Table 8.

The results of evaluating the proposed FSS-2019-nCov on CT-1 using different learning paradigms.

Methods	Semi-supervised learning	Supervised learning
DSC$\uparrow$	0.798	0.679
Sens$\uparrow$	0.803	0.744
Spec$\uparrow$	0.986	0.959
${\mathit{S}}_{\mathit{\alpha }}\uparrow$	0.834	0.774
${\mathit{E}}_{\mathit{\varphi }}\uparrow$	0.908	0.803
MAE$\downarrow$	0.065	0.105

$\uparrow$denote ‘higher is better’, $\downarrow$denote ‘lower is better’.