Medical Image Segmentation on HAM10000
0.9471mDSCCENet
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| CENet2025.05 | 0.9471 | — | — | — | 97.04 | |
| DermoSegDiff-B2023.08 | 0.943 | — | 0.9326 | 0.9839 | 97.04 | |
| DermoSegDiff-A2023.08 | 0.9386 | — | 0.9308 | 0.9814 | 96.81 | |
| TransUNet2023.08 | 0.9353 | — | 0.9225 | 0.9851 | 96.49 | |
| TransUNet2025.05 | 0.9353 | — | — | — | 96.49 | |
| UCTransNet2023.08 | 0.9346 | — | 0.9205 | 0.9825 | 96.84 | |
| UCTransNet2025.05 | 0.9346 | — | — | — | 96.84 | |
| EnsDiffbaseline=true2023.08 | 0.9277 | — | 0.9213 | 0.9771 | 96.25 | |
| Att-UNet2023.08 | 0.9268 | — | 0.9403 | 0.9684 | 96.1 | |
| Att-UNet2025.05 | 0.9268 | — | — | — | 96.1 | |
| Swin-Unet2023.08 | 0.9263 | — | 0.9316 | 0.9723 | 96.16 | |
| Swin-UnetBackbone=Swin Transformer2025.05 | 0.9263 | — | — | — | 96.16 | |
| DeepLabv3+2023.08 | 0.9251 | — | 0.9015 | 0.9794 | 96.07 | |
| DeepLabv3+2025.05 | 0.9251 | — | — | — | 96.07 | |
| MissFormer2023.08 | 0.9211 | — | 0.9287 | 0.9725 | 96.21 | |
| MissFormer2025.05 | 0.9211 | — | — | — | 96.21 | |
| U-Net2023.08 | 0.9167 | — | 0.9085 | 0.9738 | 95.67 | |
| U-Net2025.05 | 0.9167 | — | — | — | 95.67 | |
| MFSNetTraining time=9hr 41min 34sec2022.03 | 0.906 | 0.902 | 0.999 | 0.999 | — | |
| Shahin et al.2022.03 | 0.903 | 0.837 | 0.902 | 0.974 | — | |
| Saha et al.2022.03 | 0.891 | 0.819 | 0.824 | 0.981 | — | |
| Bissoto et al.2022.03 | 0.873 | 0.792 | 0.934 | 0.936 | — | |
| Abraham et al.2022.03 | 0.856 | — | — | — | — | |
| Double U-NetTraining time=11hr 21min 53sec2022.03 | 0.843 | 0.812 | 0.861 | 0.845 | — | |
| SegNetTraining time=11hr 04min 10sec2022.03 | 0.816 | 0.821 | 0.867 | 0.854 | — | |
| Ibtehaz et al.2022.03 | 0.803 | — | — | — | — | |
| U-NetTraining time=9hr 05min 31sec2022.03 | 0.781 | 0.774 | 0.799 | 0.802 | — |