MRS

MR Spectroscopy

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
🥇	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 0.848 Correct & Reconstruct →	0.848	0.908 42.34 dB / 0.992	0.837 37.32 dB / 0.979	0.798 34.26 dB / 0.963	✓ 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
🥈	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 0.846 Correct & Reconstruct →	0.846	0.877 39.62 dB / 0.987	0.842 38.09 dB / 0.982	0.820 35.72 dB / 0.972	✓ 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
🥉	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 0.838 Correct & Reconstruct →	0.838	0.900 41.55 dB / 0.991	0.840 37.74 dB / 0.981	0.775 33.17 dB / 0.954	✓ Certified	Fabian et al., dHUMUS-Net 2023 — k-space DC per module + dynamic multi-scale weighting + INR head + perceptual-structural loss + axial attention
4	ReconFormer++ + gradient ReconFormer++ + gradient Pan et al., IEEE TMI 2025 Score 0.835 Correct & Reconstruct →	0.835	0.862 39.19 dB / 0.986	0.841 36.99 dB / 0.978	0.801 35.47 dB / 0.971	✓ Certified	Pan et al., IEEE TMI 2025
5	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 0.833 Correct & Reconstruct →	0.833	0.894 41.03 dB / 0.990	0.817 35.0 dB / 0.968	0.787 33.14 dB / 0.954	✓ 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 MR-IPT + gradient Sci. Reports 2025 Score 0.829 Correct & Reconstruct →	0.829	0.894 41.16 dB / 0.990	0.810 35.74 dB / 0.972	0.784 34.24 dB / 0.963	✓ Certified	Sci. Reports 2025
7	BrainID-MRI + gradient BrainID-MRI + gradient Liu et al., CVPR 2025 Score 0.808 Correct & Reconstruct →	0.808	0.858 39.2 dB / 0.986	0.789 34.51 dB / 0.964	0.776 32.31 dB / 0.946	✓ Certified	Liu et al., CVPR 2025
8	MRI-FM + gradient MRI-FM + gradient Wang et al., Nature MI 2026 Score 0.807 Correct & Reconstruct →	0.807	0.871 40.31 dB / 0.989	0.797 34.53 dB / 0.965	0.752 32.2 dB / 0.945	✓ Certified	Wang et al., Nature MI 2026
9	MRDynamo + gradient MRDynamo + gradient Chen et al., NeurIPS 2024 Score 0.793 Correct & Reconstruct →	0.793	0.853 38.57 dB / 0.984	0.777 33.58 dB / 0.958	0.748 32.04 dB / 0.943	✓ Certified	Chen et al., NeurIPS 2024
10	ReconFormer + gradient ReconFormer + gradient Guo et al., IEEE TMI 2024 Score 0.791 Correct & Reconstruct →	0.791	0.855 37.81 dB / 0.981	0.780 32.76 dB / 0.950	0.739 30.02 dB / 0.917	✓ Certified	Guo et al., IEEE TMI 2024
11	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 0.786 Correct & Reconstruct →	0.786	0.868 39.48 dB / 0.987	0.753 31.89 dB / 0.941	0.737 30.49 dB / 0.924	✓ 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 MMR-Mamba + gradient Zhao et al., Med. Image Anal. 2025 Score 0.784 Correct & Reconstruct →	0.784	0.858 39.15 dB / 0.986	0.765 31.71 dB / 0.939	0.728 29.32 dB / 0.906	✓ Certified	Zhao et al., Med. Image Anal. 2025
13	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 0.779 Correct & Reconstruct →	0.779	0.885 40.49 dB / 0.989	0.761 31.31 dB / 0.935	0.690 28.4 dB / 0.889	✓ 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
14	SwinMR + gradient SwinMR + gradient Huang et al., MICCAI 2022 Score 0.775 Correct & Reconstruct →	0.775	0.849 36.95 dB / 0.978	0.758 32.05 dB / 0.943	0.718 29.35 dB / 0.906	✓ Certified	Huang et al., MICCAI 2022
15	MRI-DiffusionNet + gradient MRI-DiffusionNet + gradient Song et al., ICCV 2024 Score 0.775 Correct & Reconstruct →	0.775	0.848 37.76 dB / 0.981	0.752 31.54 dB / 0.937	0.726 29.93 dB / 0.916	✓ Certified	Song et al., ICCV 2024
16	PromptMR + gradient PromptMR + gradient Bai et al., ECCV 2024 Score 0.773 Correct & Reconstruct →	0.773	0.862 38.31 dB / 0.983	0.749 31.18 dB / 0.933	0.709 28.37 dB / 0.888	✓ Certified	Bai et al., ECCV 2024
17	PromptMR-SFM + gradient PromptMR-SFM + gradient PWM 2026 Score 0.772 Correct & Reconstruct →	0.772	0.863 39.41 dB / 0.986	0.759 31.69 dB / 0.939	0.695 28.62 dB / 0.893	✓ Certified	PWM 2026
18	E2E-VarNet + gradient E2E-VarNet + gradient Sriram et al., MICCAI 2020 Score 0.764 Correct & Reconstruct →	0.764	0.837 36.65 dB / 0.977	0.754 30.46 dB / 0.924	0.702 28.55 dB / 0.892	✓ Certified	Sriram et al., MICCAI 2020
19	HUMUS-Net + gradient HUMUS-Net + gradient Fabian et al., NeurIPS 2022 Score 0.758 Correct & Reconstruct →	0.758	0.834 36.72 dB / 0.977	0.755 30.75 dB / 0.928	0.685 28.17 dB / 0.884	✓ Certified	Fabian et al., NeurIPS 2022
20	PnP-DnCNN + gradient PnP-DnCNN + gradient Ahmad et al., IEEE TCI 2019 (DOI:10.1109/TCI.2019.2944521) Score 0.745 Correct & Reconstruct →	0.745	0.805 33.46 dB / 0.957	0.738 29.44 dB / 0.908	0.692 27.48 dB / 0.869	✓ Certified	Ahmad et al., IEEE TCI 2019 (DOI:10.1109/TCI.2019.2944521)
21	BM3D-MRI + gradient BM3D-MRI + gradient Eksioglu, Comput. Math. Meth. Med. 2016 Score 0.737 Correct & Reconstruct →	0.737	0.771 31.62 dB / 0.938	0.728 29.12 dB / 0.902	0.713 29.04 dB / 0.901	✓ Certified	Eksioglu, Comput. Math. Meth. Med. 2016
22	MoDL + gradient MoDL + gradient Aggarwal et al., IEEE TMI 2019 Score 0.728 Correct & Reconstruct →	0.728	0.827 35.47 dB / 0.971	0.696 28.28 dB / 0.887	0.662 25.88 dB / 0.829	✓ Certified	Aggarwal et al., IEEE TMI 2019
23	Deep-ADMM-Net + gradient Deep-ADMM-Net + gradient Yang et al., NeurIPS 2016 Score 0.725 Correct & Reconstruct →	0.725	0.808 33.66 dB / 0.958	0.696 27.47 dB / 0.869	0.671 27.09 dB / 0.860	✓ Certified	Yang et al., NeurIPS 2016
24	GRAPPA + gradient GRAPPA + gradient Griswold et al., MRM 2002 Score 0.721 Correct & Reconstruct →	0.721	0.729 29.26 dB / 0.905	0.712 29.24 dB / 0.904	0.723 29.51 dB / 0.909	✓ Certified	Griswold et al., MRM 2002
25	U-Net + gradient U-Net + gradient Zbontar et al., arXiv 2018 Score 0.720 Correct & Reconstruct →	0.720	0.818 34.8 dB / 0.966	0.694 27.16 dB / 0.862	0.647 25.72 dB / 0.824	✓ Certified	Zbontar et al., arXiv 2018
26	HybridCascade + gradient HybridCascade + gradient Fastmri, arXiv 2020 Score 0.719 Correct & Reconstruct →	0.719	0.821 35.81 dB / 0.972	0.691 27.29 dB / 0.865	0.645 25.27 dB / 0.811	✓ Certified	Fastmri, arXiv 2020
27	DCCNN + gradient DCCNN + gradient Schlemper et al., IEEE TMI 2018 Score 0.712 Correct & Reconstruct →	0.712	0.811 33.78 dB / 0.959	0.697 27.73 dB / 0.875	0.629 24.33 dB / 0.780	✓ Certified	Schlemper et al., IEEE TMI 2018
28	ALOHA + gradient ALOHA + gradient Jin et al., IEEE TMI 2016 Score 0.705 Correct & Reconstruct →	0.705	0.778 32.52 dB / 0.948	0.709 28.38 dB / 0.889	0.629 25.38 dB / 0.814	✓ Certified	Jin et al., IEEE TMI 2016
29	ESPIRiT + gradient ESPIRiT + gradient Uecker et al., MRM 2014 Score 0.679 Correct & Reconstruct →	0.679	0.787 32.37 dB / 0.947	0.670 26.34 dB / 0.841	0.579 22.65 dB / 0.717	✓ Certified	Uecker et al., MRM 2014
30	LORAKS + gradient LORAKS + gradient Haldar, IEEE TMI 2014 Score 0.670 Correct & Reconstruct →	0.670	0.789 32.27 dB / 0.946	0.647 25.36 dB / 0.813	0.575 22.95 dB / 0.729	✓ Certified	Haldar, IEEE TMI 2014
31	SENSE + gradient SENSE + gradient Pruessmann et al., MRM 1999 Score 0.666 Correct & Reconstruct →	0.666	0.718 27.79 dB / 0.876	0.665 25.63 dB / 0.821	0.614 24.43 dB / 0.783	✓ Certified	Pruessmann et al., MRM 1999
32	L1-Wavelet + gradient L1-Wavelet + gradient Lustig et al., MRM 2007 Score 0.639 Correct & Reconstruct →	0.639	0.740 29.67 dB / 0.912	0.640 25.07 dB / 0.804	0.537 21.03 dB / 0.647	✓ Certified	Lustig et al., MRM 2007
33	k-t SPARSE-SENSE + gradient k-t SPARSE-SENSE + gradient Lustig et al., MRM 2006 Score 0.621 Correct & Reconstruct →	0.621	0.746 30.2 dB / 0.920	0.584 22.36 dB / 0.705	0.532 20.99 dB / 0.645	✓ Certified	Lustig et al., MRM 2006
34	Score-MRI + gradient Score-MRI + gradient Chung & Ye, Med. Image Anal. 2022 Score 0.618 Correct & Reconstruct →	0.618	0.726 28.64 dB / 0.894	0.596 23.46 dB / 0.749	0.533 21.52 dB / 0.669	✓ Certified	Chung & Ye, Med. Image Anal. 2022
35	Zero-Filled IFFT + gradient Zero-Filled IFFT + gradient Pruessmann et al., MRM 1999 Score 0.616 Correct & Reconstruct →	0.616	0.621 23.97 dB / 0.767	0.628 24.54 dB / 0.787	0.598 23.55 dB / 0.752	✓ 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	0.908	42.34	0.992
2	HUMUS-Net++ + gradient	0.900	41.55	0.991
3	HybridCascade++ + gradient	0.894	41.03	0.99
4	MR-IPT + gradient	0.894	41.16	0.99
5	U-Net++ + gradient	0.885	40.49	0.989
6	PnP-DnCNN-Pro + gradient	0.877	39.62	0.987
7	MRI-FM + gradient	0.871	40.31	0.989
8	MoDL-Net++ + gradient	0.868	39.48	0.987
9	PromptMR-SFM + gradient	0.863	39.41	0.986
10	ReconFormer++ + gradient	0.862	39.19	0.986
11	PromptMR + gradient	0.862	38.31	0.983
12	BrainID-MRI + gradient	0.858	39.2	0.986
13	MMR-Mamba + gradient	0.858	39.15	0.986
14	ReconFormer + gradient	0.855	37.81	0.981
15	MRDynamo + gradient	0.853	38.57	0.984
16	SwinMR + gradient	0.849	36.95	0.978
17	MRI-DiffusionNet + gradient	0.848	37.76	0.981
18	E2E-VarNet + gradient	0.837	36.65	0.977
19	HUMUS-Net + gradient	0.834	36.72	0.977
20	MoDL + gradient	0.827	35.47	0.971
21	HybridCascade + gradient	0.821	35.81	0.972
22	U-Net + gradient	0.818	34.8	0.966
23	DCCNN + gradient	0.811	33.78	0.959
24	Deep-ADMM-Net + gradient	0.808	33.66	0.958
25	PnP-DnCNN + gradient	0.805	33.46	0.957
26	LORAKS + gradient	0.789	32.27	0.946
27	ESPIRiT + gradient	0.787	32.37	0.947
28	ALOHA + gradient	0.778	32.52	0.948
29	BM3D-MRI + gradient	0.771	31.62	0.938
30	k-t SPARSE-SENSE + gradient	0.746	30.2	0.92
31	L1-Wavelet + gradient	0.740	29.67	0.912
32	GRAPPA + gradient	0.729	29.26	0.905
33	Score-MRI + gradient	0.726	28.64	0.894
34	SENSE + gradient	0.718	27.79	0.876
35	Zero-Filled IFFT + gradient	0.621	23.97	0.767

Spec Ranges (4 parameters)

Parameter	Min	Max	Unit
linewidth	-2.0	4.0	Hz
freq_drift	-1.5	3.0	Hz
phase_error	-5.0	10.0	deg
baseline	-0.05	0.1

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	PnP-DnCNN-Pro + gradient	0.842	38.09	0.982
2	ReconFormer++ + gradient	0.841	36.99	0.978
3	HUMUS-Net++ + gradient	0.840	37.74	0.981
4	SwinMR++ + gradient	0.837	37.32	0.979
5	HybridCascade++ + gradient	0.817	35.0	0.968
6	MR-IPT + gradient	0.810	35.74	0.972
7	MRI-FM + gradient	0.797	34.53	0.965
8	BrainID-MRI + gradient	0.789	34.51	0.964
9	ReconFormer + gradient	0.780	32.76	0.95
10	MRDynamo + gradient	0.777	33.58	0.958
11	MMR-Mamba + gradient	0.765	31.71	0.939
12	U-Net++ + gradient	0.761	31.31	0.935
13	PromptMR-SFM + gradient	0.759	31.69	0.939
14	SwinMR + gradient	0.758	32.05	0.943
15	HUMUS-Net + gradient	0.755	30.75	0.928
16	E2E-VarNet + gradient	0.754	30.46	0.924
17	MoDL-Net++ + gradient	0.753	31.89	0.941
18	MRI-DiffusionNet + gradient	0.752	31.54	0.937
19	PromptMR + gradient	0.749	31.18	0.933
20	PnP-DnCNN + gradient	0.738	29.44	0.908
21	BM3D-MRI + gradient	0.728	29.12	0.902
22	GRAPPA + gradient	0.712	29.24	0.904
23	ALOHA + gradient	0.709	28.38	0.889
24	DCCNN + gradient	0.697	27.73	0.875
25	MoDL + gradient	0.696	28.28	0.887
26	Deep-ADMM-Net + gradient	0.696	27.47	0.869
27	U-Net + gradient	0.694	27.16	0.862
28	HybridCascade + gradient	0.691	27.29	0.865
29	ESPIRiT + gradient	0.670	26.34	0.841
30	SENSE + gradient	0.665	25.63	0.821
31	LORAKS + gradient	0.647	25.36	0.813
32	L1-Wavelet + gradient	0.640	25.07	0.804
33	Zero-Filled IFFT + gradient	0.628	24.54	0.787
34	Score-MRI + gradient	0.596	23.46	0.749
35	k-t SPARSE-SENSE + gradient	0.584	22.36	0.705

Spec Ranges (4 parameters)

Parameter	Min	Max	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

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	PnP-DnCNN-Pro + gradient	0.820	35.72	0.972
2	ReconFormer++ + gradient	0.801	35.47	0.971
3	SwinMR++ + gradient	0.798	34.26	0.963
4	HybridCascade++ + gradient	0.787	33.14	0.954
5	MR-IPT + gradient	0.784	34.24	0.963
6	BrainID-MRI + gradient	0.776	32.31	0.946
7	HUMUS-Net++ + gradient	0.775	33.17	0.954
8	MRI-FM + gradient	0.752	32.2	0.945
9	MRDynamo + gradient	0.748	32.04	0.943
10	ReconFormer + gradient	0.739	30.02	0.917
11	MoDL-Net++ + gradient	0.737	30.49	0.924
12	MMR-Mamba + gradient	0.728	29.32	0.906
13	MRI-DiffusionNet + gradient	0.726	29.93	0.916
14	GRAPPA + gradient	0.723	29.51	0.909
15	SwinMR + gradient	0.718	29.35	0.906
16	BM3D-MRI + gradient	0.713	29.04	0.901
17	PromptMR + gradient	0.709	28.37	0.888
18	E2E-VarNet + gradient	0.702	28.55	0.892
19	PromptMR-SFM + gradient	0.695	28.62	0.893
20	PnP-DnCNN + gradient	0.692	27.48	0.869
21	U-Net++ + gradient	0.690	28.4	0.889
22	HUMUS-Net + gradient	0.685	28.17	0.884
23	Deep-ADMM-Net + gradient	0.671	27.09	0.86
24	MoDL + gradient	0.662	25.88	0.829
25	U-Net + gradient	0.647	25.72	0.824
26	HybridCascade + gradient	0.645	25.27	0.811
27	DCCNN + gradient	0.629	24.33	0.78
28	ALOHA + gradient	0.629	25.38	0.814
29	SENSE + gradient	0.614	24.43	0.783
30	Zero-Filled IFFT + gradient	0.598	23.55	0.752
31	ESPIRiT + gradient	0.579	22.65	0.717
32	LORAKS + gradient	0.575	22.95	0.729
33	L1-Wavelet + gradient	0.537	21.03	0.647
34	Score-MRI + gradient	0.533	21.52	0.669
35	k-t SPARSE-SENSE + gradient	0.532	20.99	0.645

Spec Ranges (4 parameters)

Parameter	Min	Max	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

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

About the Imaging Modality

Magnetic resonance spectroscopy (MRS) measures the concentration of metabolites in a localized tissue volume by exploiting the chemical shift — the slight difference in Larmor frequency caused by the electronic environment of different molecular groups. The free induction decay (FID) or spin echo signal is Fourier-transformed to a spectrum where each metabolite produces characteristic peaks (e.g. NAA at 2.01 ppm, Cr at 3.03 ppm). Quantification involves fitting the spectrum to a linear combination of basis spectra (LCModel, OSPREY). Challenges include low SNR, spectral overlap, water/lipid suppression, and B0 inhomogeneity causing linewidth broadening.

Principle

MR Spectroscopy measures the chemical shift spectrum of nuclear spins (usually ¹H) from a localized volume in the body, providing concentrations of metabolites such as NAA, creatine, choline, lactate, myo-inositol, and glutamate/glutamine. Chemical shift differences (in ppm) arise from the varying electronic shielding of nuclei in different molecular environments.

How to Build the System

Use PRESS or STEAM single-voxel localization on a 1.5T or 3T scanner. Voxel sizes are typically 2×2×2 cm³ for brain. Suppress the dominant water signal (CHESS or VAPOR water suppression). Acquire 64-256 averages (NEX) for adequate SNR. Shimming is critical: water linewidth should be <12 Hz (3T) for the voxel. Multi-voxel CSI (Chemical Shift Imaging) maps metabolite distributions but requires longer acquisition and careful lipid suppression.

Common Reconstruction Algorithms

LCModel (frequency-domain linear combination fitting)
TARQUIN (open-source time-domain fitting)
jMRUI (time-domain quantification with AMARES/QUEST)
HSVD (Hankel SVD) for water removal and baseline correction
Deep-learning spectral quantification (DeepSpectra, convolutional fitting)

Common Mistakes

Poor shimming producing broad linewidths that overlap metabolite peaks
Voxel placed partly outside the brain, contaminating spectrum with lipid signal
Insufficient water suppression saturating the spectrum baseline
Too few averages, producing noisy spectra with unreliable metabolite estimates
Ignoring macromolecular baseline contributions in fitting

How to Avoid Mistakes

Iteratively shim the voxel to achieve <12 Hz water linewidth (3T) before acquisition
Place the voxel with margin from skull and subcutaneous fat; use outer-volume suppression
Optimize water suppression parameters; acquire separate water reference for quantification
Acquire sufficient averages: 128-256 for metabolites at low concentration (e.g., GABA)
Include macromolecular basis set or measured baseline in the fitting model

Forward-Model Mismatch Cases

The widefield fallback produces a spatial image, but MR Spectroscopy acquires frequency-domain spectra encoding chemical composition — metabolite peaks (NAA, choline, creatine, lactate) at specific ppm values are entirely absent
MRS data is a 1D free induction decay (FID) or spectrum per voxel, not a 2D spatial image — the widefield blur destroys the spectral dimension that encodes metabolite concentrations

How to Correct the Mismatch

Use the MRS operator that models the free induction decay: y(t) = sum_k(a_k * exp(i*2pi*f_k*t) * exp(-t/T2_k)) for each metabolite k, then FFT to produce the frequency spectrum
Quantify metabolite concentrations by fitting the spectrum (LCModel, TARQUIN) or using deep-learning spectral quantification with the correctly modeled spectral forward model

Experimental Setup — Signal Chain

Experimental setup diagram for MR Spectroscopy

Reconstruction Gallery — 4 Scenes × 3 Scenarios

Method: CPU_baseline | Mismatch: nominal (nominal=True, perturbed=False)

I Ideal II Mismatch (uncorrected) III Oracle correction

I — Ideal

Ground Truth

Measurement

Reconstruction

16.584200964385797 dB SSIM 0.3250905580130585

II — Mismatch

Ground Truth

Measurement

Reconstruction

12.506523882319042 dB SSIM 0.044716474341606735

III — Oracle

Ground Truth

Measurement (perturbed)

Reconstruction

15.931719256055723 dB SSIM 0.12142227244746187

I — Ideal

Ground Truth

Measurement

Reconstruction

18.884299556167566 dB SSIM 0.3252411352628261

II — Mismatch

Ground Truth

Measurement

Reconstruction

13.742128073130111 dB SSIM 0.048029248758291385

III — Oracle

Ground Truth

Measurement (perturbed)

Reconstruction

17.02391116326154 dB SSIM 0.11622117666622035

I — Ideal

Ground Truth

Measurement

Reconstruction

18.301830801494646 dB SSIM 0.32976160086547746

II — Mismatch

Ground Truth

Measurement

Reconstruction

13.36792873190369 dB SSIM 0.04964364838717723

III — Oracle

Ground Truth

Measurement (perturbed)

Reconstruction

16.550610792224443 dB SSIM 0.1259030967011571

I — Ideal

Ground Truth

Measurement

Reconstruction

16.67775192548055 dB SSIM 0.3247822274360334

II — Mismatch

Ground Truth

Measurement

Reconstruction

12.53195005733068 dB SSIM 0.044966709218701925

III — Oracle

Ground Truth

Measurement (perturbed)

Reconstruction

15.850507045662987 dB SSIM 0.12621817875162467

Mean PSNR Across All Scenes

17.61202081188214 dB

I — Ideal

13.037132686170882 dB

II — Mismatch

16.33918706430117 dB

III — Oracle

Degradation: -4.6 dB Recovery: +3.3 dB

Per-scene PSNR breakdown (4 scenes)

Scene	I (PSNR)	I (SSIM)	II (PSNR)	II (SSIM)	III (PSNR)	III (SSIM)
scene_00	16.584200964385797	0.3250905580130585	12.506523882319042	0.044716474341606735	15.931719256055723	0.12142227244746187
scene_01	18.884299556167566	0.3252411352628261	13.742128073130111	0.048029248758291385	17.02391116326154	0.11622117666622035
scene_02	18.301830801494646	0.32976160086547746	13.36792873190369	0.04964364838717723	16.550610792224443	0.1259030967011571
scene_03	16.67775192548055	0.3247822274360334	12.53195005733068	0.044966709218701925	15.850507045662987	0.12621817875162467
Mean	17.61202081188214	0.3262188803943489	13.037132686170882	0.046839020176444326	16.33918706430117	0.122441181141616

Experimental Setup

Instrument: Siemens MAGNETOM Prisma 3T

Sequence: PRESS (Point RESolved Spectroscopy)

Te Ms: 30

Tr Ms: 2000

Voxel Size Cm3: 2x2x2 (8 mL)

Transients: 64

Metabolites: ['NAA', 'Cr', 'Cho', 'Glx', 'mI']

Fitting: LCModel / OSPREY

Key References

Provencher, 'Estimation of metabolite concentrations from localized in vivo proton NMR spectra (LCModel)', MRM 30, 672-679 (1993)
Wilson et al., 'Methodological consensus on clinical proton MRS of the brain (MRSinMRS)', NMR in Biomedicine 34, e4484 (2021)

Canonical Datasets

ISMRM MRS fitting challenge datasets
Big GABA multi-site MRS data

Spec DAG — Forward Model Pipeline

F(FID) → D(g, η₁)

F Free Induction Decay (FID)

D RF Coil Receiver (g, η₁)

Mismatch Parameters

Symbol	Parameter	Description	Perturbed
Δν	linewidth	Linewidth broadening (Hz)	2.0
Δf	freq_drift	Frequency drift (Hz)	1.5
Δφ	phase_error	Zero-order phase error (deg)	5.0
B	baseline	Baseline distortion amplitude	0.05

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.