Table 9.

Comparison of the classification performance of the proposed model with COVID-19 classification models in related works.

Models (Params)	External dataset 1 (CNUH)			External dataset 2 (YNU)			External dataset 3 (KNUH)
	AUC (95% CI)			AUC (95% CI)			AUC (95% CI)
	Normal	Others	COVID-19	Normal	Others	COVID-19	Normal	Others	COVID-19
Proposed model (79.402M)	0.968 (0.954–0.981)	0.926 (0.893–0.959)	0.953 (0.935–0.971)	0.973 (0.964–0.983)	0.935 (0.914–0.955)	0.884 (0.861–0.906)	0.961 (0.950-0.972)	0.861 (0.837–0.885)	0.898 (0.878–0.918)
CoroNet (33.969M)	0.772${}^{**}$ (0.731–0.813)	0.760${}^{**}$ (0.689–0.830)	0.834${}^{**}$ (0.796–0.872)	0.803${}^{**}$ (0.774–0.833)	0.708${}^{**}$ (0.667–0.749)	0.812${}^{**}$ (0.780-0.844)	0.701${}^{**}$ (0.669–0.732)	0.753${}^{**}$ (0.723–0.783)	0.847${}^{**}$ (0.823–0.871)
COVIDNet (11.750M)	0.787${}^{**}$ (0.740-0.834)	0.744${}^{**}$ (0.667–0.822)	0.636${}^{**}$ (0.575–0.697)	0.715${}^{**}$ (0.681–0.750)	0.586${}^{**}$ (0.542–0.630)	0.844${}^{*}$ (0.816–0.872)	0.665${}^{**}$ (0.633–0.696)	0.491${}^{**}$ (0.456–0.526)	0.651${}^{**}$ (0.611–0.690)
DarkCOVIDNet (1.164M)	0.749${}^{**}$ (0.697–0.802)	0.708${}^{**}$ (0.633–0.784)	0.843${}^{**}$ (0.784–0.902)	0.952${}^{*}$ (0.938–0.966)	0.898${}^{*}$ (0.873–0.923)	0.901 (0.879–0.922)	0.466${}^{**}$ (0.432–0.499)	0.562${}^{**}$ (0.522–0.602)	0.479${}^{**}$ (0.444–0.514)
DeepCOVID (11.178M)	0.711${}^{**}$ (0.660-0.762)	0.701${}^{**}$ (0.625–0.777)	0.791${}^{**}$ (0.742–0.841)	0.893${}^{**}$ (0.871–0.916)	0.751${}^{**}$ (0.714–0.789)	0.844${}^{*}$ (0.817–0.870)	0.690${}^{**}$ (0.659–0.721)	0.625${}^{**}$ (0.589–0.660)	0.770${}^{**}$ (0.737–0.803)

Note:${}^{*}$, ${}^{**}$ denote the better performance of our model w with statistical significance ($p<0.05$, $p<0.001$). CI: confidence interval.