Table 4 Evaluation Indicators ± standard deviation of all competing methods.

From: Dual-branch attention network with deep split convolution and multi-dimensional transformers for medical image segmentation

Methods

DICE (%)

IOU (%)

RVD (%)

ASSD (mm)

MSSD (mm)

U

75.87±0.45

63.36±0.52

−8.19±0.39

5.39±0.39

32.86±3.8

U+D

76.34±0.53

64.87±0.45

−7.22±0.33

5.61± 0.34

28.64±3.41

U+D+M

76.53±0.32

65.32±0.41

−6.29±0.26

5.34±0.29

27.16±3.37

U+D+M+MS

78.68±0.24

66.84±0.38

4.58±0.21

4.85±0.37

25.22±2.97

U+D+M+MS+L

79.27±0.16

67.92±0.21

3.57±0.14

4.62±0.28

24.13±2.54

D3T-Net

81.49±0.12

61.42±0.13

1.44±0.10

4.23±0.24

22.31±2.36

Back to article page