Table 9.

Results for the extended test dataset binary classification setting. LB stands for the label balancing usage (PBC in the case for SSDL), using 400 test images. PBC results with no statistical significance (with $p>0.1$) gains over the non-balanced SSDL implementation are written in italic.

COVID-19${}^{-}$	COVID${}^{+}$	SSDL	LB	$n_l=10$		$n_l=15$		$n_l=20$
				$\overline{x}$	$s$	$\overline{x}$	$s$	$\overline{x}$	$s$
50%	50%	No	No	0.671	0.032	0.696	0.037	0.722	0.035
70%	30%	Yes	Yes	0.773	0.055	0.769	0.305	0.761	0.025
		Yes	No	0.651	0.065	0.678	0.075	0.652	0.09
		No	Yes	0.631	0.072	0.622	0.066	0.659	0.05
80%	20%	Yes	Yes	0.785	0.045	0.78	0.035	0.772	0.029
		Yes	No	0.7	0.062	0.667	0.058	0.695	0.053
		No	Yes	0.67	0.036	0.642	0.042	0.726	0.01