Deep learning framework for rapid and accurate respiratory COVID-19 prediction using chest X-ray images

Chiagoziem C Ukwuoma; Dongsheng Cai; Md Belal Bin Heyat; Olusola Bamisile; Humphrey Adun; Zaid Al-Huda; Mugahed A Al-Antari

doi:10.1016/j.jksuci.2023.101596

Deep learning framework for rapid and accurate respiratory COVID-19 prediction using chest X-ray images

J King Saud Univ Comput Inf Sci. 2023 Jul;35(7):101596. doi: 10.1016/j.jksuci.2023.101596. Epub 2023 May 25.

Authors

Chiagoziem C Ukwuoma¹, Dongsheng Cai¹, Md Belal Bin Heyat², Olusola Bamisile³, Humphrey Adun⁴, Zaid Al-Huda⁵, Mugahed A Al-Antari⁶

Affiliations

¹ The College of Nuclear Technology and Automation Engineering, Chengdu University of Technology, Sichuan, 610059, China.
² IoT Research Center, College of Computer Science and Software Engineering, Shenzhen University, Shenzhen, Guangdong 518060, China.
³ Sichuan Industrial Internet Intelligent Monitoring and Application Engineering Technology Research Center, Chengdu University of Technology, China.
⁴ Department of Mechanical and Energy Systems Engineering, Cyprus International University, Nicosia, North Nicosia, Cyprus.
⁵ School of Computing and Artificial Intelligence, Southwest Jiaotong University, Chengdu, Sichuan, China.
⁶ Department of Artificial Intelligence, College of Software & Convergence Technology, Daeyang AI Center, Sejong University, Seoul 05006, Republic of Korea.

Abstract

COVID-19 is a contagious disease that affects the human respiratory system. Infected individuals may develop serious illnesses, and complications may result in death. Using medical images to detect COVID-19 from essentially identical thoracic anomalies is challenging because it is time-consuming, laborious, and prone to human error. This study proposes an end-to-end deep-learning framework based on deep feature concatenation and a Multi-head Self-attention network. Feature concatenation involves fine-tuning the pre-trained backbone models of DenseNet, VGG-16, and InceptionV3, which are trained on a large-scale ImageNet, whereas a Multi-head Self-attention network is adopted for performance gain. End-to-end training and evaluation procedures are conducted using the COVID-19_Radiography_Dataset for binary and multi-classification scenarios. The proposed model achieved overall accuracies (96.33% and 98.67%) and F1_scores (92.68% and 98.67%) for multi and binary classification scenarios, respectively. In addition, this study highlights the difference in accuracy (98.0% vs. 96.33%) and F_1 score (97.34% vs. 95.10%) when compared with feature concatenation against the highest individual model performance. Furthermore, a virtual representation of the saliency maps of the employed attention mechanism focusing on the abnormal regions is presented using explainable artificial intelligence (XAI) technology. The proposed framework provided better COVID-19 prediction results outperforming other recent deep learning models using the same dataset.

Keywords: COVID-19 prediction; Explainable Artificial Intelligence (XAI); Hybrid ensemble deep feature extraction; Multi-head self-attention network; Visual explainable saliency maps.