Table 8.

Comparison with other existing approaches (unit: %).

Methods	Accuracy	Sensitivity	Specificity	Precision	F1-Score	Dataset
xDNN [10]	97.38	95.53	-	99.16	97.31	SARS-CoV-2
SepNorm + Contrastive [7]	90.83	85.89	-	95.75	90.87	SARS-CoV-2
FGCNet [21]	97.14	97.71	96.56	96.61	97.15	Private dataset
DarkCovidNet [22]	98.08	95.13	95.3	98.03	96.51	Mixed public datasets
Dual-Track Learning [1]	-	86.00	-	89.60	87.80	Private dataset
CGENet (ours)	97.78	97.94	97.65	97.68	97.80	SARS-CoV-2