Table 7 Comparison results of different segmentation algorithms in CHASE,[Key: Best, Second Best].

From: TCDDU-Net: combining transformer and convolutional dual-path decoding U-Net for retinal vessel segmentation

Methods

F1

Acc

AUC

Sp

Se

Yu44

–

–

–

–

–

DUNet45

78.83

96.10

98.04

97.52

81.55

Yan46

–

96.07

97.76

98.06

76.40

U-Net++47

81.34

97.62

98.35

–

–

CTF-Net48

–

–

–

–

–

CcNet49

–

–

–

–

–

Yang30

–

96.32

–

97.76

81.76

Xu28

–

96.85

98.40

97.60

86.19

Ours (w/o background decoder)

78.04 ± 1.84

97.19 ± 0.42

98.43 ± 0.27

98.21 ± 0.47

81.42 ± 3.45

Ours (TCDDU-Net)

78.38 ± 1.68

97.23 ± 0.40

98.50 ± 0.23

98.23 ± 0.45

81.87 ± 2.99

  1. Significant values are in bold and underline.