Brain Tumor Segmentation on BRATS-2015
Dice Score
- 2016-03-18 — 3D CNN + CRF: Dice Score 85.0
- 2019-06-05 — OM-Net + CGAp: Dice Score 87.0
| Rank | Model | Dice Score | Extra Training Data | Paper | Code | Year |
|---|---|---|---|---|---|---|
| 1 | OM-Net + CGAp | 87% | ✓ | One-pass Multi-task Networks with Cross-task Guided Attention for Brain Tumor Segmentation | chenhong-zhou/OM-Net | 2019 |
| 2 | CNN + 3D filters | 85.0% | ✓ | CNN-based Segmentation of Medical Imaging Data | shreyaspadhy/unet-zoo · BRML/CNNbasedMedicalSegmentation | 2017 |
| 2 | 3D CNN + CRF | 85.0% | ✓ | Efficient Multi-Scale 3D CNN with Fully Connected CRF for Accurate Brain Lesion Segmentation | Kamnitsask/deepmedic · etjoa003/medical_imaging | 2016 |
| 4 | AFN-6 | 84% | ✓ | Autofocus Layer for Semantic Segmentation | yaq007/Autofocus-Layer · perslev/Autofocus-Layer-TF · luvgold/auotofoucus3D-Brats | 2018 |