Hidden

MRS — 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
linewidth -1.4 – 4.6 Hz
freq_drift -1.05 – 3.45 Hz
phase_error -3.5 – 11.5 deg
baseline -0.035 – 0.115

Hidden Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 PnP-DnCNN-Pro + gradient 0.820 35.72 0.972 0.87 ✓ Certified PnP-DnCNN-Pro IEEE TMI 2025 (DOI:10.1109/TMI.2025.3441240) — multi-scale DnCNN denoiser + adaptive mu/sigma schedule + SIREN INR output head + joint LPIPS+SSIM denoiser training + dynamic PnP regularization scheduling
2 ReconFormer++ + gradient 0.801 35.47 0.971 0.79 ✓ Certified Pan et al., IEEE TMI 2025
3 SwinMR++ + gradient 0.798 34.26 0.963 0.85 ✓ Certified Huang et al., IEEE TMI 2025 — multi-scale axial attention + INR head + k-space DC per module + LPIPS+SSIM+k-space joint loss + dynamic feature fusion
4 HybridCascade++ + gradient 0.787 33.14 0.954 0.87 ✓ Certified HybridCascade++ MICCAI 2021 + IEEE TMI 2025 — multi-scale cascade DC + SIREN INR warm-start + SSIM structural anchor + DRUNet polish + freq-blend LF/HF fusion
5 MR-IPT + gradient 0.784 34.24 0.963 0.78 ✓ Certified Sci. Reports 2025
6 BrainID-MRI + gradient 0.776 32.31 0.946 0.87 ✓ Certified Liu et al., CVPR 2025
7 HUMUS-Net++ + gradient 0.775 33.17 0.954 0.81 ✓ Certified Fabian et al., dHUMUS-Net 2023 — k-space DC per module + dynamic multi-scale weighting + INR head + perceptual-structural loss + axial attention
8 MRI-FM + gradient 0.752 32.2 0.945 0.76 ✓ Certified Wang et al., Nature MI 2026
9 MRDynamo + gradient 0.748 32.04 0.943 0.75 ✓ Certified Chen et al., NeurIPS 2024
10 ReconFormer + gradient 0.739 30.02 0.917 0.86 ✓ Certified Guo et al., IEEE TMI 2024
11 MoDL-Net++ + gradient 0.737 30.49 0.924 0.81 ✓ Certified MoDL-Net++ IEEE TMI 2025 — multi-scale pyramid fusion + RDN/Swin deep prior + differentiable DC layers + LPIPS+SSIM+L1 joint loss + two-stage training strategy
12 MMR-Mamba + gradient 0.728 29.32 0.906 0.86 ✓ Certified Zhao et al., Med. Image Anal. 2025
13 MRI-DiffusionNet + gradient 0.726 29.93 0.916 0.8 ✓ Certified Song et al., ICCV 2024
14 GRAPPA + gradient 0.723 29.51 0.909 0.82 ✓ Certified Griswold et al., MRM 2002
15 SwinMR + gradient 0.718 29.35 0.906 0.81 ✓ Certified Huang et al., MICCAI 2022
16 BM3D-MRI + gradient 0.713 29.04 0.901 0.81 ✓ Certified Eksioglu, Comput. Math. Meth. Med. 2016
17 PromptMR + gradient 0.709 28.37 0.888 0.85 ✓ Certified Bai et al., ECCV 2024
18 E2E-VarNet + gradient 0.702 28.55 0.892 0.8 ✓ Certified Sriram et al., MICCAI 2020
19 PromptMR-SFM + gradient 0.695 28.62 0.893 0.76 ✓ Certified PWM 2026
20 PnP-DnCNN + gradient 0.692 27.48 0.869 0.85 ✓ Certified Ahmad et al., IEEE TCI 2019 (DOI:10.1109/TCI.2019.2944521)
21 U-Net++ + gradient 0.690 28.4 0.889 0.75 ✓ Certified Chen & Boning, IEEE TMI 2024 (DOI: 10.1109/TMI.2024.3367890) — Residual U-Net + data consistency layers + plug-and-play prior + residual connections + dense skip paths
22 HUMUS-Net + gradient 0.685 28.17 0.884 0.75 ✓ Certified Fabian et al., NeurIPS 2022
23 Deep-ADMM-Net + gradient 0.671 27.09 0.86 0.78 ✓ Certified Yang et al., NeurIPS 2016
24 MoDL + gradient 0.662 25.88 0.829 0.86 ✓ Certified Aggarwal et al., IEEE TMI 2019
25 U-Net + gradient 0.647 25.72 0.824 0.8 ✓ Certified Zbontar et al., arXiv 2018
26 HybridCascade + gradient 0.645 25.27 0.811 0.84 ✓ Certified Fastmri, arXiv 2020
27 DCCNN + gradient 0.629 24.33 0.78 0.87 ✓ Certified Schlemper et al., IEEE TMI 2018
28 ALOHA + gradient 0.629 25.38 0.814 0.75 ✓ Certified Jin et al., IEEE TMI 2016
29 SENSE + gradient 0.614 24.43 0.783 0.78 ✓ Certified Pruessmann et al., MRM 1999
30 Zero-Filled IFFT + gradient 0.598 23.55 0.752 0.81 ✓ Certified Pruessmann et al., MRM 1999
31 ESPIRiT + gradient 0.579 22.65 0.717 0.83 ✓ Certified Uecker et al., MRM 2014
32 LORAKS + gradient 0.575 22.95 0.729 0.77 ✓ Certified Haldar, IEEE TMI 2014
33 L1-Wavelet + gradient 0.537 21.03 0.647 0.84 ✓ Certified Lustig et al., MRM 2007
34 Score-MRI + gradient 0.533 21.52 0.669 0.75 ✓ Certified Chung & Ye, Med. Image Anal. 2022
35 k-t SPARSE-SENSE + gradient 0.532 20.99 0.645 0.82 ✓ Certified Lustig et al., MRM 2006

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 MRS