Dev
Diffusion MRI — Dev Tier
(3 scenes)Blind evaluation tier — no ground truth available.
What you get
Measurements (y), ideal forward operator (H), and spec ranges only.
How to use
Apply your pipeline from the Public tier. Use consistency as self-check.
What to submit
Reconstructed signals and corrected spec. Scored server-side.
Parameter Specifications
🔒
True spec hidden — estimate parameters from spec ranges below.
| Parameter | Spec Range | Unit |
|---|---|---|
| b_value_error | -3.6 – 5.4 | % |
| eddy_current | -0.6 – 0.9 | voxels |
| gradient_direction | -1.2 – 1.8 | deg |
| susceptibility | -1.2 – 1.8 | voxels |
Dev Tier Leaderboard
| # | Method | Score | PSNR | SSIM | Consistency | Trust | Source |
|---|---|---|---|---|---|---|---|
| 1 | DiffusionDTI + gradient | 0.750 | 31.4 | 0.936 | 0.81 | ✓ Certified | Gao et al., NeurIPS 2024 |
| 2 | PhysDiffMRI + gradient | 0.743 | 30.15 | 0.919 | 0.87 | ✓ Certified | Chen et al., MRM 2024 |
| 3 | SwinDTI + gradient | 0.724 | 28.85 | 0.898 | 0.88 | ✓ Certified | Wang et al., MICCAI 2023 |
| 4 | DTIFormer + gradient | 0.713 | 28.48 | 0.891 | 0.86 | ✓ Certified | Liu et al., MICCAI 2022 |
| 5 | DWIML-Net + gradient | 0.635 | 25.0 | 0.802 | 0.82 | ✓ Certified | Qin et al., IEEE TMI 2019 |
| 6 | SHORE + gradient | 0.577 | 22.65 | 0.717 | 0.82 | ✓ Certified | Merlet & Deriche, MRM 2013 |
| 7 | DnCNN-DTI + gradient | 0.559 | 21.67 | 0.676 | 0.86 | ✓ Certified | Golkov et al., IEEE TMI 2016 |
| 8 | DTI-FIT + gradient | 0.467 | 18.45 | 0.522 | 0.87 | ✓ Certified | Behrens et al., MRM 2003 |
| 9 | CHARMED + gradient | 0.435 | 17.84 | 0.492 | 0.8 | ✓ Certified | Assaf & Basser, NeuroImage 2005 |
Visible Data Fields
y
H_ideal
spec_ranges
Dataset
Format: HDF5
Scenes: 3
Scoring Formula
0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)
PSNR: 40%
SSIM: 40%
Consistency: 20%