{"url":"/dataset/fastmri","name":"fastMRI","full_name":null,"description_markdown":"The **fastMRI** dataset includes two types of MRI scans: knee MRIs and the brain (neuro) MRIs, and containing training, validation, and masked test sets.\r\nThe deidentified imaging dataset provided by NYU Langone comprises raw k-space data in several sub-dataset groups. Curation of these data are part of an IRB approved study. Raw and DICOM data have been deidentified via conversion to the vendor-neutral ISMRMD format and the RSNA clinical trial processor, respectively. Also, each DICOM image is manually inspected for the presence of any unexpected protected health information (PHI), with spot checking of both metadata and image content.\r\n**Knee MRI**: Data from more than 1,500 fully sampled knee MRIs obtained on 3 and 1.5 Tesla magnets and DICOM images from 10,000 clinical knee MRIs also obtained at 3 or 1.5 Tesla. The raw dataset includes coronal proton density-weighted images with and without fat suppression. The DICOM dataset contains coronal proton density-weighted with and without fat suppression, axial proton density-weighted with fat suppression, sagittal proton density, and sagittal T2-weighted with fat suppression.\r\n**Brain MRI**: Data from 6,970 fully sampled brain MRIs obtained on 3 and 1.5 Tesla magnets. The raw dataset includes axial T1 weighted, T2 weighted and FLAIR images. Some of the T1 weighted acquisitions included admissions of contrast agent.\r\n\r\nSource: [https://fastmri.med.nyu.edu/](https://fastmri.med.nyu.edu/)\r\nImage Source: [https://fastmri.med.nyu.edu/](https://fastmri.med.nyu.edu/)","description_withheld":null,"homepage":"https://fastmri.med.nyu.edu/","introduced_date":null,"introduced_date_note":null,"introduced_by":{"paper":"/paper/fastmri-an-open-dataset-and-benchmarks-for","title":"fastMRI: An Open Dataset and Benchmarks for Accelerated MRI","first_author":"Jure Zbontar","url":null},"license":{"name":"Custom (internal research-only)","url":"https://fastmri.med.nyu.edu/"},"modalities":[{"name":"Images","url":"/datasets/modality/images"},{"name":"Medical","url":"/datasets/modality/medical"},{"name":"MRI","url":"/datasets/modality/mri"}],"tasks":[{"name":"MRI Reconstruction","url":"/task/mri-reconstruction","datasets_with_task":"/datasets/task/mri-reconstruction"}],"languages":[],"variants":["fastMRI","fastMRI Knee 4x","fastMRI Knee 8x","fastMRI Brain 4x","fastMRI Brain 8x","fastMRI Knee Val 8x "],"data_loaders":[],"num_papers_in_archive":332,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/mri-reconstruction-on-fastmri-knee-8x","task":"MRI Reconstruction","dataset_variant":"fastMRI Knee 8x","rows":4,"metrics":["SSIM","PSNR"],"first_row_in_archive_order":{"model":"HUMUS-Net (train+val data)","paper":"/paper/humus-net-hybrid-unrolled-multi-scale-network","metrics":{"PSNR":"37.3","SSIM":"0.8945"},"code_links":[{"title":"z-fabian/HUMUS-Net","url":"https://github.com/z-fabian/HUMUS-Net"},{"title":"MathFLDS/HUMUS-Net","url":"https://github.com/MathFLDS/HUMUS-Net"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/mri-reconstruction-on-fastmri-knee-val-8x","task":"MRI Reconstruction","dataset_variant":"fastMRI Knee Val 8x","rows":4,"metrics":["SSIM","PSNR","NMSE","Params (M)"],"first_row_in_archive_order":{"model":"PromptMR","paper":"/paper/fill-the-k-space-and-refine-the-image","metrics":{"NMSE":"0.0080","PSNR":"37.78","Params (M)":"80","SSIM":"0.8983"},"code_links":[{"title":"hellopipu/promptmr","url":"https://github.com/hellopipu/promptmr"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/mri-reconstruction-on-fastmri-brain-4x","task":"MRI Reconstruction","dataset_variant":"fastMRI Brain 4x","rows":2,"metrics":["SSIM","PSNR"],"first_row_in_archive_order":{"model":"End-to-end variational network","paper":"/paper/end-to-end-variational-networks-for","metrics":{"PSNR":"41","SSIM":"0.959"},"code_links":[{"title":"facebookresearch/fastMRI","url":"https://github.com/facebookresearch/fastMRI"},{"title":"z-fabian/MRAugment","url":"https://github.com/z-fabian/MRAugment"},{"title":"MathFLDS/MRAugment","url":"https://github.com/MathFLDS/MRAugment"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/mri-reconstruction-on-fastmri-brain-8x","task":"MRI Reconstruction","dataset_variant":"fastMRI Brain 8x","rows":2,"metrics":["SSIM","PSNR"],"first_row_in_archive_order":{"model":"End-to-end variational network","paper":"/paper/end-to-end-variational-networks-for","metrics":{"PSNR":"38","SSIM":"0.943"},"code_links":[{"title":"facebookresearch/fastMRI","url":"https://github.com/facebookresearch/fastMRI"},{"title":"z-fabian/MRAugment","url":"https://github.com/z-fabian/MRAugment"},{"title":"MathFLDS/MRAugment","url":"https://github.com/MathFLDS/MRAugment"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/mri-reconstruction-on-fastmri-knee-4x","task":"MRI Reconstruction","dataset_variant":"fastMRI Knee 4x","rows":2,"metrics":["SSIM","PSNR"],"first_row_in_archive_order":{"model":"End-to-end variational network","paper":"/paper/end-to-end-variational-networks-for","metrics":{"PSNR":"40","SSIM":"0.930"},"code_links":[{"title":"facebookresearch/fastMRI","url":"https://github.com/facebookresearch/fastMRI"},{"title":"z-fabian/MRAugment","url":"https://github.com/z-fabian/MRAugment"},{"title":"MathFLDS/MRAugment","url":"https://github.com/MathFLDS/MRAugment"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/fill-the-k-space-and-refine-the-image","title":"Fill the K-Space and Refine the Image: Prompting for Dynamic and Multi-Contrast MRI Reconstruction","date":"2023-09-25","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/humus-net-hybrid-unrolled-multi-scale-network","title":"HUMUS-Net: Hybrid unrolled multi-scale network architecture for accelerated MRI reconstruction","date":"2022-03-15","rows_on_this_dataset":4,"code_links":2,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":5,"samples_ran":5,"samples_unverified":0,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/xpdnet-for-mri-reconstruction-an-application","title":"XPDNet for MRI Reconstruction: an application to the 2020 fastMRI challenge","date":"2020-10-15","rows_on_this_dataset":4,"code_links":3,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":1,"samples_ran":1,"samples_unverified":0,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/end-to-end-variational-networks-for","title":"End-to-End Variational Networks for Accelerated MRI Reconstruction","date":"2020-04-14","rows_on_this_dataset":5,"code_links":3,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":5,"samples_ran":1,"samples_unverified":4,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":3,"samples_harvested":11,"samples_ran":7,"samples_unverified":4,"pointer_only_for_licence":0,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}