Hidden

Diffusion MRI — Hidden Tier

(3 scenes)

Fully blind server-side evaluation — no data download.

What you get

No data downloadable. Algorithm runs server-side on hidden measurements.

How to use

Package algorithm as Docker container / Python script. Submit via link.

What to submit

Containerized algorithm accepting y + H, outputting x_hat + corrected spec.

Parameter Specifications

🔒

True spec hidden — blind evaluation, only ranges available.

Parameter Spec Range Unit
b_value_error -2.1 – 6.9 %
eddy_current -0.35 – 1.15 voxels
gradient_direction -0.7 – 2.3 deg
susceptibility -0.7 – 2.3 voxels

Hidden Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 DiffusionDTI + gradient 0.724 29.08 0.902 0.86 ✓ Certified Gao et al., NeurIPS 2024
2 PhysDiffMRI + gradient 0.696 27.77 0.876 0.84 ✓ Certified Chen et al., MRM 2024
3 DTIFormer + gradient 0.672 27.23 0.864 0.77 ✓ Certified Liu et al., MICCAI 2022
4 SwinDTI + gradient 0.650 26.34 0.841 0.75 ✓ Certified Wang et al., MICCAI 2023
5 DWIML-Net + gradient 0.588 23.08 0.734 0.82 ✓ Certified Qin et al., IEEE TMI 2019
6 SHORE + gradient 0.536 21.5 0.668 0.77 ✓ Certified Merlet & Deriche, MRM 2013
7 DnCNN-DTI + gradient 0.516 20.08 0.603 0.87 ✓ Certified Golkov et al., IEEE TMI 2016
8 DTI-FIT + gradient 0.434 17.99 0.5 0.77 ✓ Certified Behrens et al., MRM 2003
9 CHARMED + gradient 0.397 16.48 0.425 0.81 ✓ Certified Assaf & Basser, NeuroImage 2005

Dataset

Scenes: 3

Scoring Formula

0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

PSNR: 40% SSIM: 40% Consistency: 20%
Back to Diffusion MRI