Fuzzy Clustering Applied to ROI Detection in Helical Thoracic CT Scans with a New Proposal and Variants

Table 3

Results for slices from different studies using 2D and 3D neighborhoods.

			3 × 3		5 × 5		7 × 7
			2D	3D	2D	3D	2D	3D
1	MKSFCM	VP	61.8	61.8	43.6	60	41.8	50.9
		FP	6.9	7.3	3.8	6.3	3.6	5.9
	SKFCM	VP	49.1	63.6	49.1	63.6	60	63.6
		FP	5	34.2	5.2	29.4	5.8	31.3
2	MKSFCM	VP	30.9	49.4	27.2	45.7	23.5	42
		FP	3.3	5.7	3.1	5	2.9	4.4
	SKFCM	VP	34.6	51.9	34.6	51.9	35.8	53.1
		FP	3.7	26.3	3.7	29.6	3.70	31.1
3	MKSFCM	VP	100	95.7	65.2	30.4	47.8	21.7
		FP	4.8	3.6	3.8	1.9	1.8	1.7
	SKFCM	VP	95.7	100	100	100	100	100
		FP	3.5	10.1	8	19.2	14.9	23.4
4	MKSFCM	VP	100	95	95	95	90	95
		FP	9.4	6.3	6	5.5	5	5.3
	SKFCM	VP	100	100	100	100	100	100
		FP	30.8	23.8	32.8	26.6	34.3	41.7
5	MKSFCM	VP	78.9	78.9	78.9	78.9	0	76.3
		FP	13.6	13.7	9.5	13.9	3	14.5
	SKFCM	VP	89.5	89.5	89.5	89.5	89.5	92.1
		FP	20.9	35.2	26.1	38.5	28.7	40.1
6	MKSFCM	VP	96.9	90.70	93.8	88.7	85.6	67
		FP	17.4	6.7	15.8	4.3	15.7	3.7
	SKFCM	VP	97.9	97.3	99	97.9	99	97.9
		FP	31.1	21.3	33.5	27.5	35.3	30.2
7	MKSFCM	VP	43.8	43.8	42.2	42.2	42.2	32.8
		FP	7	6.8	6.2	5.5	5.3	4.7
	SKFCM	VP	43.8	43.8	43.8	43.8	43.8	43.8
		FP	9.7	8.4	15.3	15.9	21	21.2
8	MKSFCM	VP	98.8	97.6	97.6	84.5	100	84.5
		FP	4.6	3.8	3.6	2.6	4.4	2.4
	SKFCM	VP	100	100	100	100	100	100
		FP	6	5.2	8.6	10.1	11.7	14.2
9	MKSFCM	VP	97.8	95.1	92.2	92.1	91.6	91.6
		FP	5.5	1.7	1.1	1.1	0.9	0.9
	SKFCM	VP	97	97.8	100	100	100	100
		FP	2.4	5.9	13	14.7	17.1	19.2