A hybrid rexception network for COVID-19 classification from chest X-ray images

Aburaed, Nour and Al-Saad, Mina and Panthakkan, Alavikunhu and al Mansoori, Saeed and Al-Ahmad, Hussain and Marshall, Stephen; (2022) A hybrid rexception network for COVID-19 classification from chest X-ray images. In: 2021 28th IEEE International Conference on Electronics, Circuits, and Systems, (ICECS), 2021. 2021 28th IEEE International Conference on Electronics, Circuits, and Systems, ICECS 2021 - Proceedings . Institute of Electrical and Electronics Engineers Inc., ARE, pp. 1-5. ISBN 9781728182810 (https://doi.org/10.1109/ICECS53924.2021.9665598)

[thumbnail of Aburaed-etal-ICECS-2021-A-hybrid-rexception-network-for-COVID-19-classification-from-chest-X-ray-images]
Preview
 Text. Filename: Aburaed_etal_ICECS_2021_A_hybrid_rexception_network_for_COVID_19_classification_from_chest_X_ray_images.pdf
Accepted Author Manuscript
License: Strathprints license 1.0
Download (1MB)| Preview

Abstract

Nowadays, with the rapid spread of Coronavirus disease (COVID-19) across the globe, the necessity to develop an intelligent system for early diagnosis and detection the COVID-19 infectious disease increases. In recent researches, Chest Xray (CXR) of individual lungs became a common method to identify COVID-19 virus. Manual interpretation of the CXR images can be a lengthy process and subjective to human errors. In this paper, a hybrid Deep Learning model called ReXception is implemented, trained, and evaluated using two types of datasets; Mutliclass and Binary. The network is evaluated based on its overall accuracy, loss, precision, and recall, in addition to the running time and network size. The results show positive indications of the network's performance, especially when compared to other state-of-the-art networks.

CORE (COnnecting REpositories)