Interstitial lung disease classification on 7 Lung Regions (test)
85.23AUCRadiomics-Guided (REN)
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Radiomics-Guided (REN)Integration Strategy=Validation-Weighted, Backbone Architecture=Radiomics-Guided Regional Expert Network2025.10 | 85.23 | 0.043 | 0.796 | 10.9 | 0.014 | |
| ViT WeightedIntegration Strategy=Validation-Weighted, Backbone Architecture=ViT2025.10 | 80.21 | 0.0577 | 0.73 | 4.4 | 0.297 | |
| Mamba WeightedIntegration Strategy=Validation-Weighted, Backbone Architecture=Mamba2025.10 | 79.83 | 0.049 | 0.737 | 3.9 | 0.351 | |
| CNN WeightedIntegration Strategy=Validation-Weighted, Backbone Architecture=CNN2025.10 | 79.6 | 0.0512 | 0.732 | 3.6 | 0.382 | |
| Mamba Gated MoEIntegration Strategy=End-to-End Gated MoE, Backbone Architecture=Mamba2025.10 | 79.5 | 0.0486 | 0.735 | 3.4 | 0.278 | |
| ViT Gated MoEIntegration Strategy=End-to-End Gated MoE, Backbone Architecture=ViT2025.10 | 79.46 | 0.0328 | 0.754 | 3.4 | 0.242 | |
| CNN Gated MoEIntegration Strategy=End-to-End Gated MoE, Backbone Architecture=CNN2025.10 | 77.78 | 0.0545 | 0.71 | 1.2 | 0.442 | |
| SwinUNETR (Baseline)Integration Strategy=Baseline, Backbone Architecture=SwinUNETR2025.10 | 76.85 | 0.0759 | 0.674 | — | — | |
| CNN (Baseline)Integration Strategy=Baseline, Backbone Architecture=CNN2025.10 | 75.84 | 0.0919 | 0.667 | — | — | |
| Mamba (Baseline)Integration Strategy=Baseline, Backbone Architecture=Mamba2025.10 | 67.75 | 0.0454 | 0.632 | — | — | |
| ViT (Baseline)Integration Strategy=Baseline, Backbone Architecture=ViT2025.10 | 65.35 | 0.0356 | 0.618 | — | — |