Skip to main content

Table 3 Quantitative results of different segmentation methods

From: Lesion segmentation in breast ultrasound images using the optimized marked watershed method

  \({ACC \,(\%)}\) \({TPR \,(\%)}\) \({FPR \,(\%)}\) \({DSC\, (\%)}\) \({JI \,(\%)}\) \({AER\, (\%)}\) \(\text {HE}\) \(\text {MAE}\)
MW 94.37
(±0.03)
97.64
(±0.02)
79.17
(±0.05)
71.80
(±0.08)
56.68
(±0.10)
81.53
(±0.41)
69.21
(±26.97)
27.88
(±8.81)
level set 94.69
(±0.03)
95.56
(±0.05)
5.24
(±0.03)
69.58
(±0.13)
54.83
(±0.14)
98.11
(±0.78)
72.54
(±26.35)
28.09
(±9.06)
MS 95.76
(±0.04)
63.05
(±0.13)
11.93
(±0.16)
71.57
(±0.10)
56.72
(±0.12)
48.88
(±0.17)
69.91
(±50.21)
22.68
(±13.67)
AMS 94.38
(±0.05)
81.70
(±0.15)
45.32
(±0.28)
72.69
(±0.09)
58.01
(±0.11)
63.62
(±0.32)
72.72
(±44.87)
24.94
(±14.53)
MS+MW 95.92
(±0.03)
67.65
(±0.04)
15.40
(±0.11)
73.57
(±0.12)
59.14
(±0.14)
47.74
(±0.73)
67.00
(±25.24)
21.75
(±7.13)
AMSMW 96.59
(±0.02)
83.29
(±0.11)
1.94
(±0.21)
78.40
(±0.09)
65.34
(±0.11)
46.45
(±0.22)
51.33
(±28.66)
17.70
(±7.37)
level set+MW 96.25
(±0.03)
91.58
(±0.06)
50.52
(±0.20)
77.47
(±0.09)
64.18
(±0.12)
58.94
(±0.45)
55.85
(±27.01)
19.50
(±7.36)
FSMW 96.19
(±0.03)
92.88
(±0.05)
52.19
(±0.14)
77.53
(±0.09)
64.27
(±0.12)
59.30
(±0.45)
56.08
(±28.61)
19.50
(±7.21)
  1. Bold indicates the best results in the current column
  2. MW: marked watershed [31]; Level set [5]; MS: morphological snake [9]; AMS: adaptive morphological snake; FSMW [30]