Framework for Classification of Chest X-Rays into Normal/COVID-19 Using Brownian-Mayfly-Algorithm Selected Hybrid Features

Table 5

Investigational outcome of VGG16 with deep and hybrid features.

Feature	Classifier	TP	FN	TN	FP	AC	PR	SE	SP	F1S	NPV
Deep features	SoftMax	224	14	233	9	0.9521	0.9614	0.9412	0.9628	0.9512	0.9433
	DT	227	8	231	14	0.9542	0.9419	0.9660	0.9429	0.9538	0.9665
	RF	228	15	226	11	0.9458	0.9540	0.9383	0.9536	0.9461	0.9378
	NB	230	11	224	15	0.9458	0.9388	0.9544	0.9372	0.9465	0.9532
	KNN	229	7	225	19	0.9458	0.9234	0.9703	0.9221	0.9463	0.9698
	SVM	226	10	232	12	0.9542	0.9496	0.9576	0.9508	0.9536	0.9587
Optimal Deep + HF	SoftMax	236	3	234	7	0.9792	0.9712	0.9874	0.9710	0.9793	0.9873
	DT	238	4	232	6	0.9792	0.9754	0.9835	0.9748	0.9794	0.9831
	RF	240	6	231	3	0.9812	0.9877	0.9756	0.9872	0.9816	0.9747
	NB	239	5	232	4	0.9812	0.9835	0.9795	0.9831	0.9815	0.9789
	KNN	244	0	232	4	0.9917	0.9839	1.0000	0.9831	0.9919	1.0000
	SVM	238	6	231	5	0.9771	0.9794	0.9754	0.9788	0.9774	0.9747