U-Net (e)

Reported on 5 benchmarks across 2 tasks · 1 paper · 2 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Medical5 results

Colorectal Gland Segmentation:onCRAG
Dice· 2015-05-18
0.844
best: 0.892 (PatchCL)
SOTA
U-Net: Convolutional Networks for Biomedical Image Segmentation arXiv:1505.04597
Multi-tissue Nucleus SegmentationonKumar
Hausdorff Distance (mm)· 2015-05-18
47.8
best: 60 (DSF-CNN (C8))
SOTA
U-Net: Convolutional Networks for Biomedical Image Segmentation arXiv:1505.04597
Colorectal Gland Segmentation:onCRAG
Hausdorff Distance (mm)· 2015-05-18
196.9
best: 318.9 (VF-CNN (C4))
U-Net: Convolutional Networks for Biomedical Image Segmentation arXiv:1505.04597
Colorectal Gland Segmentation:onSTARE
AUC· 2015-05-18
0.827
best: 0.835 (U-Net)
U-Net: Convolutional Networks for Biomedical Image Segmentation arXiv:1505.04597
Multi-tissue Nucleus SegmentationonKumar
Dice· 2015-05-18
0.758
best: 0.843 (GC-MHVN)
U-Net: Convolutional Networks for Biomedical Image Segmentation arXiv:1505.04597