Dev

fMRI — 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
B0_inhomog -2.4 – 3.6 ppm
head_motion -1.2 – 1.8 mm
hemodynamic_delay 4.8 – 7.8 s
physiological_noise -0.024 – 0.036

Dev Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 SwinMR++ + gradient 0.848 37.77 0.981 0.89 ✓ 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
2 HUMUS-Net++ + gradient 0.842 38.96 0.985 0.79 ✓ Certified Fabian et al., dHUMUS-Net 2023 — k-space DC per module + dynamic multi-scale weighting + INR head + perceptual-structural loss + axial attention
3 HybridCascade++ + gradient 0.828 35.85 0.973 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
4 ReconFormer++ + gradient 0.823 36.29 0.975 0.85 ✓ Certified Pan et al., IEEE TMI 2025
5 PromptMR-SFM + gradient 0.807 35.01 0.968 0.85 ✓ Certified PWM 2026
6 PnP-DnCNN-Pro + gradient 0.797 34.06 0.961 0.86 ✓ 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
7 MRI-FM + gradient 0.796 33.69 0.958 0.88 ✓ Certified Wang et al., Nature MI 2026
8 MoDL-Net++ + gradient 0.791 33.89 0.96 0.84 ✓ 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
9 U-Net++ + gradient 0.789 34.34 0.963 0.8 ✓ 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
10 SwinMR + gradient 0.788 33.55 0.957 0.85 ✓ Certified Huang et al., MICCAI 2022
11 PromptMR + gradient 0.784 33.52 0.957 0.83 ✓ Certified Bai et al., ECCV 2024
12 E2E-VarNet + gradient 0.781 32.4 0.947 0.89 ✓ Certified Sriram et al., MICCAI 2020
13 MR-IPT + gradient 0.774 31.94 0.942 0.89 ✓ Certified Sci. Reports 2025
14 HUMUS-Net + gradient 0.769 31.59 0.938 0.89 ✓ Certified Fabian et al., NeurIPS 2022
15 ReconFormer + gradient 0.758 31.23 0.934 0.86 ✓ Certified Guo et al., IEEE TMI 2024
16 BrainID-MRI + gradient 0.755 31.86 0.941 0.8 ✓ Certified Liu et al., CVPR 2025
17 MMR-Mamba + gradient 0.746 30.32 0.922 0.87 ✓ Certified Zhao et al., Med. Image Anal. 2025
18 MRI-DiffusionNet + gradient 0.741 30.79 0.928 0.81 ✓ Certified Song et al., ICCV 2024
19 MRDynamo + gradient 0.740 29.95 0.916 0.87 ✓ Certified Chen et al., NeurIPS 2024
20 PnP-DnCNN + gradient 0.740 30.22 0.92 0.85 ✓ Certified Ahmad et al., IEEE TCI 2019 (DOI:10.1109/TCI.2019.2944521)
21 BM3D-MRI + gradient 0.734 30.46 0.924 0.8 ✓ Certified Eksioglu, Comput. Math. Meth. Med. 2016
22 MoDL + gradient 0.733 29.3 0.906 0.89 ✓ Certified Aggarwal et al., IEEE TMI 2019
23 HybridCascade + gradient 0.728 29.12 0.902 0.88 ✓ Certified Fastmri, arXiv 2020
24 DCCNN + gradient 0.726 29.47 0.908 0.84 ✓ Certified Schlemper et al., IEEE TMI 2018
25 GRAPPA + gradient 0.707 28.69 0.895 0.81 ✓ Certified Griswold et al., MRM 2002
26 ALOHA + gradient 0.699 27.94 0.88 0.84 ✓ Certified Jin et al., IEEE TMI 2016
27 SENSE + gradient 0.688 27.32 0.866 0.84 ✓ Certified Pruessmann et al., MRM 1999
28 Deep-ADMM-Net + gradient 0.678 26.55 0.847 0.87 ✓ Certified Yang et al., NeurIPS 2016
29 U-Net + gradient 0.644 24.71 0.793 0.9 ✓ Certified Zbontar et al., arXiv 2018
30 L1-Wavelet + gradient 0.623 24.58 0.789 0.81 ✓ Certified Lustig et al., MRM 2007
31 LORAKS + gradient 0.614 23.54 0.752 0.89 ✓ Certified Haldar, IEEE TMI 2014
32 k-t SPARSE-SENSE + gradient 0.613 24.05 0.77 0.82 ✓ Certified Lustig et al., MRM 2006
33 Score-MRI + gradient 0.602 23.69 0.757 0.81 ✓ Certified Chung & Ye, Med. Image Anal. 2022
34 ESPIRiT + gradient 0.600 23.24 0.74 0.86 ✓ Certified Uecker et al., MRM 2014
35 Zero-Filled IFFT + gradient 0.568 22.52 0.712 0.79 ✓ Certified Pruessmann et al., MRM 1999

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 fMRI