# Whole brain measurement reconstruction on MUDI challenge (leave-one-out cross validation)

Source: [Whole brain measurement reconstruction on MUDI challenge (leave-one-out cross validation)](<https://www.sota2.com/research/sota/whole-brain-measurement-reconstruction-on-mudi-challenge-leave-one-out-cross-validation>)

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

Data updated: 2026-02-26T05:51:15.783Z

Evaluation of whole brain measurement reconstruction performance using the MUDI challenge dataset under a leave-one-out cross-validation scheme, typically measured using Mean Squared Error (MSE) across various undersampling factors (M).

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 | Whole brain measurement reconstruction |
| Dataset (including any stated split) | MUDI challenge (leave-one-out cross validation) |
| Metric | MSE (M=500) |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 6 of 6 |
| 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

- [MSE (M=500)](<https://www.sota2.com/research/sota/whole-brain-measurement-reconstruction-on-mudi-challenge-leave-one-out-cross-validation.md?metric=MSE%20(M%3D500)>): Lower is better; unit: SCALAR.

- [MSE (M=250)](<https://www.sota2.com/research/sota/whole-brain-measurement-reconstruction-on-mudi-challenge-leave-one-out-cross-validation.md?metric=MSE%20(M%3D250)>): Lower is better; unit: SCALAR.

- [MSE (M=100)](<https://www.sota2.com/research/sota/whole-brain-measurement-reconstruction-on-mudi-challenge-leave-one-out-cross-validation.md?metric=MSE%20(M%3D100)>): Lower is better; unit: SCALAR.

- [MSE (M=50)](<https://www.sota2.com/research/sota/whole-brain-measurement-reconstruction-on-mudi-challenge-leave-one-out-cross-validation.md?metric=MSE%20(M%3D50)>): Lower is better; unit: SCALAR.

- [MSE (M=40)](<https://www.sota2.com/research/sota/whole-brain-measurement-reconstruction-on-mudi-challenge-leave-one-out-cross-validation.md?metric=MSE%20(M%3D40)>): Lower is better; unit: SCALAR.

- [MSE (M=30)](<https://www.sota2.com/research/sota/whole-brain-measurement-reconstruction-on-mudi-challenge-leave-one-out-cross-validation.md?metric=MSE%20(M%3D30)>): Lower is better; unit: SCALAR.

- [MSE (M=20)](<https://www.sota2.com/research/sota/whole-brain-measurement-reconstruction-on-mudi-challenge-leave-one-out-cross-validation.md?metric=MSE%20(M%3D20)>): Lower is better; unit: SCALAR.

- [MSE (M=10)](<https://www.sota2.com/research/sota/whole-brain-measurement-reconstruction-on-mudi-challenge-leave-one-out-cross-validation.md?metric=MSE%20(M%3D10)>): Lower is better; unit: SCALAR.

## Recorded results

| Method | MSE (M=500) | Result date | Report paper |
| --- | --- | --- | --- |
| PROSUB | 0.49 | Not recorded | 68fa5f381e209131ce9e858f |
| PROSUB w/o NAS | 0.66 | Not recorded | 68fa5f381e209131ce9e858f |
| SARDU-Net-v2-NAS | 0.82 | Not recorded | 68fa5f381e209131ce9e858f |
| SARDU-Net-v2-BOF | 0.83 | Not recorded | 68fa5f381e209131ce9e858f |
| SARDU-Net-v2 | 0.88 | Not recorded | 68fa5f381e209131ce9e858f |
| SARDU-Net-v1 | 1.45 | Not recorded | 68fa5f381e209131ce9e858f |

### PROSUB

MSE (M=500): 0.49. Result ID: 698bbf137a967eddb7710d27.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f381e209131ce9e858f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5f381e209131ce9e858f.md>)

[Implementation](<https://github.com/sbb-gh/PROSUB>)

Recorded evaluation configuration: {"Category":"Ours"}. This record does not assert independent reproduction.

### PROSUB w/o NAS

MSE (M=500): 0.66. Result ID: 698bbf137a967eddb7710d26.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f381e209131ce9e858f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5f381e209131ce9e858f.md>)

[Implementation](<https://github.com/sbb-gh/PROSUB>)

Recorded evaluation configuration: {"Category":"Ours"}. This record does not assert independent reproduction.

### SARDU-Net-v2-NAS

MSE (M=500): 0.82. Result ID: 698bbf137a967eddb7710d25.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f381e209131ce9e858f.md>)

Recorded evaluation configuration: {"Category":"Baseline"}. This record does not assert independent reproduction.

### SARDU-Net-v2-BOF

MSE (M=500): 0.83. Result ID: 698bbf137a967eddb7710d24.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f381e209131ce9e858f.md>)

Recorded evaluation configuration: {"Category":"Baseline"}. This record does not assert independent reproduction.

### SARDU-Net-v2

MSE (M=500): 0.88. Result ID: 698bbf137a967eddb7710d23.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f381e209131ce9e858f.md>)

Recorded evaluation configuration: {"Category":"Baseline"}. This record does not assert independent reproduction.

### SARDU-Net-v1

MSE (M=500): 1.45. Result ID: 698bbf137a967eddb7710d22.

[Reporting paper](<https://www.sota2.com/research/paper/68fa5f381e209131ce9e858f.md>)

Recorded evaluation configuration: {"Category":"Baseline"}. This record does not assert independent reproduction.

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