Table 1 Comparisons with state-of-the-art models on CVC-ClinicDB dataset.
From: Multi-scale conv-attention U-Net for medical image segmentation
Method | DSC | IoU |
|---|---|---|
U-net5 | 88.42 ± 0.32 | 81.22 ± 0.54 |
Attention U-Net7 | 88.44 ± 0.04 | 81.32 ± 0.18 |
DualNet19 | 85.74 ± 0.15 | 77.19 ± 0.10 |
MedT26 | 67.56 ± 0.03 | 55.76 ± 0.18 |
TransUNet22 | 89.07 ± 0.03 | 81.75 ± 0.18 |
Unext12 | 79.30 ± 0.08 | 69.39 ± 0.19 |
EGE-UNet8 | 80.86 ± 0.45 | 71.39 ± 0.77 |
WRANet13 | 89.14 ± 0.08 | 82.08 ± 0.09 |
SeTformer32 | 89.43 ± 0.10 | 82.31 ± 0.05 |
Ours | 89.77 ± 0.24 | 82.87 ± 0.13 |
