3D Shape-Weighted Level Set Method for Breast MRI 3D Tumor Segmentation

Table 3

Accuracy, specificity, and false alarm rate (%) of the different algorithms and noise densities in 5 × 5 mask blurred computer-simulated images.

Noise density		Methods
		ACM	LSM	SLSM	3D-LSM	3D-SLSM
5%	CCR	87.65	98.88	98.94	99.18	99.43
	SP	87.01	98.81	98.88	99.13	99.9
	FAR	13.08	1.19	1.13	0.87	0.6
10%	CCR	89.85	99.01	99.39	99.36	99.38
	SP	89.37	98.96	99.72	99.34	99.9
	FAR	10.74	1.04	0.65	0.68	0.66
15%	CCR	90.57	99.07	99.34	99.28	99.35
	SP	90.13	99.02	99.7	99.25	99.92
	FAR	9.98	0.99	0.7	0.77	0.68
20%	CCR	90.37	99.11	99.25	99.29	99.39
	SP	89.93	99.07	99.69	99.28	99.86
	FAR	10.2	0.94	0.8	0.75	0.65
25%	CCR	89.71	99.06	99.16	99.38	99.29
	SP	89.26	99.05	99.7	99.42	99.91
	FAR	10.9	0.99	0.89	0.65	0.75
30%	CCR	88.68	99.05	99.14	99.26	99.15
	SP	88.17	99.07	99.7	99.3	99.91
	FAR	11.98	1.01	0.92	0.79	0.9
35%	CCR	88.08	99.12	99.01	99.34	98.94
	SP	87.62	99.22	99.73	99.56	99.96
	FAR	12.62	0.93	1.04	0.7	1.12
40%	CCR	89.18	98.96	98.89	99.25	99.31
	SP	88.77	99.14	99.68	99.45	99.81
	FAR	11.46	1.1	1.17	0.8	0.73
45%	CCR	89.34	98.91	98.8	99.24	99.27
	SP	88.94	99.14	99.72	99.63	99.74
	FAR	11.29	1.16	1.27	0.8	0.77
50%	CCR	88	98.8	98.55	99.01	99.08
	SP	87.55	99.21	99.68	99.46	99.76
	FAR	12.71	1.27	1.53	1.05	0.97