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 (%)	F-score (%)
RF	N	60	227	0	0	97.75 ± 2.63	98.33 ± 2.35	99.38 ± 0.74	98.00 ± 1.50	95.66
	MS	60	227	6	0	88.60 ± 3.77	96.99 ± 1.39	96.72 ± 1.30	92.56 ± 1.90
	MR	49	238	1	11	96.31 ± 1.87	87.99 ± 4.62	99.13 ± 0.42	91.94 ± 3.20
	AS	59	228	1	1	92.38 ± 3.59	95.33 ± 1.82	97.90 ± 1.04	93.79 ± 1.63
	MVP	59	228	5	1	98.22 ± 1.24	93.32 ± 4.56	99.56 ± 0.31	95.67 ± 2.79
KNN	N	60	231	0	0	98.69 ± 1.35	99.66 ± 0.74	99.64 ± 0.37	99.17 ± 0.82	97.00
	MS	57	234	3	3	90.16 ± 1.65	97.66 ± 0.91	97.28 ± 0.49	93.76 ± 1.28
	MR	59	232	2	1	97.91 ± 1.46	93.99 ± 3.45	99.48 ± 0.35	95.89 ± 2.19
	AS	58	233	3	2	98.24 ± 2.11	91.99 ± 3.20	99.57 ± 0.52	94.99 ± 2.20
	MVP	57	234	1	3	96.41 ± 2.56	97.33 ± 1.90	99.04 ± 0.71	96.85 ± 1.76
KSRC	N	60	236	0	0	100.0 ± 0.00	100.0 ± 0.00	100.0 ± 0.00	100.0 ± 0.00	98.66
	MS	59	237	1	1	96.13 ± 2.12	98.99 ± 0.91	98.98 ± 0.56	97.54 ± 1.52
	MR	60	236	0	0	98.66 ± 2.17	97.66 ± 1.90	99.66 ± 0.55	98.15 ± 1.89
	AS	58	238	0	2	99.00 ± 1.49	96.99 ± 2.47	99.74 ± 0.38	97.96 ± 1.29
	MVP	59	237	3	0	96.70 ± 1.99	96.66 ± 1.17	99.15 ± 0.51	96.66 ± 1.01
DLKSRN	N	60	238	0	0	100.0 ± 0.00	100.0 ± 0.00	100.0 ± 0.00	100.0 ± 0.00	99.23
	MS	59	239	1	1	98.24 ± 1.02	99.01 ± 0.10	99.00 ± 0.63	99.13 ± 0.14
	MR	60	238	1	0	99.02 ± 0.23	98.76 ± 1.25	99.88 ± 0.16	99.00 ± 0.26
	AS	60	238	0	1	99.18 ± 0.64	97.95 ± 1.38	99.82 ± 0.01	98.72 ± 1.33
	MVP	59	239	1	1	96.89 ± 1.02	97.36 ± 1.29	99.56 ± 0.11	97.51 ± 1.82