Table 3.

Performance comparison of the proposed hybrid deep learning model and the other existing pre-trained CNN model for each class.

Deep learning models	Class	Precision	recall	f1-score
Vgg16	Normal	98	78	87
	Covid-19	61	94	74
Vgg19	Normal	97	73	83
	Covid-19	55	93	69
EfficientNetB0	Normal	96	70	81
	Covid-19	52	91	66
ResNet50	Normal	98	72	83
	Covid-19	55	95	70
Hybrid deep learning model (max pooling layer)-NN	Normal	99	88	93
	Covid-19	74	98	85
Hybrid deep learning model (average pooling layer)-Naive Bayes	Normal	89	57	70
	Covid-19	40	80	53
Hybrid deep learning model(average pooling layer)-Random Forest	Normal	96	65	78
	Covid-19	49	93	64
Hybrid deep learning model(average pooling layer)-KNN	Normal	95	60	74
	Covid-19	45	91	60
Hybrid deep learning model(average pooling layer)-SVM(RBF)	Normal	99	83	90
	Covid-19	68	97	80
Hybrid deep learning model(average pooling layer)-SVM(sigmoid)	Normal	97	80	88
	Covid-19	62	94	75
Hybrid deep learning model(average pooling layer)-SVM(linear)	Normal	99	90	94
	Covid-19	77	96	86
Hybrid deep learning model (average pooling layer)-NN	Normal	99	90	94
	Covid-19	77	97	86