Table 1 Comparison with state-of-the-art methods on the ACDC dataset.

From: Multi-perspective dynamic consistency learning for semi-supervised medical image segmentation

Method

#Scans

Metrics

Labeled

Unlabeled

Dice (%)↑

Jaccard (%)↑

95HD (voxel)↓

ASD (voxel)↓

SupOnly

7 (10%)

63 (90%)

79.37

67.78

10.73

3.12

URPC18

 

83.10

72.41

4.84

1.53

MCNet22

86.44

77.04

5.50

1.84

SSNet50

86.78

77.67

6.07

1.40

DCNet23

89.42

81.37

1.28

0.38

MPDC (Ours)

90.24

82.70

1.21

0.31

Back to article page