Disease Classification on VinDr
0.925AUROCX-WIN
Evaluation Results
| Method | Links | |
|---|---|---|
| X-WINPretrain. data=372K CXRs + 32K CT volumes, Backbone=ViT-Large, Evaluation protocol=linear probing2025.11 | 0.925 | |
| Ark+Pretrain. data=704K CXRs, Backbone=Swin-Base, Evaluation protocol=linear probing2025.11 | 0.906 | |
| X-WINPretrain. data=372K CXRs + 32K CT volumes, Backbone=ViT-Base, Evaluation protocol=linear probing2025.11 | 0.905 | |
| CheXWorldPretrain. data=448K CXRs, Backbone=ViT-Base, Evaluation protocol=linear probing2025.11 | 0.903 | |
| CheXFoundPretrain. data=987K CXRs, Backbone=ViT-Large, Evaluation protocol=linear probing2025.11 | 0.869 | |
| CheXAgentPretrain. data=1.07M CXRs w/ text, Backbone=ViT-Large, Evaluation protocol=linear probing2025.11 | 0.823 | |
| MaCoPretrain. data=377K CXRs w/ reports, Backbone=ViT-Base, Evaluation protocol=linear probing2025.11 | 0.817 | |
| DINOv2Pretrain. data=LVD-142M, Backbone=ViT-Base, Evaluation protocol=linear probing2025.11 | 0.795 | |
| RAD-DINOPretrain. data=LVD-142M + 838K CXRs, Backbone=ViT-Base, Evaluation protocol=linear probing2025.11 | 0.795 | |
| CXR-AlignPretrain. data=325K CXRs w/ reports, Backbone=ViT-Base, Evaluation protocol=linear probing2025.11 | 0.793 | |
| CARZeroPretrain. data=377K CXRs w/ reports, Backbone=ViT-Base, Evaluation protocol=linear probing2025.11 | 0.776 | |
| I-JEPAPretrain. data=ImageNet-1K, Backbone=ViT-Large, Evaluation protocol=linear probing2025.11 | 0.77 |