COVID-19 Classification in Chest CT Images Using Deep Convolution Neural Networks




Elgadi, Malek

Journal Title

Journal ISSN

Volume Title



The coronavirus disease (COVID-19) has rapidly spread over the world since the end of 2019. The immediate and accurate diagnosis of COVID-19 is essential for improving the prognosis of this disease and reducing the pandemic spread. Although the PCR test is a standard test to diagnose COVID-19, radiography techniques such as chest X-rays and computed tomography (CT) scans are preferred for detection of COVID-19 disease. Deep learning and convolutional neural Networks (CNNs) play an important role in the early and accurate detection of COVID-19 using radiography images. In this project, a deep convolutional neural network framework based on a transfer learning technique with fine-tuning is suggested for detection and classification of COVID-19. Two pre-trained models i.e., VGG16 and DenseNet201 are trained using COVID-19 CT images dataset. Various experiments are performed to evaluate the performance of the pre-trained models using several evaluation parameters. The results show that the best accuracy of 99.4%, recall of 99.39%, precision of 99.4%, F1-score of 99.39%, and Area Under the Curve (AUC) of 99.93% are achieved by VGG-16 model. DenseNet201 model also shown a competitive result with an accuracy of 99.13 since it has lesser execution time and fewer parameters compared to other deep learning models.