Multimodal MRI Brain Tumor Image Segmentation Using Sparse Subspace Clustering Algorithm

Table 9

Comparison of segmentation performance of three algorithms under different noise ratios.

Noise ratio	Algorithm	Dice	Jaccard	Precision	Recall
5%	FCM	0.7002	0.5316	0.7010	0.6753
	SVM	0.7912	0.6902	0.8956	0.7412
	SSC	0.8410	0.7367	0.9515	0.7657
10%	FCM	0.6833	0.5241	0.6931	0.6595
	SVM	0.7800	0.6789	0.8763	0.7286
	SSC	0.8362	0.7353	0.9402	0.7542
15%	FCM	0.6658	0.5056	0.6767	0.6323
	SVM	0.7682	0.6574	0.8553	0.7001
	SSC	0.8268	0.7246	0.9362	0.7405
20%	FCM	0.6312	0.4712	0.6420	0.6125
	SVM	0.7404	0.6310	0.8211	0.6803
	SSC	0.8182	0.7131	0.9265	0.7327