Skin Lesion Segmentation on PH2
0.9763DICMCGU-Net
Evaluation results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| MCGU-NetNumber of dense blocks (d)=32020.03 | 0.9763 | 0.8322 | 0.9714 | 0.9537 | 0.9537 | — | — | — | — | — | |
| MCGU-NetNumber of dense blocks (d)=12020.03 | 0.9762 | 0.8727 | 0.9925 | 0.9536 | 0.9536 | — | — | — | — | — | |
| IARS SegNetShort name=SN+SC+RC+AM, Skip Connections=true, Residual Convolutions=true, Attention Mechanism=true2023.10 | 0.9712 | 96.46 | 98.94 | — | 0.9233 | 0.04 | 0.01 | — | — | — | |
| PVT-GDLA2026.03 | 0.9559 | — | — | 0.972 | — | — | — | — | — | — | |
| MFSNetTraining time=46min 37sec2022.03 | 0.954 | 0.995 | 0.997 | — | 0.914 | — | — | — | — | — | |
| SegNet + Skip Connections + Residual ConvolutionsShort name=SN+SC+RC, Skip Connections=true, Residual Convolutions=true2023.10 | 0.9518 | 94.22 | 96.32 | — | 0.9139 | 0.04 | 0.02 | — | — | — | |
| PVT-DiffAttn2026.03 | 0.951 | — | — | 0.9713 | — | — | — | — | — | — | |
| PVT-SelfAttn2026.03 | 0.9508 | — | — | 0.9718 | — | — | — | — | — | — | |
| CENet2025.05 | 0.9504 | — | — | 0.9719 | — | — | — | — | — | — | |
| CENet2026.03 | 0.9504 | — | — | 0.9719 | — | — | — | — | — | — | |
| PVT-LinearAttn2026.03 | 0.9478 | — | — | 0.9668 | — | — | — | — | — | — | |
| DermoSegDiff-B2023.08 | 0.9467 | 0.9308 | 0.9814 | 0.965 | — | — | — | — | — | — | |
| SegNet + Skip ConnectionsShort name=SN+SC, Skip Connections=true2023.10 | 0.9453 | 92.5 | 95.51 | — | 0.8844 | 0.08 | 0.07 | — | — | — | |
| DermoSegDiff-A2023.08 | 0.945 | 0.9296 | 0.981 | 0.9637 | — | — | — | — | — | — | |
| Swin-Unet2023.08 | 0.9449 | 0.941 | 0.9564 | 0.9678 | — | — | — | — | — | — | |
| Swin-UnetBackbone=Swin Transformer2025.05 | 0.9449 | — | — | 0.9678 | — | — | — | — | — | — | |
| Swin-Unet2026.03 | 0.9449 | — | — | 0.9678 | — | — | — | — | — | — | |
| MLFFM-SegDiff2026.06 | 0.9384 | 0.9354 | 0.9697 | 0.9591 | 0.8852 | — | — | — | 0.9384 | 0.9424 | |
| H-vmunetParams (M)=8.97, GFLOPs=0.7422026.04 | 0.9322 | — | — | — | 0.873 | — | — | 10.06 | — | — | |
| Ozturk et al.2022.03 | 0.93 | 0.969 | 0.953 | — | 0.871 | — | — | — | — | — | |
| WTCM-UNetParams (M)=28.74, GFLOPs=3.122026.04 | 0.9289 | — | — | — | 0.8672 | — | — | 14.76 | — | — | |
| SegNetShort name=SN2023.10 | 0.9277 | 90.23 | 94.3 | — | 0.8641 | 0.11 | 0.09 | — | — | — | |
| Bi et al. [10]2022.03 | 0.921 | 0.962 | 0.945 | — | 0.859 | — | — | — | — | — | |
| DeepLabv3+2023.08 | 0.9202 | 0.8818 | 0.9832 | 0.9503 | — | — | — | — | — | — | |
| DeepLabv3+2025.05 | 0.9202 | — | — | 0.9503 | — | — | — | — | — | — | |
| DeepLabv3+2026.03 | 0.9202 | — | — | 0.9503 | — | — | — | — | — | — | |
| DAGAN2023.08 | 0.9201 | 0.832 | 0.964 | 0.9425 | — | — | — | — | — | — | |
| Xie et al.2022.03 | 0.919 | 0.963 | 0.942 | — | 0.857 | — | — | — | — | — | |
| UltraLBM-UNet-TVenue (year)=-, Params (M)=0.011, FLOPs (G)=0.0192025.12 | 0.9185 | — | — | — | 0.8492 | — | — | — | — | — | |
| Al et al.2022.03 | 0.918 | 0.937 | 0.957 | — | 0.848 | — | — | — | — | — | |
| FrCN2020.03 | 0.9177 | 0.9372 | 0.9565 | 0.9508 | 0.8479 | — | — | — | — | — | |
| UltraLBM-UNetVenue (year)=-, Params (M)=0.034, FLOPs (G)=0.062025.12 | 0.9154 | — | — | — | 0.8441 | — | — | — | — | — | |
| Yuan et al.2022.03 | 0.915 | — | — | — | — | — | — | — | — | — | |
| U-KANVenue (year)=AAAI (2025), Params (M)=25.359, FLOPs (G)=6.8892025.12 | 0.9131 | — | — | — | 0.8402 | — | — | — | — | — | |
| MK-UNetVenue (year)=ICCVW (2025), Params (M)=0.316, FLOPs (G)=0.3282025.12 | 0.9122 | — | — | — | 0.8385 | — | — | — | — | — | |
| MALUNetVenue (year)=BIBM (2022), Params (M)=0.178, FLOPs (G)=0.0832025.12 | 0.912 | — | — | — | 0.8383 | — | — | — | — | — | |
| EnsDiffbaseline=true2023.08 | 0.9117 | 0.8752 | 0.9774 | 0.9431 | — | — | — | — | — | — | |
| UNeXtVenue (year)=MICCAI (2022), Params (M)=1.472, FLOPs (G)=0.5732025.12 | 0.9102 | — | — | — | 0.8352 | — | — | — | — | — | |
| UCTransNet2023.08 | 0.9093 | 0.9698 | 0.8835 | 0.9408 | — | — | — | — | — | — | |
| UCTransNet2025.05 | 0.9093 | — | — | 0.9408 | — | — | — | — | — | — | |
| EGE-UNetVenue (year)=MICCAI (2023), Params (M)=0.053, FLOPs (G)=0.0722025.12 | 0.9093 | — | — | — | 0.8336 | — | — | — | — | — | |
| UCTransNet2026.03 | 0.9093 | — | — | 0.9408 | — | — | — | — | — | — | |
| UltraLight VM-UNetVenue (year)=Patterns (2024), Params (M)=0.045, FLOPs (G)=0.0692025.12 | 0.9089 | — | — | — | 0.8331 | — | — | — | — | — | |
| Attention U-NetVenue (year)=MIDL (2018), Params (M)=34.879, FLOPs (G)=66.6322025.12 | 0.9084 | — | — | — | 0.8322 | — | — | — | — | — | |
| Double U-NetTraining time=1hr 2min 12sec2022.03 | 0.907 | 0.945 | 0.966 | — | 0.899 | — | — | — | — | — | |
| Goyal et al.2022.03 | 0.907 | 0.932 | 0.929 | — | 0.839 | — | — | — | — | — | |
| Bi et al. [11]2022.03 | 0.907 | 0.949 | 0.94 | — | 0.84 | — | — | — | — | — | |
| MambaU-LiteVenue (year)=ICISN (2025), Params (M)=0.416, FLOPs (G)=0.9342025.12 | 0.9066 | — | — | — | 0.8293 | — | — | — | — | — | |
| CMUNeXt-SVenue (year)=ISBI (2024), Params (M)=0.418, FLOPs (G)=1.092025.12 | 0.9064 | — | — | — | 0.8289 | — | — | — | — | — | |
| U-LiteVenue (year)=APSIPA (2023), Params (M)=0.878, FLOPs (G)=0.7572025.12 | 0.9059 | — | — | — | 0.8282 | — | — | — | — | — | |
| UNet++Venue (year)=DLMIA (2018), Params (M)=9.163, FLOPs (G)=34.9032025.12 | 0.905 | — | — | — | 0.8265 | — | — | — | — | — | |
| TinyU-NetVenue (year)=MICCAI (2024), Params (M)=0.481, FLOPs (G)=1.6592025.12 | 0.904 | — | — | — | 0.8248 | — | — | — | — | — | |
| Rolling-UNet-SVenue (year)=AAAI (2024), Params (M)=1.783, FLOPs (G)=2.1022025.12 | 0.9009 | — | — | — | 0.8199 | — | — | — | — | — | |
| U-NetVenue (year)=MICCAI (2015), Params (M)=7.766, FLOPs (G)=13.7462025.12 | 0.9005 | — | — | — | 0.8191 | — | — | — | — | — | |
| Att-UNet2023.08 | 0.9003 | 0.9205 | 0.964 | 0.9276 | — | — | — | — | — | — | |
| Att-UNet2025.05 | 0.9003 | — | — | 0.9276 | — | — | — | — | — | — | |
| Att-UNet2026.03 | 0.9003 | — | — | 0.9276 | — | — | — | — | — | — | |
| MambaLiteUNetTraining dataset=ISIC2018, Fine-tuning=none2026.04 | 0.8993 | 0.9426 | 0.9268 | 0.9319 | 0.8171 | — | — | 15.58 | — | — | |
| ULVM-UNetTraining dataset=ISIC2018, Fine-tuning=none2026.04 | 0.8972 | 0.9518 | 0.919 | 0.9296 | 0.8135 | — | — | 17.07 | — | — | |
| LB-UNetTraining dataset=ISIC2018, Fine-tuning=none2026.04 | 0.8961 | 0.9476 | 0.9203 | 0.9291 | 0.8117 | — | — | 17.38 | — | — | |
| EGE-UNetTraining dataset=ISIC2018, Fine-tuning=none2026.04 | 0.8957 | 0.9207 | 0.9356 | 0.9308 | 0.8111 | — | — | 17.36 | — | — | |
| LightM-UNetTraining dataset=ISIC2018, Fine-tuning=none2026.04 | 0.8956 | 0.9352 | 0.9271 | 0.9297 | 0.811 | — | — | 16.63 | — | — | |
| VM-UNet2Training dataset=ISIC2018, Fine-tuning=none2026.04 | 0.8947 | 0.9525 | 0.9158 | 0.9276 | 0.8094 | — | — | 17.76 | — | — | |
| SegNetTraining time=58min 21 sec2022.03 | 0.894 | 0.865 | 0.966 | — | 0.808 | — | — | — | — | — | |
| SegNet2020.03 | 0.8936 | 0.8653 | 0.9661 | 0.9336 | 0.8077 | — | — | — | — | — | |
| U-Net2023.08 | 0.8936 | 0.9125 | 0.9588 | 0.9233 | — | — | — | — | — | — | |
| U-Net2025.05 | 0.8936 | — | — | 0.9233 | — | — | — | — | — | — | |
| U-Net2026.03 | 0.8936 | — | — | 0.9233 | — | — | — | — | — | — | |
| VM-UNetTraining dataset=ISIC2018, Fine-tuning=none2026.04 | 0.8935 | 0.9374 | 0.9234 | 0.9279 | 0.8075 | — | — | 17.12 | — | — | |
| UNeXt-STraining dataset=ISIC2018, Fine-tuning=none2026.04 | 0.8932 | 0.9452 | 0.9185 | 0.9271 | 0.807 | — | — | 18.42 | — | — | |
| TransFuseTraining dataset=ISIC2018, Fine-tuning=none2026.04 | 0.8923 | 0.9529 | 0.9129 | 0.9258 | 0.8056 | — | — | 18.7 | — | — | |
| FCN2020.03 | 0.8903 | 0.903 | 0.9402 | 0.9282 | 0.8022 | — | — | — | — | — | |
| U-Net2026.06 | 0.89 | 0.857 | 0.9679 | 0.9313 | 0.8022 | — | — | — | 0.89 | 0.9282 | |
| UTNetV2Training dataset=ISIC2018, Fine-tuning=none2026.04 | 0.8885 | 0.9023 | 0.9386 | 0.9269 | 0.7994 | — | — | 18.82 | — | — | |
| MALUNetTraining dataset=ISIC2018, Fine-tuning=none2026.04 | 0.8881 | 0.9274 | 0.9232 | 0.9246 | 0.7987 | — | — | 19.62 | — | — | |
| C2SDGTraining dataset=ISIC2018, Fine-tuning=none2026.04 | 0.8879 | 0.9454 | 0.9122 | 0.9229 | 0.7983 | — | — | 21.53 | — | — | |
| TransUNet2023.08 | 0.884 | 0.9063 | 0.9427 | 0.92 | — | — | — | — | — | — | |
| TransUNet2025.05 | 0.884 | — | — | 0.92 | — | — | — | — | — | — | |
| TransUNet2026.03 | 0.884 | — | — | 0.92 | — | — | — | — | — | — | |
| SCR-NetTraining dataset=ISIC2018, Fine-tuning=none2026.04 | 0.8823 | 0.9094 | 0.9276 | 0.9217 | 0.7893 | — | — | 19.54 | — | — | |
| Unver et al.2022.03 | 0.881 | 0.836 | 0.94 | — | 0.795 | — | — | — | — | — | |
| U-net2020.03 | 0.8761 | 0.8163 | 0.9776 | 0.9255 | 0.7795 | — | — | — | — | — | |
| U-Net2023.10 | 0.8761 | — | — | — | 0.7795 | — | — | — | — | — | |
| U-NetTraining time=30min 54 sec2022.03 | 0.876 | 0.816 | 0.978 | — | 0.78 | — | — | — | — | — | |
| U-NetTraining dataset=ISIC2018, Fine-tuning=none2026.04 | 0.8702 | 0.9463 | 0.8911 | 0.9089 | 0.7702 | — | — | 22.95 | — | — | |
| ASwin U-NetTraining dataset=ISIC2018, Fine-tuning=none2026.04 | 0.8572 | 0.8807 | 0.9171 | 0.9053 | 0.7501 | — | — | 21.76 | — | — | |
| MissFormer2023.08 | 0.855 | 0.9738 | 0.7817 | 0.905 | — | — | — | — | — | — | |
| MissFormer2025.05 | 0.855 | — | — | 0.905 | — | — | — | — | — | — | |
| MissFormer2026.03 | 0.855 | — | — | 0.905 | — | — | — | — | — | — | |
| DermoSegDiff2026.06 | 0.8235 | 0.8252 | 0.9157 | 0.885 | 0.7005 | — | — | — | 0.8235 | 0.8319 | |
| SwinUNETR2026.06 | 0.8116 | 0.8088 | 0.9153 | 0.8789 | 0.6837 | — | — | — | 0.8116 | 0.8181 | |
| Generic U-Net2026.06 | 0.7174 | 0.911 | 0.7061 | 0.7692 | 0.5623 | — | — | — | 0.7174 | 0.5988 | |
| Hasan et al.Training time=35min 08sec2022.03 | — | 0.929 | 0.969 | — | 0.87 | — | — | — | — | — |