network for chest X-ray images to detect and classify coronavirus infections. In this system, image pre-processing steps are applied to the X-ray image which is then fed to the CNN model to classify the chest X-ray images into two classes of COVID cases and normal cases. To build and test the CNN model, a dataset containing X-ray images is used which is categorized into training and testing sets. The experimental results are depicted in terms of accuracy, specificity, sensitivity, precision, and ROC. The system achieves the desired results on the given dataset, which can be further improved with the availability of the number of COVID-19 images.