Skip to main content

Table 7 Registration results with different number of labels (mean ± std)

From: macJNet: weakly-supervised multimodal image deformable registration using joint learning framework and multi-sampling cascaded MIND

Amount

Liver

Image

TRE (mm)

DSC (%)

Hd95 (mm)

MI (%)

SSIM (%)

0%

4.90 ± 1.48

93.64 ± 1.07

5.39 ± 1.14

43.76 ± 4.83

53.00 ± 11.64

5%

4.93 ± 1.52

94.42 ± 0.85

4.94 ± 1.18

43.98 ± 4.70

53.72 ± 11.53

10%

5.18 ± 1.63

94.50 ± 0.92

4.80 ± 1.12

43.93 ± 4.60

54.41 ± 11.61

20%

5.24 ± 1.75

94.55 ± 0.91

4.63 ± 1.03

44.12 ± 4.58

54.85 ± 11.53

30%

4.83 ± 1.49

94.75 ± 0.82

4.53 ± 1.11

44.12 ± 4.63

54.43 ± 11.62

40%

4.79 ± 1.67

94.64 ± 0.82

4.52 ± 0.88

43.96 ± 4.55

54.22 ± 11.63

50%

5.05 ± 1.38

94.76 ± 0.85

4.46 ± 1.17

44.32 ± 4.64

54.39 ± 11.60

60%

5.10 ± 1.36

94.72 ± 0.89

4.59 ± 1.15

44.30 ± 4.71

54.05 ± 11.71

70%

5.27 ± 1.18

94.69 ± 0.86

4.53 ± 1.12

44.45 ± 4.68

54.24 ± 11.59

80%

5.31 ± 1.36

94.78 ± 0.79

4.40 ± 0.87

44.18 ± 4.72

54.17 ± 11.63

90%

4.97 ± 1.21

94.86 ± 0.89

4.20 ± 1.09

44.20 ± 4.52

53.94 ± 11.62

100%

5.10 ± 1.48

94.91 ± 0.81

4.21 ± 0.91

44.38 ± 4.65

54.09 ± 11.59

  1. Bold values indicate better results than other methods