COVID-19 Diagnosis Using an Enhanced Inception-ResNetV2 Deep Learning Model in CXR Images

Table 6

Model classification results for 3-category (COVID-19 vs. normal vs. viral pneumonia) classification task for the original dataset.

Model	Category	Accuracy (%)	F1-score (%)	Precision (%)	Recall (%)
Inception-ResNetV2	COVID-19	99.83	99.05	98.11	100.00
	Normal	98.80	98.62	98.81	98.43
	Viral pneumonia	98.97	98.91	98.91	98.91
Xception	COVID-19	99.83	98.90	97.83	100.00
	Normal	98.28	98.16	99.63	96.74
	Viral pneumonia	98.45	98.29	97.00	99.62
VGG16	COVID-19	99.48	96.55	100.00	93.33
	Normal	97.59	97.48	96.79	98.19
	Viral pneumonia	98.11	97.88	98.07	97.69
ResNet50V2	COVID-19	99.66	97.83	95.75	100.00
	Normal	97.76	97.62	98.89	96.38
	Viral pneumonia	97.76	97.52	96.60	98.46
InceptionV3	COVID-19	99.83	98.90	97.83	100.00
	Normal	98.11	98.00	98.54	97.46
	Viral pneumonia	97.93	97.70	97.33	98.08
MobileNetV2	COVID-19	99.48	96.55	100.00	93.33
	Normal	98.28	98.21	97.16	99.28
	Viral pneumonia	98.45	98.26	98.83	97.69
DenseNet121	COVID-19	99.83	98.90	97.83	100.00
	Normal	98.28	98.16	99.63	96.74
	Viral pneumonia	98.45	98.29	97.00	99.62
ResNet101V2	COVID-19	99.48	96.63	97.73	95.56
	Normal	97.93	97.83	97.83	97.83
	Viral pneumonia	97.42	97.12	96.94	97.31