A Computer-Aided Diagnosis System for Dynamic Contrast-Enhanced MR Images Based on Level Set Segmentation and ReliefF Feature Selection

Table 2

Classification results of the different segmentation methods (leave-half-case-out).

Segmentation method	Classification model	Accuracy (%)	Sensitivity (%)	Specificity (%)
FCM	Fisher	79.5	87.7	57.1
	SVM	74.4	82.5	52.4
FCM-GVF-snake	Fisher	80.8	86.0	66.7
	SVM	82.1	86.0	71.4
CV-level set	Fisher	91.0	96.5	76.2
	SVM	92.3	98.2	76.2