Medical Image Segmentation on DRIVE
81.63DiceMedCAGD
Evaluation results
| Method | Links | ||
|---|---|---|---|
| MedCAGDParams=30.60 M, Flops=05.00 G2026.07 | 81.63 | — | |
| MCADSParams=50.90 M, Flops=61.89 G2026.07 | 78.42 | — | |
| EMCADParams=26.76 M, Flops=05.60 G2026.07 | 77.15 | — | |
| nnU-NetParams=31.29 M, Flops=55.26 G2026.07 | 75.43 | — | |
| UCTransNetParams=65.60 M, Flops=56.70 G2026.07 | 75.42 | — | |
| PraNetParams=32.55 M, Flops=06.93 G2026.07 | 75.21 | — | |
| TransUNetParams=105.32 M, Flops=38.52 G2026.07 | 74.98 | — | |
| Swin-UnetParams=27.17 M, Flops=06.20 G2026.07 | 74.93 | — | |
| UNeXtParams=1.470 M, Flops=0.570 G2026.07 | 74.77 | — | |
| C2GR2026.06 | 74.29 | — | |
| Swin-UMambaParams=60.00 M, Flops=68.00 G2026.07 | 73.32 | — | |
| VM-UNetParams=27.43 M, Flops=04.12 G2026.07 | 73.25 | — | |
| JointTrain2026.06 | 73.08 | — | |
| CoNuSeg2026.06 | 72.97 | — | |
| UNet++Params=09.16 M, Flops=34.65 G2026.07 | 72.94 | — | |
| GR2026.06 | 72.57 | — | |
| MedSeqFT2026.06 | 71.93 | — | |
| AttnUNetParams=34.88 M, Flops=66.64 G2026.07 | 71.68 | — | |
| FineTune2026.06 | 71.45 | — | |
| U-NetParams=34.53 M, Flops=65.53 G2026.07 | 71.2 | — | |
| Scratch2026.06 | 70.73 | — | |
| DeepLabv3+Params=39.76 M, Flops=14.92 G2026.07 | 69.59 | — | |
| SR-DSFW2026.06 | 65.78 | — | |
| Medical SAM32026.01 | 55.8 | 39.2 | |
| SAM32026.01 | 24.8 | 14.2 | |
| RobustMedSAMPrompt type=point prompts, Prompt count (K)=3, Inference mode=single-image, Input condition=degraded2026.04 | 22.4 | — | |
| RobustSAMPrompt type=point prompts, Prompt count (K)=3, Inference mode=single-image, Input condition=degraded2026.04 | 21.6 | — | |
| SAMPrompt type=point prompts, Prompt count (K)=3, Inference mode=single-image, Input condition=degraded2026.04 | 20.9 | — | |
| RobustMedSAM+SVDPrompt type=point prompts, Prompt count (K)=3, Inference mode=single-image, Input condition=degraded2026.04 | 14.1 | — | |
| MedSAMPrompt type=point prompts, Prompt count (K)=3, Inference mode=single-image, Input condition=degraded2026.04 | 9.3 | — |