Public

Susceptibility-Weighted Imaging (SWI) — Public Tier

(3 scenes)

Full-access development tier with all data visible.

What you get

Measurements (y), ideal forward operator (H), spec ranges, ground truth (x_true), and true mismatch spec.

How to use

Load HDF5 → compare reconstruction vs x_true → check consistency → iterate.

What to submit

Reconstructed signals (x_hat) and corrected spec as HDF5.

Parameter Specifications

✓

True spec visible — use these exact values for Scenario III oracle reconstruction.

Parameter Spec Range True Value Unit
phase_unwrapping_error -1.0 – 2.0 0.5 -
background_field_removal_error -2.0 – 4.0 1.0 -
dipole_inversion_regularization -0.15 – 0.15 0.0 -

InverseNet Baseline Scores

Method: CPU_baseline — Mismatch parameter: nominal

Scenario I (Ideal)

17.61 dB

SSIM 0.3262

Scenario II (Mismatch)

13.39 dB

SSIM 0.0553

Scenario III (Oracle)

16.66 dB

SSIM 0.1359

Per-scene breakdown (4 scenes)
Scene PSNR I SSIM I PSNR II SSIM II PSNR III SSIM III
scene_00 16.58 0.3251 12.85 0.0515 16.16 0.1317
scene_01 18.88 0.3252 14.12 0.0562 17.37 0.1299
scene_02 18.30 0.3298 13.74 0.0603 16.95 0.1425
scene_03 16.68 0.3248 12.85 0.0534 16.15 0.1394

Public Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 SwinMR++ + gradient 0.908 42.11 0.992 0.95 ✓ 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 0.901 41.61 0.991 0.94 ✓ 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 0.895 41.46 0.991 0.92 ✓ 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 MRI-FM + gradient 0.888 40.48 0.989 0.94 ✓ Certified Wang et al., Nature MI 2026
5 MoDL-Net++ + gradient 0.886 40.06 0.988 0.95 ✓ 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
6 PromptMR-SFM + gradient 0.881 39.82 0.987 0.94 ✓ Certified PWM 2026
7 MR-IPT + gradient 0.874 40.63 0.989 0.86 ✓ Certified Sci. Reports 2025
8 ReconFormer++ + gradient 0.864 39.49 0.987 0.87 ✓ Certified Pan et al., IEEE TMI 2025
9 PromptMR + gradient 0.863 38.2 0.983 0.94 ✓ Certified Bai et al., ECCV 2024
10 U-Net++ + gradient 0.861 38.58 0.984 0.91 ✓ 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
11 BrainID-MRI + gradient 0.858 38.44 0.984 0.9 ✓ Certified Liu et al., CVPR 2025
12 PnP-DnCNN-Pro + gradient 0.856 38.07 0.982 0.91 ✓ 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
13 MMR-Mamba + gradient 0.856 38.23 0.983 0.9 ✓ Certified Zhao et al., Med. Image Anal. 2025
14 ReconFormer + gradient 0.855 37.64 0.981 0.93 ✓ Certified Guo et al., IEEE TMI 2024
15 HUMUS-Net + gradient 0.854 37.58 0.98 0.93 ✓ Certified Fabian et al., NeurIPS 2022
16 MRDynamo + gradient 0.853 38.17 0.983 0.89 ✓ Certified Chen et al., NeurIPS 2024
17 SwinMR + gradient 0.848 36.74 0.977 0.95 ✓ Certified Huang et al., MICCAI 2022
18 MRI-DiffusionNet + gradient 0.848 38.09 0.982 0.87 ✓ Certified Song et al., ICCV 2024
19 E2E-VarNet + gradient 0.841 37.63 0.981 0.86 ✓ Certified Sriram et al., MICCAI 2020
20 MoDL + gradient 0.826 35.42 0.97 0.92 ✓ Certified Aggarwal et al., IEEE TMI 2019
21 HybridCascade + gradient 0.818 34.8 0.966 0.92 ✓ Certified Fastmri, arXiv 2020
22 U-Net + gradient 0.817 34.17 0.962 0.95 ✓ Certified Zbontar et al., arXiv 2018
23 PnP-DnCNN + gradient 0.808 33.96 0.961 0.92 ✓ Certified Ahmad et al., IEEE TCI 2019 (DOI:10.1109/TCI.2019.2944521)
24 Deep-ADMM-Net + gradient 0.791 33.41 0.956 0.87 ✓ Certified Yang et al., NeurIPS 2016
25 DCCNN + gradient 0.786 32.52 0.948 0.91 ✓ Certified Schlemper et al., IEEE TMI 2018
26 ALOHA + gradient 0.775 31.85 0.941 0.9 ✓ Certified Jin et al., IEEE TMI 2016
27 BM3D-MRI + gradient 0.770 31.5 0.937 0.9 ✓ Certified Eksioglu, Comput. Math. Meth. Med. 2016
28 k-t SPARSE-SENSE + gradient 0.769 30.92 0.93 0.94 ✓ Certified Lustig et al., MRM 2006
29 LORAKS + gradient 0.763 30.81 0.928 0.92 ✓ Certified Haldar, IEEE TMI 2014
30 ESPIRiT + gradient 0.762 31.24 0.934 0.88 ✓ Certified Uecker et al., MRM 2014
31 L1-Wavelet + gradient 0.743 30.17 0.919 0.87 ✓ Certified Lustig et al., MRM 2007
32 Score-MRI + gradient 0.730 29.21 0.904 0.88 ✓ Certified Chung & Ye, Med. Image Anal. 2022
33 GRAPPA + gradient 0.726 28.87 0.898 0.89 ✓ Certified Griswold et al., MRM 2002
34 SENSE + gradient 0.688 26.76 0.852 0.9 ✓ Certified Pruessmann et al., MRM 1999
35 Zero-Filled IFFT + gradient 0.620 23.87 0.764 0.88 ✓ Certified Pruessmann et al., MRM 1999

Visible Data Fields

y H_ideal spec_ranges x_true true_spec

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 Susceptibility-Weighted Imaging (SWI)