# Multiclass Neurodegenerative Disease Classification on ADNI3 CN vs. EMCI vs. AD

Source: [Multiclass Neurodegenerative Disease Classification on ADNI3 CN vs. EMCI vs. AD](<https://www.sota2.com/research/sota/multiclass-neurodegenerative-disease-classification-on-adni3-cn-vs-emci-vs-ad>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:31:36.450Z

Evaluation of multi-class classification performance distinguishing between Cognitively Normal (CN), Early Mild Cognitive Impairment (EMCI), and Alzheimer's Disease (AD) subjects using the ADNI3 dataset.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Multiclass Neurodegenerative Disease Classification |
| Dataset (including any stated split) | ADNI3 CN vs. EMCI vs. AD |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Accuracy](<https://www.sota2.com/research/sota/multiclass-neurodegenerative-disease-classification-on-adni3-cn-vs-emci-vs-ad.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Balanced Accuracy](<https://www.sota2.com/research/sota/multiclass-neurodegenerative-disease-classification-on-adni3-cn-vs-emci-vs-ad.md?metric=Balanced%20Accuracy>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/multiclass-neurodegenerative-disease-classification-on-adni3-cn-vs-emci-vs-ad.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [Specificity](<https://www.sota2.com/research/sota/multiclass-neurodegenerative-disease-classification-on-adni3-cn-vs-emci-vs-ad.md?metric=Specificity>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| DW-DGAT | 68.65 | Not recorded | 6969c5dfe25ad2fa9d1ec6b5 |
| ViT-small | 64.03 | Not recorded | 6969c5dfe25ad2fa9d1ec6b5 |
| ChebNetII | 62.9 | Not recorded | 6969c5dfe25ad2fa9d1ec6b5 |
| GATv2 | 62.07 | Not recorded | 6969c5dfe25ad2fa9d1ec6b5 |
| LG-GNN | 61.64 | Not recorded | 6969c5dfe25ad2fa9d1ec6b5 |
| VGG-19-BN | 60.64 | Not recorded | 6969c5dfe25ad2fa9d1ec6b5 |
| GCN | 59.41 | Not recorded | 6969c5dfe25ad2fa9d1ec6b5 |
| BrainGNN | 56.9 | Not recorded | 6969c5dfe25ad2fa9d1ec6b5 |
| MLP | 56.69 | Not recorded | 6969c5dfe25ad2fa9d1ec6b5 |
| BrainNet-T | 56.47 | Not recorded | 6969c5dfe25ad2fa9d1ec6b5 |
| RA-GCN | 55.07 | Not recorded | 6969c5dfe25ad2fa9d1ec6b5 |
| ResNet-34 | 54.42 | Not recorded | 6969c5dfe25ad2fa9d1ec6b5 |
| BrainNetCNN | 53.66 | Not recorded | 6969c5dfe25ad2fa9d1ec6b5 |
| MA-GCNN | 53.28 | Not recorded | 6969c5dfe25ad2fa9d1ec6b5 |
| GCN-MHSA | 50.86 | Not recorded | 6969c5dfe25ad2fa9d1ec6b5 |

### DW-DGAT

Accuracy: 68.65. Result ID: 6997ba7e5698665ec41103a0.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5dfe25ad2fa9d1ec6b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6969c5dfe25ad2fa9d1ec6b5.md>)

Recorded evaluation configuration: {"Validation Protocol":"Ten-fold cross-validation"}. This record does not assert independent reproduction.

### ViT-small

Accuracy: 64.03. Result ID: 6997ba7e5698665ec411039f.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5dfe25ad2fa9d1ec6b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a0b1e209131cea25b04.md>)

[Implementation](<https://github.com/google-research/vision_transformer>)

Recorded evaluation configuration: {"Validation Protocol":"Ten-fold cross-validation"}. This record does not assert independent reproduction.

### ChebNetII

Accuracy: 62.9. Result ID: 6997ba7e5698665ec4110395.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5dfe25ad2fa9d1ec6b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72751e209131cea567fc.md>)

Recorded evaluation configuration: {"Validation Protocol":"Ten-fold cross-validation"}. This record does not assert independent reproduction.

### GATv2

Accuracy: 62.07. Result ID: 6997ba7e5698665ec4110396.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5dfe25ad2fa9d1ec6b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6bba1e209131cea2feef.md>)

[Implementation](<https://github.com/tech-srl/how_attentive_are_gats>)

Recorded evaluation configuration: {"Validation Protocol":"Ten-fold cross-validation"}. This record does not assert independent reproduction.

### LG-GNN

Accuracy: 61.64. Result ID: 6997ba7e5698665ec4110399.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5dfe25ad2fa9d1ec6b5.md>)

Recorded evaluation configuration: {"Validation Protocol":"Ten-fold cross-validation"}. This record does not assert independent reproduction.

### VGG-19-BN

Accuracy: 60.64. Result ID: 6997ba7e5698665ec411039e.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5dfe25ad2fa9d1ec6b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61d01e209131ce9f62f5.md>)

[Implementation](<http://www.robots.ox.ac.uk/~vgg/research/very_deep/>)

Recorded evaluation configuration: {"Validation Protocol":"Ten-fold cross-validation"}. This record does not assert independent reproduction.

### GCN

Accuracy: 59.41. Result ID: 6997ba7e5698665ec4110397.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5dfe25ad2fa9d1ec6b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa631d1e209131ce9fce45.md>)

[Implementation](<https://github.com/tkipf/gcn>)

Recorded evaluation configuration: {"Validation Protocol":"Ten-fold cross-validation"}. This record does not assert independent reproduction.

### BrainGNN

Accuracy: 56.9. Result ID: 6997ba7e5698665ec4110392.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5dfe25ad2fa9d1ec6b5.md>)

Recorded evaluation configuration: {"Validation Protocol":"Ten-fold cross-validation"}. This record does not assert independent reproduction.

### MLP

Accuracy: 56.69. Result ID: 6997ba7e5698665ec411039b.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5dfe25ad2fa9d1ec6b5.md>)

Recorded evaluation configuration: {"Validation Protocol":"Ten-fold cross-validation"}. This record does not assert independent reproduction.

### BrainNet-T

Accuracy: 56.47. Result ID: 6997ba7e5698665ec4110393.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5dfe25ad2fa9d1ec6b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6dac1e209131cea3b62a.md>)

[Implementation](<https://github.com/Wayfear/BrainNetworkTransformer>)

Recorded evaluation configuration: {"Validation Protocol":"Ten-fold cross-validation"}. This record does not assert independent reproduction.

### RA-GCN

Accuracy: 55.07. Result ID: 6997ba7e5698665ec411039c.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5dfe25ad2fa9d1ec6b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b321e209131cea2ccd2.md>)

Recorded evaluation configuration: {"Validation Protocol":"Ten-fold cross-validation"}. This record does not assert independent reproduction.

### ResNet-34

Accuracy: 54.42. Result ID: 6997ba7e5698665ec411039d.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5dfe25ad2fa9d1ec6b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62321e209131ce9f8328.md>)

Recorded evaluation configuration: {"Validation Protocol":"Ten-fold cross-validation"}. This record does not assert independent reproduction.

### BrainNetCNN

Accuracy: 53.66. Result ID: 6997ba7e5698665ec4110394.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5dfe25ad2fa9d1ec6b5.md>)

Recorded evaluation configuration: {"Validation Protocol":"Ten-fold cross-validation"}. This record does not assert independent reproduction.

### MA-GCNN

Accuracy: 53.28. Result ID: 6997ba7e5698665ec411039a.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5dfe25ad2fa9d1ec6b5.md>)

Recorded evaluation configuration: {"Validation Protocol":"Ten-fold cross-validation"}. This record does not assert independent reproduction.

### GCN-MHSA

Accuracy: 50.86. Result ID: 6997ba7e5698665ec4110398.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5dfe25ad2fa9d1ec6b5.md>)

Recorded evaluation configuration: {"Validation Protocol":"Ten-fold cross-validation"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
