Medical Image Segmentation On Glas
Métriques
Dice
F1
IoU
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | Dice | F1 | IoU |
|---|---|---|---|
| masked-diffusion-as-self-supervised | 91.95 | 91.95 | 85.13 |
| histoseg-quick-attention-with-multi-loss | - | - | 76.73 |
| hi-gmisnet-generalized-medical-image | 93.25 | 93.25 | - |
| uctransnet-rethinking-the-skip-connections-in | 85.45 | 85.45 | 74.78 |
| uctransnet-rethinking-the-skip-connections-in | 90.18 | 90.18 | 82.96 |
| medical-transformer-gated-axial-attention-for | 76.26 | 76.26 | 63.03 |
| medical-transformer-gated-axial-attention-for | 81.02 | 81.02 | 69.61 |
| medical-transformer-gated-axial-attention-for | 79.68 | 79.68 | 67.69 |
| trans2unet-neural-fusion-for-nuclei-semantic | 89.84 | 89.84 | 82.54 |
| uctransnet-rethinking-the-skip-connections-in | 87.56 | 87.56 | 79.13 |