MR Angiography (MRA)

Join the Competition Contribute to Benchmark

Standard reconstruction benchmark — forward model perfectly known, no calibration needed. Score = 0.5 × clip((PSNR−15)/30, 0, 1) + 0.5 × SSIM

#	Method	Score	PSNR (dB)	SSIM	Source
🥇	SwinMR++ SwinMR++ Huang et al., IEEE TMI 2025 — 5 improvements: multi-scale axial attention (cross-scale long-range modeling), INR coordinate-query head (high-acceleration k-space interpolation), k-space DC per unrolled module, joint LPIPS+SSIM+k-space consistency loss, dynamic conv-Transformer branch weighting 43.8 dB SSIM 0.983 Checkpoint unavailable	0.971	43.8	0.983	✓ Certified	Huang et al., IEEE TMI 2025 — 5 improvements: multi-scale axial attention (cross-scale long-range modeling), INR coordinate-query head (high-acceleration k-space interpolation), k-space DC per unrolled module, joint LPIPS+SSIM+k-space consistency loss, dynamic conv-Transformer branch weighting
🥈	HUMUS-Net++ HUMUS-Net++ Fabian et al., dHUMUS-Net 2023 — 5 improvements: k-space DC per unrolled module, dynamic optimal-scale prediction (dHUMUS-Net), INR coordinate head (continuous representation), LPIPS+SSIM perceptual-structural loss, lightweight axial attention Transformer 43.1 dB SSIM 0.979 Checkpoint unavailable	0.958	43.1	0.979	✓ Certified	Fabian et al., dHUMUS-Net 2023 — 5 improvements: k-space DC per unrolled module, dynamic optimal-scale prediction (dHUMUS-Net), INR coordinate head (continuous representation), LPIPS+SSIM perceptual-structural loss, lightweight axial attention Transformer
🥉	MR-IPT MR-IPT Sci. Reports 2025 42.48 dB SSIM 0.983 Checkpoint unavailable	0.950	42.48	0.983	✓ Certified	Sci. Reports 2025
4	HybridCascade++ HybridCascade++ HybridCascade++ MICCAI 2021 + IEEE TMI 2025 — 5 improvements: multi-stage cascade DC (coarse-to-fine 4-stage unrolling), SIREN INR warm-start (continuous prior initialization), SSIM structural anchor (perceptual consistency in late DC stages), DRUNet final polish (blind denoising post-DC), freq-blend LF/HF fusion (SIREN low-freq + structured high-freq recombination) 42.5 dB SSIM 0.981 Checkpoint unavailable	0.949	42.5	0.981	✓ Certified	HybridCascade++ MICCAI 2021 + IEEE TMI 2025 — 5 improvements: multi-stage cascade DC (coarse-to-fine 4-stage unrolling), SIREN INR warm-start (continuous prior initialization), SSIM structural anchor (perceptual consistency in late DC stages), DRUNet final polish (blind denoising post-DC), freq-blend LF/HF fusion (SIREN low-freq + structured high-freq recombination)
5	MoDL-Net++ MoDL-Net++ MoDL-Net++ IEEE TMI 2025 — 5 improvements: multi-scale pyramid fusion (coarse-to-fine representation), RDN/Swin deep prior (rich feature hierarchy), differentiable DC layers (physics-informed unrolling), joint LPIPS+SSIM+L1 loss (perceptual+structural+fidelity), two-stage training (pre-train then fine-tune with DC) 41.8 dB SSIM 0.978 Checkpoint unavailable	0.936	41.8	0.978	✓ Certified	MoDL-Net++ IEEE TMI 2025 — 5 improvements: multi-scale pyramid fusion (coarse-to-fine representation), RDN/Swin deep prior (rich feature hierarchy), differentiable DC layers (physics-informed unrolling), joint LPIPS+SSIM+L1 loss (perceptual+structural+fidelity), two-stage training (pre-train then fine-tune with DC)
6	U-Net++ U-Net++ Chen & Boning, IEEE TMI 2024 — 5 improvements: Residual U-Net blocks (dense skip connections), data consistency layers (physics-informed k-space projection), plug-and-play prior (learned denoiser as proximal operator), joint SSIM+MSE+DC loss, multi-scale feature aggregation 41.5 dB SSIM 0.978 Checkpoint unavailable	0.931	41.5	0.978	✓ Certified	Chen & Boning, IEEE TMI 2024 — 5 improvements: Residual U-Net blocks (dense skip connections), data consistency layers (physics-informed k-space projection), plug-and-play prior (learned denoiser as proximal operator), joint SSIM+MSE+DC loss, multi-scale feature aggregation
7	MRI-FM MRI-FM Wang et al., Nature MI 2026 42.1 dB SSIM 0.948 Checkpoint unavailable	0.926	42.1	0.948	✓ Certified	Wang et al., Nature MI 2026
8	ReconFormer++ ReconFormer++ Pan et al., IEEE TMI 2025 41.5 dB SSIM 0.969 Checkpoint unavailable	0.926	41.5	0.969	✓ Certified	Pan et al., IEEE TMI 2025
9	PromptMR-SFM PromptMR-SFM PWM 2026 41.3 dB SSIM 0.971 Checkpoint unavailable	0.924	41.3	0.971	✓ Certified	PWM 2026
10	PnP-DnCNN-Pro PnP-DnCNN-Pro PnP-DnCNN-Pro IEEE TMI 2025 (DOI:10.1109/TMI.2025.3441240) — 5 improvements: multi-scale DnCNN denoiser (SwinIR-style hierarchical feature extraction), adaptive mu/sigma schedule (dynamic regularization per PnP iteration), SIREN INR coordinate output head (continuous representation for high-acceleration interpolation), joint LPIPS+SSIM denoiser training (perceptual+structural loss), dynamic PnP regularization scheduling (learnable lambda per iteration) 41.0 dB SSIM 0.968 Try in SpecLab →	0.917	41.0	0.968	✓ Certified	PnP-DnCNN-Pro IEEE TMI 2025 (DOI:10.1109/TMI.2025.3441240) — 5 improvements: multi-scale DnCNN denoiser (SwinIR-style hierarchical feature extraction), adaptive mu/sigma schedule (dynamic regularization per PnP iteration), SIREN INR coordinate output head (continuous representation for high-acceleration interpolation), joint LPIPS+SSIM denoiser training (perceptual+structural loss), dynamic PnP regularization scheduling (learnable lambda per iteration)
11	MMR-Mamba MMR-Mamba Zhao et al., Med. Image Anal. 2025 40.98 dB SSIM 0.969 Checkpoint unavailable	0.917	40.98	0.969	✓ Certified	Zhao et al., Med. Image Anal. 2025
12	BrainID-MRI BrainID-MRI Liu et al., CVPR 2025 41.0 dB SSIM 0.942 Checkpoint unavailable	0.904	41.0	0.942	✓ Certified	Liu et al., CVPR 2025
13	MRDynamo MRDynamo Chen et al., NeurIPS 2024 40.5 dB SSIM 0.938 Checkpoint unavailable	0.894	40.5	0.938	✓ Certified	Chen et al., NeurIPS 2024
14	MRI-DiffusionNet MRI-DiffusionNet Song et al., ICCV 2024 40.1 dB SSIM 0.932 Checkpoint unavailable	0.884	40.1	0.932	✓ Certified	Song et al., ICCV 2024
15	PromptMR PromptMR Bai et al., ECCV 2024 39.7 dB SSIM 0.926 Checkpoint unavailable	0.875	39.7	0.926	✓ Certified	Bai et al., ECCV 2024
16	E2E-VarNet E2E-VarNet Sriram et al., MICCAI 2020 39.4 dB SSIM 0.924 Checkpoint unavailable	0.869	39.4	0.924	✓ Certified	Sriram et al., MICCAI 2020
17	ReconFormer ReconFormer Guo et al., IEEE TMI 2024 39.0 dB SSIM 0.922 Checkpoint unavailable	0.861	39.0	0.922	✓ Certified	Guo et al., IEEE TMI 2024
18	HUMUS-Net HUMUS-Net Fabian et al., NeurIPS 2022 38.9 dB SSIM 0.923 Checkpoint unavailable	0.860	38.9	0.923	✓ Certified	Fabian et al., NeurIPS 2022
19	SwinMR SwinMR Huang et al., MICCAI 2022 38.5 dB SSIM 0.921 Checkpoint unavailable	0.852	38.5	0.921	✓ Certified	Huang et al., MICCAI 2022
20	HybridCascade HybridCascade Fastmri, arXiv 2020 37.8 dB SSIM 0.917 Checkpoint unavailable	0.839	37.8	0.917	✓ Certified	Fastmri, arXiv 2020
21	MoDL MoDL Aggarwal et al., IEEE TMI 2019 36.5 dB SSIM 0.912 Checkpoint unavailable	0.814	36.5	0.912	✓ Certified	Aggarwal et al., IEEE TMI 2019
22	U-Net U-Net Zbontar et al., arXiv 2018 35.9 dB SSIM 0.904 Checkpoint unavailable	0.800	35.9	0.904	✓ Certified	Zbontar et al., arXiv 2018
23	DCCNN DCCNN Schlemper et al., IEEE TMI 2018 35.5 dB SSIM 0.908 Checkpoint unavailable	0.796	35.5	0.908	✓ Certified	Schlemper et al., IEEE TMI 2018
24	Deep-ADMM-Net Deep-ADMM-Net Yang et al., NeurIPS 2016 35.3 dB SSIM 0.907 Checkpoint unavailable	0.792	35.3	0.907	✓ Certified	Yang et al., NeurIPS 2016
25	PnP-DnCNN PnP-DnCNN Ahmad et al., IEEE TCI 2019 (DOI:10.1109/TCI.2019.2944521) 35.0 dB SSIM 0.905 Try in SpecLab →	0.786	35.0	0.905	✓ Certified	Ahmad et al., IEEE TCI 2019 (DOI:10.1109/TCI.2019.2944521)
26	ALOHA ALOHA Jin et al., IEEE TMI 2016 34.5 dB SSIM 0.900 Try in SpecLab →	0.775	34.5	0.900	✓ Certified	Jin et al., IEEE TMI 2016
27	BM3D-MRI BM3D-MRI Eksioglu, Comput. Math. Meth. Med. 2016 34.2 dB SSIM 0.897 Try in SpecLab →	0.769	34.2	0.897	✓ Certified	Eksioglu, Comput. Math. Meth. Med. 2016
28	LORAKS LORAKS Haldar, IEEE TMI 2014 33.8 dB SSIM 0.893 Try in SpecLab →	0.760	33.8	0.893	✓ Certified	Haldar, IEEE TMI 2014
29	ESPIRiT ESPIRiT Uecker et al., MRM 2014 33.4 dB SSIM 0.890 Try in SpecLab →	0.752	33.4	0.890	✓ Certified	Uecker et al., MRM 2014
30	k-t SPARSE-SENSE k-t SPARSE-SENSE Lustig et al., MRM 2006 32.5 dB SSIM 0.875 Try in SpecLab →	0.729	32.5	0.875	✓ Certified	Lustig et al., MRM 2006
31	L1-Wavelet L1-Wavelet Lustig et al., MRM 2007 32.1 dB SSIM 0.870 Try in SpecLab →	0.720	32.1	0.870	✓ Certified	Lustig et al., MRM 2007
32	Score-MRI Score-MRI Chung & Ye, Med. Image Anal. 2022 31.4 dB SSIM 0.890 Checkpoint unavailable	0.718	31.4	0.890	✓ Certified	Chung & Ye, Med. Image Anal. 2022
33	GRAPPA GRAPPA Griswold et al., MRM 2002 31.2 dB SSIM 0.860 Try in SpecLab →	0.700	31.2	0.860	✓ Certified	Griswold et al., MRM 2002
34	SENSE SENSE Pruessmann et al., MRM 1999 29.5 dB SSIM 0.830 Try in SpecLab →	0.657	29.5	0.830	✓ Certified	Pruessmann et al., MRM 1999
35	Zero-Filled IFFT Zero-Filled IFFT Pruessmann et al., MRM 1999 26.0 dB SSIM 0.620 Try in SpecLab →	0.493	26.0	0.620	✓ Certified	Pruessmann et al., MRM 1999

Dataset: PWM Benchmark (35 algorithms)

Blind Reconstruction Challenge — forward model has unknown mismatch, must calibrate from data. Score = 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

#	Method	Overall Score	Public PSNR / SSIM	Dev PSNR / SSIM	Hidden PSNR / SSIM	Trust	Source
🥇	HUMUS-Net++ + gradient HUMUS-Net++ + gradient Fabian et al., dHUMUS-Net 2023 — k-space DC per module + dynamic multi-scale weighting + INR head + perceptual-structural loss + axial attention Score 0.852 Correct & Reconstruct →	0.852	0.879 40.18 dB / 0.988	0.853 38.51 dB / 0.984	0.824 36.53 dB / 0.976	✓ Certified	Fabian et al., dHUMUS-Net 2023 — k-space DC per module + dynamic multi-scale weighting + INR head + perceptual-structural loss + axial attention
🥈	SwinMR++ + gradient SwinMR++ + gradient 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 Score 0.832 Correct & Reconstruct →	0.832	0.907 42.05 dB / 0.992	0.826 37.22 dB / 0.979	0.763 32.67 dB / 0.950	✓ 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
🥉	MRI-FM + gradient MRI-FM + gradient Wang et al., Nature MI 2026 Score 0.831 Correct & Reconstruct →	0.831	0.888 40.48 dB / 0.989	0.814 35.1 dB / 0.968	0.790 33.67 dB / 0.958	✓ Certified	Wang et al., Nature MI 2026
4	ReconFormer++ + gradient ReconFormer++ + gradient Pan et al., IEEE TMI 2025 Score 0.824 Correct & Reconstruct →	0.824	0.861 38.58 dB / 0.984	0.839 37.63 dB / 0.981	0.772 33.83 dB / 0.960	✓ Certified	Pan et al., IEEE TMI 2025
5	PnP-DnCNN-Pro + gradient PnP-DnCNN-Pro + gradient 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 Score 0.814 Correct & Reconstruct →	0.814	0.877 39.63 dB / 0.987	0.796 34.91 dB / 0.967	0.769 32.37 dB / 0.947	✓ 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
6	HUMUS-Net + gradient HUMUS-Net + gradient Fabian et al., NeurIPS 2022 Score 0.811 Correct & Reconstruct →	0.811	0.833 36.48 dB / 0.976	0.810 34.71 dB / 0.966	0.791 34.21 dB / 0.962	✓ Certified	Fabian et al., NeurIPS 2022
7	MoDL-Net++ + gradient MoDL-Net++ + gradient 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 Score 0.809 Correct & Reconstruct →	0.809	0.867 39.44 dB / 0.986	0.788 34.4 dB / 0.964	0.771 33.13 dB / 0.954	✓ 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
8	MR-IPT + gradient MR-IPT + gradient Sci. Reports 2025 Score 0.806 Correct & Reconstruct →	0.806	0.873 39.64 dB / 0.987	0.795 34.51 dB / 0.964	0.751 32.28 dB / 0.946	✓ Certified	Sci. Reports 2025
9	HybridCascade++ + gradient HybridCascade++ + gradient HybridCascade++ MICCAI 2021 + IEEE TMI 2025 — multi-scale cascade DC + SIREN INR warm-start + SSIM structural anchor + DRUNet polish + freq-blend LF/HF fusion Score 0.803 Correct & Reconstruct →	0.803	0.893 40.92 dB / 0.990	0.772 32.6 dB / 0.949	0.743 30.75 dB / 0.928	✓ 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
10	ReconFormer + gradient ReconFormer + gradient Guo et al., IEEE TMI 2024 Score 0.803 Correct & Reconstruct →	0.803	0.832 36.04 dB / 0.974	0.818 35.89 dB / 0.973	0.760 32.92 dB / 0.952	✓ Certified	Guo et al., IEEE TMI 2024
11	U-Net++ + gradient U-Net++ + gradient 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 Score 0.801 Correct & Reconstruct →	0.801	0.883 39.81 dB / 0.987	0.769 33.03 dB / 0.953	0.750 30.55 dB / 0.925	✓ 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
12	PromptMR + gradient PromptMR + gradient Bai et al., ECCV 2024 Score 0.794 Correct & Reconstruct →	0.794	0.863 38.51 dB / 0.984	0.767 31.56 dB / 0.938	0.751 31.3 dB / 0.935	✓ Certified	Bai et al., ECCV 2024
13	MRI-DiffusionNet + gradient MRI-DiffusionNet + gradient Song et al., ICCV 2024 Score 0.792 Correct & Reconstruct →	0.792	0.867 38.58 dB / 0.984	0.769 32.37 dB / 0.947	0.741 30.79 dB / 0.928	✓ Certified	Song et al., ICCV 2024
14	MMR-Mamba + gradient MMR-Mamba + gradient Zhao et al., Med. Image Anal. 2025 Score 0.792 Correct & Reconstruct →	0.792	0.878 39.7 dB / 0.987	0.777 33.12 dB / 0.954	0.721 30.41 dB / 0.923	✓ Certified	Zhao et al., Med. Image Anal. 2025
15	BrainID-MRI + gradient BrainID-MRI + gradient Liu et al., CVPR 2025 Score 0.787 Correct & Reconstruct →	0.787	0.855 38.06 dB / 0.982	0.772 32.94 dB / 0.952	0.734 30.83 dB / 0.929	✓ Certified	Liu et al., CVPR 2025
16	E2E-VarNet + gradient E2E-VarNet + gradient Sriram et al., MICCAI 2020 Score 0.785 Correct & Reconstruct →	0.785	0.861 38.34 dB / 0.983	0.765 32.25 dB / 0.945	0.730 30.07 dB / 0.918	✓ Certified	Sriram et al., MICCAI 2020
17	PromptMR-SFM + gradient PromptMR-SFM + gradient PWM 2026 Score 0.783 Correct & Reconstruct →	0.783	0.882 40.08 dB / 0.988	0.759 30.93 dB / 0.930	0.707 28.16 dB / 0.884	✓ Certified	PWM 2026
18	SwinMR + gradient SwinMR + gradient Huang et al., MICCAI 2022 Score 0.779 Correct & Reconstruct →	0.779	0.849 36.8 dB / 0.977	0.779 32.88 dB / 0.951	0.709 28.25 dB / 0.886	✓ Certified	Huang et al., MICCAI 2022
19	MRDynamo + gradient MRDynamo + gradient Chen et al., NeurIPS 2024 Score 0.769 Correct & Reconstruct →	0.769	0.850 37.76 dB / 0.981	0.754 30.71 dB / 0.927	0.703 29.15 dB / 0.903	✓ Certified	Chen et al., NeurIPS 2024
20	PnP-DnCNN + gradient PnP-DnCNN + gradient Ahmad et al., IEEE TCI 2019 (DOI:10.1109/TCI.2019.2944521) Score 0.758 Correct & Reconstruct →	0.758	0.807 33.72 dB / 0.959	0.742 30.99 dB / 0.931	0.724 29.19 dB / 0.904	✓ Certified	Ahmad et al., IEEE TCI 2019 (DOI:10.1109/TCI.2019.2944521)
21	U-Net + gradient U-Net + gradient Zbontar et al., arXiv 2018 Score 0.750 Correct & Reconstruct →	0.750	0.816 34.14 dB / 0.962	0.750 30.87 dB / 0.929	0.684 28.18 dB / 0.885	✓ Certified	Zbontar et al., arXiv 2018
22	HybridCascade + gradient HybridCascade + gradient Fastmri, arXiv 2020 Score 0.746 Correct & Reconstruct →	0.746	0.841 36.13 dB / 0.974	0.728 28.97 dB / 0.900	0.668 26.96 dB / 0.857	✓ Certified	Fastmri, arXiv 2020
23	MoDL + gradient MoDL + gradient Aggarwal et al., IEEE TMI 2019 Score 0.745 Correct & Reconstruct →	0.745	0.806 34.71 dB / 0.966	0.724 29.1 dB / 0.902	0.706 28.1 dB / 0.883	✓ Certified	Aggarwal et al., IEEE TMI 2019
24	BM3D-MRI + gradient BM3D-MRI + gradient Eksioglu, Comput. Math. Meth. Med. 2016 Score 0.715 Correct & Reconstruct →	0.715	0.773 31.98 dB / 0.942	0.708 28.64 dB / 0.894	0.665 25.91 dB / 0.829	✓ Certified	Eksioglu, Comput. Math. Meth. Med. 2016
25	GRAPPA + gradient GRAPPA + gradient Griswold et al., MRM 2002 Score 0.703 Correct & Reconstruct →	0.703	0.729 29.26 dB / 0.905	0.691 27.07 dB / 0.860	0.688 27.45 dB / 0.869	✓ Certified	Griswold et al., MRM 2002
26	DCCNN + gradient DCCNN + gradient Schlemper et al., IEEE TMI 2018 Score 0.698 Correct & Reconstruct →	0.698	0.789 33.02 dB / 0.953	0.690 26.82 dB / 0.854	0.616 24.12 dB / 0.773	✓ Certified	Schlemper et al., IEEE TMI 2018
27	Deep-ADMM-Net + gradient Deep-ADMM-Net + gradient Yang et al., NeurIPS 2016 Score 0.693 Correct & Reconstruct →	0.693	0.789 33.28 dB / 0.955	0.678 27.02 dB / 0.859	0.611 23.5 dB / 0.750	✓ Certified	Yang et al., NeurIPS 2016
28	k-t SPARSE-SENSE + gradient k-t SPARSE-SENSE + gradient Lustig et al., MRM 2006 Score 0.676 Correct & Reconstruct →	0.676	0.772 31.38 dB / 0.936	0.647 25.1 dB / 0.805	0.610 23.7 dB / 0.758	✓ Certified	Lustig et al., MRM 2006
29	SENSE + gradient SENSE + gradient Pruessmann et al., MRM 1999 Score 0.670 Correct & Reconstruct →	0.670	0.695 27.3 dB / 0.865	0.685 27.19 dB / 0.863	0.631 24.48 dB / 0.785	✓ Certified	Pruessmann et al., MRM 1999
30	ALOHA + gradient ALOHA + gradient Jin et al., IEEE TMI 2016 Score 0.663 Correct & Reconstruct →	0.663	0.798 32.91 dB / 0.952	0.637 25.01 dB / 0.803	0.555 22.27 dB / 0.701	✓ Certified	Jin et al., IEEE TMI 2016
31	LORAKS + gradient LORAKS + gradient Haldar, IEEE TMI 2014 Score 0.657 Correct & Reconstruct →	0.657	0.765 31.17 dB / 0.933	0.639 24.56 dB / 0.788	0.568 22.84 dB / 0.725	✓ Certified	Haldar, IEEE TMI 2014
32	L1-Wavelet + gradient L1-Wavelet + gradient Lustig et al., MRM 2007 Score 0.650 Correct & Reconstruct →	0.650	0.765 30.78 dB / 0.928	0.622 24.67 dB / 0.792	0.562 21.86 dB / 0.684	✓ Certified	Lustig et al., MRM 2007
33	ESPIRiT + gradient ESPIRiT + gradient Uecker et al., MRM 2014 Score 0.635 Correct & Reconstruct →	0.635	0.758 30.73 dB / 0.927	0.617 24.4 dB / 0.782	0.529 21.31 dB / 0.660	✓ Certified	Uecker et al., MRM 2014
34	Score-MRI + gradient Score-MRI + gradient Chung & Ye, Med. Image Anal. 2022 Score 0.611 Correct & Reconstruct →	0.611	0.727 28.81 dB / 0.897	0.580 23.06 dB / 0.733	0.526 20.49 dB / 0.622	✓ Certified	Chung & Ye, Med. Image Anal. 2022
35	Zero-Filled IFFT + gradient Zero-Filled IFFT + gradient Pruessmann et al., MRM 1999 Score 0.597 Correct & Reconstruct →	0.597	0.651 24.81 dB / 0.796	0.589 23.1 dB / 0.735	0.550 21.29 dB / 0.659	✓ Certified	Pruessmann et al., MRM 1999

Complete score requires all 3 tiers (Public + Dev + Hidden).

Join the competition →

Scoring: 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖) PSNR 40% · SSIM 40% · Consistency 20%

Public 3 scenes

Full-access development tier with all data visible.

What you get & how to use

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.

Public Leaderboard

#	Method	Score	PSNR	SSIM
1	SwinMR++ + gradient	0.907	42.05	0.992
2	HybridCascade++ + gradient	0.893	40.92	0.99
3	MRI-FM + gradient	0.888	40.48	0.989
4	U-Net++ + gradient	0.883	39.81	0.987
5	PromptMR-SFM + gradient	0.882	40.08	0.988
6	HUMUS-Net++ + gradient	0.879	40.18	0.988
7	MMR-Mamba + gradient	0.878	39.7	0.987
8	PnP-DnCNN-Pro + gradient	0.877	39.63	0.987
9	MR-IPT + gradient	0.873	39.64	0.987
10	MoDL-Net++ + gradient	0.867	39.44	0.986
11	MRI-DiffusionNet + gradient	0.867	38.58	0.984
12	PromptMR + gradient	0.863	38.51	0.984
13	ReconFormer++ + gradient	0.861	38.58	0.984
14	E2E-VarNet + gradient	0.861	38.34	0.983
15	BrainID-MRI + gradient	0.855	38.06	0.982
16	MRDynamo + gradient	0.850	37.76	0.981
17	SwinMR + gradient	0.849	36.8	0.977
18	HybridCascade + gradient	0.841	36.13	0.974
19	HUMUS-Net + gradient	0.833	36.48	0.976
20	ReconFormer + gradient	0.832	36.04	0.974
21	U-Net + gradient	0.816	34.14	0.962
22	PnP-DnCNN + gradient	0.807	33.72	0.959
23	MoDL + gradient	0.806	34.71	0.966
24	ALOHA + gradient	0.798	32.91	0.952
25	DCCNN + gradient	0.789	33.02	0.953
26	Deep-ADMM-Net + gradient	0.789	33.28	0.955
27	BM3D-MRI + gradient	0.773	31.98	0.942
28	k-t SPARSE-SENSE + gradient	0.772	31.38	0.936
29	LORAKS + gradient	0.765	31.17	0.933
30	L1-Wavelet + gradient	0.765	30.78	0.928
31	ESPIRiT + gradient	0.758	30.73	0.927
32	GRAPPA + gradient	0.729	29.26	0.905
33	Score-MRI + gradient	0.727	28.81	0.897
34	SENSE + gradient	0.695	27.3	0.865
35	Zero-Filled IFFT + gradient	0.651	24.81	0.796

Spec Ranges (3 parameters)

Parameter	Min	Max	Unit
contrast_timing_error	-0.6	1.2	s
background_suppression	-4.0	8.0	-
velocity_encoding_error	-3.0	6.0	-

View Details →

Dev 3 scenes

Blind evaluation tier — no ground truth available.

What you get & how to use

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.

Dev Leaderboard

#	Method	Score	PSNR	SSIM
1	HUMUS-Net++ + gradient	0.853	38.51	0.984
2	ReconFormer++ + gradient	0.839	37.63	0.981
3	SwinMR++ + gradient	0.826	37.22	0.979
4	ReconFormer + gradient	0.818	35.89	0.973
5	MRI-FM + gradient	0.814	35.1	0.968
6	HUMUS-Net + gradient	0.810	34.71	0.966
7	PnP-DnCNN-Pro + gradient	0.796	34.91	0.967
8	MR-IPT + gradient	0.795	34.51	0.964
9	MoDL-Net++ + gradient	0.788	34.4	0.964
10	SwinMR + gradient	0.779	32.88	0.951
11	MMR-Mamba + gradient	0.777	33.12	0.954
12	HybridCascade++ + gradient	0.772	32.6	0.949
13	BrainID-MRI + gradient	0.772	32.94	0.952
14	U-Net++ + gradient	0.769	33.03	0.953
15	MRI-DiffusionNet + gradient	0.769	32.37	0.947
16	PromptMR + gradient	0.767	31.56	0.938
17	E2E-VarNet + gradient	0.765	32.25	0.945
18	PromptMR-SFM + gradient	0.759	30.93	0.93
19	MRDynamo + gradient	0.754	30.71	0.927
20	U-Net + gradient	0.750	30.87	0.929
21	PnP-DnCNN + gradient	0.742	30.99	0.931
22	HybridCascade + gradient	0.728	28.97	0.9
23	MoDL + gradient	0.724	29.1	0.902
24	BM3D-MRI + gradient	0.708	28.64	0.894
25	GRAPPA + gradient	0.691	27.07	0.86
26	DCCNN + gradient	0.690	26.82	0.854
27	SENSE + gradient	0.685	27.19	0.863
28	Deep-ADMM-Net + gradient	0.678	27.02	0.859
29	k-t SPARSE-SENSE + gradient	0.647	25.1	0.805
30	LORAKS + gradient	0.639	24.56	0.788
31	ALOHA + gradient	0.637	25.01	0.803
32	L1-Wavelet + gradient	0.622	24.67	0.792
33	ESPIRiT + gradient	0.617	24.4	0.782
34	Zero-Filled IFFT + gradient	0.589	23.1	0.735
35	Score-MRI + gradient	0.580	23.06	0.733

Spec Ranges (3 parameters)

Parameter	Min	Max	Unit
contrast_timing_error	-0.72	1.08	s
background_suppression	-4.8	7.2	-
velocity_encoding_error	-3.6	5.4	-

Submit Reconstruction View Details →

Hidden 3 scenes

Fully blind server-side evaluation — no data download.

What you get & how to use

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.

Hidden Leaderboard

#	Method	Score	PSNR	SSIM
1	HUMUS-Net++ + gradient	0.824	36.53	0.976
2	HUMUS-Net + gradient	0.791	34.21	0.962
3	MRI-FM + gradient	0.790	33.67	0.958
4	ReconFormer++ + gradient	0.772	33.83	0.96
5	MoDL-Net++ + gradient	0.771	33.13	0.954
6	PnP-DnCNN-Pro + gradient	0.769	32.37	0.947
7	SwinMR++ + gradient	0.763	32.67	0.95
8	ReconFormer + gradient	0.760	32.92	0.952
9	MR-IPT + gradient	0.751	32.28	0.946
10	PromptMR + gradient	0.751	31.3	0.935
11	U-Net++ + gradient	0.750	30.55	0.925
12	HybridCascade++ + gradient	0.743	30.75	0.928
13	MRI-DiffusionNet + gradient	0.741	30.79	0.928
14	BrainID-MRI + gradient	0.734	30.83	0.929
15	E2E-VarNet + gradient	0.730	30.07	0.918
16	PnP-DnCNN + gradient	0.724	29.19	0.904
17	MMR-Mamba + gradient	0.721	30.41	0.923
18	SwinMR + gradient	0.709	28.25	0.886
19	PromptMR-SFM + gradient	0.707	28.16	0.884
20	MoDL + gradient	0.706	28.1	0.883
21	MRDynamo + gradient	0.703	29.15	0.903
22	GRAPPA + gradient	0.688	27.45	0.869
23	U-Net + gradient	0.684	28.18	0.885
24	HybridCascade + gradient	0.668	26.96	0.857
25	BM3D-MRI + gradient	0.665	25.91	0.829
26	SENSE + gradient	0.631	24.48	0.785
27	DCCNN + gradient	0.616	24.12	0.773
28	Deep-ADMM-Net + gradient	0.611	23.5	0.75
29	k-t SPARSE-SENSE + gradient	0.610	23.7	0.758
30	LORAKS + gradient	0.568	22.84	0.725
31	L1-Wavelet + gradient	0.562	21.86	0.684
32	ALOHA + gradient	0.555	22.27	0.701
33	Zero-Filled IFFT + gradient	0.550	21.29	0.659
34	ESPIRiT + gradient	0.529	21.31	0.66
35	Score-MRI + gradient	0.526	20.49	0.622

Spec Ranges (3 parameters)

Parameter	Min	Max	Unit
contrast_timing_error	-0.42	1.38	s
background_suppression	-2.8	9.2	-
velocity_encoding_error	-2.1	6.9	-

Submit Algorithm View Details →

Blind Reconstruction Challenge

Challenge

Given measurements with unknown mismatch and spec ranges (not exact params), reconstruct the original signal. A method must be evaluated on all three tiers for a complete score. Scored on a composite metric: 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖).

Input

Measurements y, ideal forward model H, spec ranges

Output

Reconstructed signal x̂

Spec DAG — Forward Model Pipeline

M → F → S → D

M Modulation

F Fourier

S Sampling

D Detector

Mismatch Parameters

Symbol	Parameter	Description	Perturbed
c_t	contrast_timing_error	Contrast timing error (s)	0.6
b_s	background_suppression	Background suppression (-)	4.0
v_e	velocity_encoding_error	Velocity encoding error (-)	3.0

Credits System

40%

Platform Profit Pool

Revenue allocated to benchmark rewards

30%

Winner Share

Top algorithm receives from pool

$100

Min Withdrawal

Minimum payout threshold

Spec Primitives Reference (11 primitives)

P Propagation

Free-space or medium propagation kernel (Fresnel, Rayleigh-Sommerfeld).

M Mask / Modulation

Spatial or spatio-temporal amplitude modulation (coded aperture, SLM pattern).

Π Projection

Geometric projection operator (Radon transform, fan-beam, cone-beam).

F Fourier Sampling

Sampling in the Fourier / k-space domain (MRI, ptychography).

C Convolution

Shift-invariant convolution with a point-spread function (PSF).

Σ Summation / Integration

Summation along a physical dimension (spectral, temporal, angular).

D Detector

Sensor readout with gain g and noise model η (Gaussian, Poisson, mixed).

S Structured Illumination

Patterned illumination (block, Hadamard, random) applied to the scene.

W Wavelength Dispersion

Spectral dispersion element (prism, grating) with shift α and aperture a.

R Rotation / Motion

Sample or gantry rotation (CT, electron tomography).

Λ Wavelength Selection

Spectral filter or monochromator selecting a wavelength band.