Table 3. Network performances on CT tasks reported in Dice score (%).
| Networks | Average | Aorta | Gallbladder | Kidney (L) | Kidney (R) | Liver | Pancreas | Spleen | Stomach |
|---|---|---|---|---|---|---|---|---|---|
| TransUNet† (35) | 77.48 | 87.23 | 63.13 | 81.87 | 77.02 | 94.08 | 55.86 | 85.08 | 75.62 |
| TransClaw-Unet (47) | 78.09 | 85.87 | 61.38 | 84.83 | 79.36 | 94.28 | 57.65 | 87.74 | 73.55 |
| LeVit-Unet (46) | 78.53 | 87.33 | 62.23 | 84.61 | 80.25 | 93.11 | 59.07 | 88.86 | 72.76 |
| Mixed Transformer (40) | 78.59 | 87.92 | 64.99 | 81.47 | 77.29 | 93.06 | 59.46 | 87.75 | 76.81 |
| Swin-UNet† (37) | 79.13 | 85.47 | 66.53 | 83.28 | 79.61 | 94.29 | 56.58 | 90.66 | 76.60 |
| MISSFormer (39) | 81.96 | 86.99 | 68.65 | 85.21 | 82.00 | 94.41 | 65.67 | 91.92 | 80.81 |
| UNetR (28) | 79.42 | 88.92 | 69.80 | 81.38 | 79.71 | 94.28 | 58.93 | 86.14 | 76.22 |
| VT-UNet† (45) | 79.02 | 88.54 | 70.07 | 84.43 | 87.14 | 94.79 | 71.47 | 91.12 | 82.16 |
| MedFormer (51) | 84.52 | 92.43 | 77.36 | 92.40 | 91.21 | 95.85 | 81.92 | 93.15 | 90.23 |
| Swin UNETR (44) | 85.78 | 92.78 | 76.55 | 85.25 | 89.12 | 96.91 | 77.22 | 88.70 | 79.72 |
| CoTr (32) | 86.33 | 92.10 | 81.47 | 85.33 | 86.41 | 96.87 | 80.20 | 92.21 | 76.08 |
| nnFormer† (33) | 87.40 | 92.04 | 71.09 | 87.64 | 87.34 | 96.53 | 82.49 | 92.91 | 89.17 |
| PHTrans (54) | 88.55 | 92.54 | 80.89 | 85.25 | 91.30 | 97.04 | 83.42 | 91.20 | 86.75 |
| D-Former (31) | 88.83 | 92.12 | 80.09 | 92.60 | 91.91 | 96.99 | 76.67 | 93.78 | 86.44 |
†, the model was pre-trained on ImageNet. Top section reports 2D networks with bottom section reporting 3D networks. CT, computed tomography; L, left; R, right; 2D, two-dimensional; 3D, three-dimensional.