Codesota · Models · nnU-Net v2German Cancer Research Center (DKFZ)4 results · 4 benchmarks
Model card

nnU-Net v2.

German Cancer Research Center (DKFZ)open-sourcevaries (~31M default) paramsU-Net (self-configuring)

Self-configuring U-Net framework. De facto baseline for medical image segmentation. v2 adds improved preprocessing and more architectures.

§ 01 · Benchmarks

Every benchmark nnU-Net v2 has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01ACDCMedical · Medical Image Segmentationmean-dsc91.6%#2/62023-03-17source ↗
02BraTS 2023Medical · Medical Image Segmentationmean-dice-wt-tc-et0.9%#3/32024-06-11source ↗
03BTCVMedical · Medical Image Segmentationmean-dsc81.8%#3/62023-04-13source ↗
04Synapse Multi-Organ CTMedical · Medical Image Segmentationmean-dsc82.5%#7/112023-03-17source ↗
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 02 · Strengths by area

Where nnU-Net v2 actually performs.

Medical
4
benchmarks
avg rank #3.8
§ 03 · Papers

3 papers with results for nnU-Net v2.

  1. 2024-06-11· Medical· 1 result

    nnU-Net for BraTS 2023: Adaptation of nnU-Net for BraTS 2023

  2. 2023-04-13· Medical· 1 result

    STU-Net: Scalable and Transferable Medical Image Segmentation Models Empowered by Large-Scale Supervised Pre-training

  3. 2023-03-17· Medical· 2 results

    MedNeXt: Transformer-driven Scaling of ConvNets for Medical Image Segmentation

§ 04 · Related models

Other German Cancer Research Center (DKFZ) models scored on Codesota.

MedNeXt-L
62M params · 3 results · 2 SOTA
§ 05 · Sources & freshness

Where these numbers come from.

arxiv
4
results
4 of 4 rows marked verified. · first result 2023-03-17, latest 2024-06-11.