Table 1.

COVID-19 detection performance results on Normal vs. COVID-19 test set when trained on real data; real + synthetic data; and only synthetic data (i.e., only ${\mathcal{G}}_{\mathit{NC}}$ is used for positive class examples in training each model)

Model	Real			Real + ${\mathcal{G}}_{\mathit{NC}}$			Real + ${\mathcal{G}}_{\mathit{NC}}$ + ${\mathcal{G}}_{\mathit{PC}}$			Only Synthetic
	SEN (%) $\uparrow$	FN $\downarrow$	Skew$\downarrow$	SEN (%) $\uparrow$	FN $\downarrow$	Skew$\downarrow$	SEN (%) $\uparrow$	FN $\downarrow$	Skew$\downarrow$	SEN (%) $\uparrow$	FN $\downarrow$
VGG-16	19.56	37	91.87	54.34	21	0.98	63.04	17	0.79	50.00	23
ResNet-50	32.61	31	91.87	41.30	27	0.98	43.47	26	0.79	10.86	41
DenseNet-102	26.08	34	91.87	28.27	33	0.98	34.73	30	0.79	8.69	42
DenseNet-121 + BGT	36.95	29	91.87	45.65	25	0.98	52.17	22	0.79	21.73	36

SEN is short for sensitivity. Boldface numbers indicate the best performance