Image Classification on OrganCMNIST MedMNIST (test)
96.02AccuracyOmniRad base
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| OmniRad baseSize=base2026.02 | 96.02 | 95.45 | 99.87 | — | — | |
| OmniRad smallSize=small2026.02 | 95.49 | 94.86 | 99.83 | — | — | |
| DINOv3Size=small2026.02 | 95.31 | 94.74 | 99.78 | — | — | |
| DINOv3Size=base2026.02 | 95.18 | 94.46 | 99.79 | — | — | |
| Radio DINOSize=base2026.02 | 95.11 | 94.57 | 99.86 | — | — | |
| Radio DINOSize=small2026.02 | 94.3 | 93.63 | 99.8 | — | — | |
| DINOSize=base2026.02 | 94.28 | 94.06 | 99.72 | — | — | |
| DINOSize=small2026.02 | 93.67 | 92.91 | 99.67 | — | — | |
| Med ViTSize=base2026.02 | 92.2 | — | 99.4 | — | — | |
| Resnet18Architecture=ResNet-182026.02 | 92 | — | 99.4 | — | — | |
| ResNet-18Input Resolution=2242026.02 | 92 | — | 99.4 | — | — | |
| Med ViTSize=small2026.02 | 91.6 | — | 99.3 | — | — | |
| DINOv2Size=small2026.02 | 91.37 | 89.83 | 99.38 | — | — | |
| Resnet50Architecture=ResNet-502026.02 | 91.1 | — | 99.3 | — | — | |
| ResNet-50Input Resolution=2242026.02 | 91.1 | — | 99.3 | — | — | |
| EfficientNetB0Architecture=EfficientNet-B02026.02 | 90.5 | — | 99.2 | — | — | |
| ResNet-50Input Resolution=282026.02 | 90.5 | — | 99.2 | — | — | |
| Wasserstein FIObjective Function=Wasserstein FI-based regularizer, Backbone=ResNet-18, Number of repetitions=10, Input Resolution=28x282025.02 | 90.28 | — | 98.31 | 0.4291 | 7.69 | |
| Med ViTSize=tiny2026.02 | 90.1 | — | 99.1 | — | — | |
| ResNet-18Input Resolution=282026.02 | 90 | — | 99.2 | — | — | |
| ERMObjective Function=Empirical Risk Minimization, Backbone=ResNet-18, Number of repetitions=10, Input Resolution=28x282025.02 | 89.41 | — | 99.02 | 0.4566 | 8.27 | |
| AutoKeras2026.02 | 87.9 | — | 99 | — | — | |
| AutoKeras2026.02 | 87.9 | — | 99 | — | — | |
| Google AutoML Vision2026.02 | 87.7 | — | 98.8 | — | — | |
| Google AutoML Vision2026.02 | 87.7 | — | 98.8 | — | — | |
| DINOv2Size=base2026.02 | 85.73 | 82.61 | 98.84 | — | — | |
| auto-sklearn2026.02 | 82.9 | — | 97.6 | — | — | |
| auto-sklearn2026.02 | 82.9 | — | 97.6 | — | — | |
| G-LoG bi-filtration (MLP)Sigma (σ)=0.52026.02 | 60.1 | — | 93.3 | — | — | |
| G-LoG bi-filtration (MLP)Sigma (σ)=02026.02 | 59.3 | — | 93.5 | — | — | |
| G-LoG bi-filtration (MLP)Sigma (σ)=12026.02 | 58.8 | — | 93.1 | — | — | |
| G-LoG bi-filtration (MLP)Sigma (σ)=1.52026.02 | 56.5 | — | 92.2 | — | — | |
| Topo-Med (MLP)2026.02 | 48.9 | — | 89.4 | — | — |