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

Table 2

Results (%) for a subset of slices displaying the greatest problems for the algorithms using spatial information.

		SFCM			SKFCM			MSKFCM
		3 × 3	5 × 5	7 × 7	3 × 3	5 × 5	7 × 7	3 × 3	5 × 5	7 × 7
1	TP	72.0	72.0	71.0	73.0	73.0	66.0	64.0	58.0	56.0
	FP	5.9	5.5	5.1	9.1	9.7	12.6	9.5	9.7	9.4
2	TP	21.6	19.6	19.6	31.4	33.3	33.3	29.4	29.4	29.4
	FP	11.9	11.7	11.6	14.3	17.3	18.5	21.2	21.3	21.1
7	TP	11.4	8.6	5.7	95.7	95.7	94.3	61.4	64.3	60.0
	FP	6.4	5.9	5.4	14.0	12.4	10.3	9.9	10.2	9.7
13	TP	62.9	59.3	55.6	96.3	100.0	96.3	96.3	100.0	92.6
	FP	5.5	4.9	4.6	11.8	11.2	10.2	8.9	10.1	8.9
17	TP	88.2	88.2	85.3	100.0	100.0	100.0	97.1	97.1	97.1
	FP	5.1	4.8	4.6	10.4	10.7	9.8	6.6	6.8	7.0
21	TP	15.5	8.4	5.6	100.0	100.0	100.0	69.0	81.7	81.7
	FP	5.5	4.9	4.5	14.4	11.3	9.0	8.5	9.9	9.4