A Novel Model for Detection and Classification Coronavirus (COVID-19) Based on Chest X-Ray Images Using CNN-CapsNet

Yousra, D.; Abdelhakim, A. B.; Mohamed, B. A.

The coronavirus disease 2019 (COVID-19) first emerged in Wuhan China, and spread across the globe with unprecedented effect. It has became a major healthcare challenge threatening health of billions of humans. Due to absence of therapeutic drugs or vaccines for all, discovering this virus in the early stages will help in diagnosis, evaluation and fast recovery using one and most commonly of the key screening approaches being radiological imaging --Chest X-Ray--. With the advances in Artificial Intelligence algorithms and especially Deep Learning models, and within order to help the radiologists to analyze the vast amount of Chest X-Ray images, which can be crucial for diagnosis and detection of COVID-19. In this paper, we proposed an effective Chest X-Rays image classification model named CNN-CapsNet. Our main idea is to make full use of the merits of these two models: CNN and CapsNet. First, we used a CNN without fully connected layers are used as an initial feature maps extractor. In detail, we have used a pre-trained deep CNN model --VGG19-- that was fully trained on the ImageNet dataset is selected as a feature extractor. Then, we have fed the initial feature maps into a newly designed CapsNet to obtain the final classification result. The performance of our model was evaluated with the four metrics: Accuracy, Sensitivity, Precision and F1 Score. The result is based on the data available in the repositories of Kaggle and Mendeley. Where the experimental results demonstrate that the proposed method can lead to a competitive classification performance compared with the state-of-the-art methods, as achieved high accuracy of 94%. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.