Automated Diagnosis of Chest X-Ray for Early Detection of COVID-19 Disease

Table 5

Comparison of the performance of our proposed system with existing system.

Previous studies	Number of class	Technique	Overall accuracy (%)	COVID-19 sensitivity (%)
Apostolopoulos et al. [38]	3 classes	MobileNet v2	92.80	94.00
		VGGNet-19	93.50	86.00
Khan et al. [39]	3 classes	Xception	90.20	89.00
Ibrahim et al. [40]	4 classes	ResNet152V2+Bi-GRU	93.36	92.95
Loey et al. [6]	3 classes	GoogLeNet	81.50	81.80
Wang et al. [41]	3 classes	COVID-Net	93.30	91.00
		VGGNet-19	83.00	58.70
		ResNet-50	90.60	83.00
Muhammad et al. [42]	3 classes	SqueezeNet	84.40	84.30
		ResNet-50	90.00	87.40
Ismael et al. [43]	2 classes	Fine-tuning of ResNet50	92.63	88.00
Proposed model	4 classes	ResNet-50	95.00	97.10
Proposed model	2 classes	ResNet-50	98.00	97.40
Proposed model	4 classes	AlexNet	92.00	94.50
Proposed model	2 classes	AlexNet	93.00	99.30