Dev

MRS — 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
linewidth -2.4 – 3.6 Hz
freq_drift -1.8 – 2.7 Hz
phase_error -6.0 – 9.0 deg
baseline -0.06 – 0.09

Dev Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 PnP-DnCNN-Pro + gradient 0.842 38.09 0.982 0.84 ✓ 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 0.841 36.99 0.978 0.9 ✓ Certified Pan et al., IEEE TMI 2025
3 HUMUS-Net++ + gradient 0.840 37.74 0.981 0.85 ✓ Certified Fabian et al., dHUMUS-Net 2023 — k-space DC per module + dynamic multi-scale weighting + INR head + perceptual-structural loss + axial attention
4 SwinMR++ + gradient 0.837 37.32 0.979 0.86 ✓ 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
5 HybridCascade++ + gradient 0.817 35.0 0.968 0.9 ✓ 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
6 MR-IPT + gradient 0.810 35.74 0.972 0.82 ✓ Certified Sci. Reports 2025
7 MRI-FM + gradient 0.797 34.53 0.965 0.83 ✓ Certified Wang et al., Nature MI 2026
8 BrainID-MRI + gradient 0.789 34.51 0.964 0.79 ✓ Certified Liu et al., CVPR 2025
9 ReconFormer + gradient 0.780 32.76 0.95 0.86 ✓ Certified Guo et al., IEEE TMI 2024
10 MRDynamo + gradient 0.777 33.58 0.958 0.79 ✓ Certified Chen et al., NeurIPS 2024
11 MMR-Mamba + gradient 0.765 31.71 0.939 0.86 ✓ Certified Zhao et al., Med. Image Anal. 2025
12 U-Net++ + gradient 0.761 31.31 0.935 0.87 ✓ 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
13 PromptMR-SFM + gradient 0.759 31.69 0.939 0.83 ✓ Certified PWM 2026
14 SwinMR + gradient 0.758 32.05 0.943 0.8 ✓ Certified Huang et al., MICCAI 2022
15 HUMUS-Net + gradient 0.755 30.75 0.928 0.88 ✓ Certified Fabian et al., NeurIPS 2022
16 E2E-VarNet + gradient 0.754 30.46 0.924 0.9 ✓ Certified Sriram et al., MICCAI 2020
17 MoDL-Net++ + gradient 0.753 31.89 0.941 0.79 ✓ 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
18 MRI-DiffusionNet + gradient 0.752 31.54 0.937 0.81 ✓ Certified Song et al., ICCV 2024
19 PromptMR + gradient 0.749 31.18 0.933 0.82 ✓ Certified Bai et al., ECCV 2024
20 PnP-DnCNN + gradient 0.738 29.44 0.908 0.9 ✓ Certified Ahmad et al., IEEE TCI 2019 (DOI:10.1109/TCI.2019.2944521)
21 BM3D-MRI + gradient 0.728 29.12 0.902 0.88 ✓ Certified Eksioglu, Comput. Math. Meth. Med. 2016
22 GRAPPA + gradient 0.712 29.24 0.904 0.79 ✓ Certified Griswold et al., MRM 2002
23 ALOHA + gradient 0.709 28.38 0.889 0.85 ✓ Certified Jin et al., IEEE TMI 2016
24 DCCNN + gradient 0.697 27.73 0.875 0.85 ✓ Certified Schlemper et al., IEEE TMI 2018
25 MoDL + gradient 0.696 28.28 0.887 0.79 ✓ Certified Aggarwal et al., IEEE TMI 2019
26 Deep-ADMM-Net + gradient 0.696 27.47 0.869 0.87 ✓ Certified Yang et al., NeurIPS 2016
27 U-Net + gradient 0.694 27.16 0.862 0.89 ✓ Certified Zbontar et al., arXiv 2018
28 HybridCascade + gradient 0.691 27.29 0.865 0.86 ✓ Certified Fastmri, arXiv 2020
29 ESPIRiT + gradient 0.670 26.34 0.841 0.85 ✓ Certified Uecker et al., MRM 2014
30 SENSE + gradient 0.665 25.63 0.821 0.9 ✓ Certified Pruessmann et al., MRM 1999
31 LORAKS + gradient 0.647 25.36 0.813 0.84 ✓ Certified Haldar, IEEE TMI 2014
32 L1-Wavelet + gradient 0.640 25.07 0.804 0.84 ✓ Certified Lustig et al., MRM 2007
33 Zero-Filled IFFT + gradient 0.628 24.54 0.787 0.84 ✓ Certified Pruessmann et al., MRM 1999
34 Score-MRI + gradient 0.596 23.46 0.749 0.81 ✓ Certified Chung & Ye, Med. Image Anal. 2022
35 k-t SPARSE-SENSE + gradient 0.584 22.36 0.705 0.89 ✓ Certified Lustig et al., MRM 2006

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%
Back to MRS