Deep Layer Kernel Sparse Representation Network for the Detection of Heart Valve Ailments from the Time-Frequency Representation of PCG Recordings

Table 2

Classification results obtained for automated detection of HVAs using various classifiers with SCT domain features and hold-out CV.

Classifier	Class	Performance measure								OA (%)
		TP	TN	FP	FN	Precision (%)	Sensitivity (%)	Specificity (%)	-score (%)
RF	N	60	227	0	0					95.66
	MS	60	227	6	0
	MR	49	238	1	11
	AS	59	228	1	1
	MVP	59	228	5	1
KNN	N	60	231	0	0					97.00
	MS	57	234	3	3
	MR	59	232	2	1
	AS	58	233	3	2
	MVP	57	234	1	3
KSRC	N	60	236	0	0					98.66
	MS	59	237	1	1
	MR	60	236	0	0
	AS	58	238	0	2
	MVP	59	237	3	0
DLKSRN	N	60	238	0	0					99.23
	MS	59	239	1	1
	MR	60	238	1	0
	AS	60	238	0	1
	MVP	59	239	1	1