Table 4.

Class-wise Precision, Recall and F1 Score for all the models.

Performance measures	Class	VGG16	DenseNet121	Xception	NASNet	EfficientNet
Precision	COVID	79%	90%	88%	85%	93%
Precision	Normal	79%	90%	88%	85%	94%
Precision	Other	79%	90%	88%	85%	93%
Recall	COVID	76%	90%	88%	87%	93%
Recall	Normal	82%	89%	87%	81%	93%
Recall	Other	79%	90%	89%	88%	91%
F1 Score	COVID	0.78	0.90	0.88	0.86	0.93
F1 Score	Normal	0.80	0.90	0.88	0.83	0.94
F1 Score	Other	0.79	0.90	0.89	0.86	0.93