MCI vs. AD Classification on ADNI
86.84Accuracy3D CNNs
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| 3D CNNsArchitecture=3D CNNs, Modality=Unimodal (MRI)2026.06 | 86.84 | — | — | — | — | — | — | |
| Gated self-attentionArchitecture=Gated self-attention, Modality=Multimodal2026.06 | 82.08 | — | 81.43 | 80.48 | 85.24 | — | — | |
| 3D Multi-Scale CNNArchitecture=3D Multi-Scale CNN, Modality=Multimodal2026.06 | 80.8 | — | 71.19 | — | 85.94 | — | — | |
| PIMEEvaluation Protocol=5-fold CV2026.02 | 78.03 | — | — | — | — | — | — | |
| BrainNetTFEvaluation Protocol=5-fold CV2026.02 | 77.56 | — | — | — | — | — | — | |
| BrainIB++Evaluation Protocol=5-fold CV2026.02 | 77.19 | — | — | — | — | — | — | |
| BrainIBEvaluation Protocol=5-fold CV2026.02 | 76.7 | — | — | — | — | — | — | |
| ProtGNNEvaluation Protocol=5-fold CV2026.02 | 76.19 | — | — | — | — | — | — | |
| GMN4ADvalidation=5-fold cross validation, optimization=grid search2026.06 | 76.13 | — | 75.16 | — | 76.57 | 71.65 | 73.17 | |
| GCNvalidation=5-fold cross validation, optimization=grid search2026.06 | 76.09 | — | 73.91 | — | 77.78 | 72.07 | 72.76 | |
| LETetCNN2025.03 | 75.5 | — | 61.5 | — | 83.5 | — | — | |
| GATvalidation=5-fold cross validation, optimization=grid search2026.06 | 75.49 | — | 74.13 | — | 76.61 | 71.05 | 72.38 | |
| GINvalidation=5-fold cross validation, optimization=grid search2026.06 | 75.44 | — | 71.13 | — | 78.65 | 71.93 | 71.5 | |
| FILvalidation=5-fold cross validation, optimization=grid search2026.06 | 75.29 | — | 71.95 | — | 77.89 | 71.53 | 71.64 | |
| PGIBEvaluation Protocol=5-fold CV2026.02 | 75.19 | — | — | — | — | — | — | |
| SIBEvaluation Protocol=5-fold CV2026.02 | 74.61 | — | — | — | — | — | — | |
| DIR-GNNEvaluation Protocol=5-fold CV2026.02 | 74.36 | — | — | — | — | — | — | |
| Graph TransformerEvaluation Protocol=5-fold CV2026.02 | 72.79 | — | — | — | — | — | — | |
| GAT2025.03 | 72.7 | — | 63 | — | 77.3 | — | — | |
| GINEvaluation Protocol=5-fold CV2026.02 | 72.68 | — | — | — | — | — | — | |
| GATEvaluation Protocol=5-fold CV2026.02 | 71.95 | — | — | — | — | — | — | |
| GCNEvaluation Protocol=5-fold CV2026.02 | 71.63 | — | — | — | — | — | — | |
| TetCNN2025.03 | 70.9 | — | 66 | — | 76.9 | — | — | |
| ChebyNet2025.03 | 70.3 | — | 79 | — | 61.6 | — | — | |
| FedADAexperiments=five repeated experiments2025.12 | 67.33 | — | 62.96 | 67.93 | — | — | — | |
| BLvalidation=5-fold cross validation, optimization=grid search, cross-graph feature representation learning=disabled2026.06 | 62.99 | — | 64.2 | — | 62.25 | 58.61 | 59.35 |