Medical Image Segmentation on ISIC 2016
92.9Dice ScoreMedSAM
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| MedSAMprompt=bounding box, aggregation=mean ± std across degradation types2026.04 | 92.9 | 86.9 | — | — | |
| Robust MedSAMprompt=bounding box, aggregation=mean ± std across degradation types2026.04 | 91 | 84.4 | — | — | |
| SAMprompt=bounding box, aggregation=mean ± std across degradation types2026.04 | 85.1 | 74.7 | — | — | |
| SpectraFlow2026.05 | 0.9298 | 0.8688 | 0.9322 | 0.9274 | |
| UHR-Net2026.04 | 0.929 | 0.876 | — | — | |
| ConDSeg2026.04 | 0.925 | 0.868 | — | — | |
| ConDSeg2026.05 | 0.9224 | 0.8628 | 0.9325 | 0.9066 | |
| BGDiffSeg2026.04 | 0.922 | 0.855 | — | — | |
| DCSAU-Net2026.05 | 0.919 | 0.8531 | 0.9114 | 0.9079 | |
| FAT-Net2026.04 | 0.916 | 0.853 | — | — | |
| DoubleAANet2026.05 | 0.9148 | 0.8514 | 0.9216 | 0.9224 | |
| DCSAU-Net2026.04 | 0.914 | 0.853 | — | — | |
| MSCB-UNet2026.04 | 0.914 | 0.842 | — | — | |
| PraNet2026.05 | 0.9112 | 0.8441 | 0.9074 | 0.9198 | |
| CENet2026.04 | 0.909 | 0.846 | — | — | |
| XBFormer2026.05 | 0.9082 | 0.8446 | 0.9104 | 0.9088 | |
| U-Net2026.05 | 0.9032 | 0.8361 | 0.8843 | 0.8945 | |
| U-Net2026.04 | 0.903 | 0.836 | — | — | |
| Attn U-Net2026.05 | 0.9023 | 0.8341 | 0.8856 | 0.8977 | |
| TGANet2026.05 | 0.9012 | 0.8243 | 0.911 | 0.9084 | |
| U-Net++2026.05 | 0.8989 | 0.8354 | 0.8976 | 0.9042 | |
| CASF-Net2026.05 | 0.8866 | 0.8145 | 0.8912 | 0.8856 | |
| DTAN2026.05 | 0.8756 | 0.8114 | 0.8941 | 0.8823 |