A Novel Method for COVID-19 Detection Based on DCNNs and Hierarchical Structure

Table 2

Sensitivity, specificity, precision, -score, and accuracy values of different models.

Method	Performance metrics
	Sensitivity	Specificity	Precision	-score	Accuracy
Single ResNet	0.9289	0.9365	0.9497	0.9348	0.9351
Single DenseNet	0.8990	0.9453	0.8900	0.8945	0.9300
The proposed method	0.9900	1.0000	1.0000	0.9950	0.9967