NYU fastMRI
Raw k-space and reconstructed MRI from clinical knee, brain, prostate, and breast exams, built for accelerated MRI reconstruction research
Open the source page ↗ See it in the directory
At a glance
- What it is
- Raw k-space and reconstructed MRI from clinical knee, brain, prostate, and breast exams, built for accelerated MRI reconstruction research
- Area
- MSK, Neuro, Body, Breast
- Modality
- MR
- Format
- DICOM, HDF5
- Size
- Over 1,500 raw knee and 10,000 DICOM knee exams, 6,970 raw brain exams, 312 prostate and 300 breast exams
- Normals included
- Partial (mixed or unlabeled)
- License
- NYU fastMRI Dataset Sharing Agreement (internal research or education only)
- Commercial use
- Not allowed under the free terms
- Access
- Free account and data-use terms required
- Status
- Verified: source page reviewed by Hounsfield Lab
- Added
- 2026-10-19
- Last checked
- 2026-11-11
- Source
- fastmri.med.nyu.edu/
How to access
Read the Dataset Sharing Agreement on the fastMRI page and submit the online application with your institutional affiliation and intended use. Approved users get a personal download link. Each sub-dataset has its own agreement.
Limitations
Internal research or education only. No redistribution, no sharing of the download link, no selling or monetizing, and no attempt to re-identify anyone; the agreement also asks you to destroy copies when your work is done. Most data are raw k-space for reconstruction, not diagnostic labels. DICOM is described for the knee set only. Not for clinical use.
Citation required?
Yes. The agreement requires citing “NYU fastMRI” and Knoll et al., Radiology: Artificial Intelligence 2020 (doi 10.1148/ryai.2020190007), plus arXiv:1811.08839.
Related open-source tools
- TotalSegmentator: Segments more than 100 anatomical structures in CT and MR
- TotalSpineSeg: Two-stage nnU-Net tool that segments and individually labels every vertebra and intervertebral disc plus the spinal cord and canal, using an odd/even vertebra scheme and landmark discs (C2-C3, C7-T1, T12-L1, L5-S) to avoid fused or split vertebrae; works on localizers and short-FOV scans; peer-reviewed in European Radiology (Nov 7, 2026)
- nnU-Net: Self-configuring segmentation framework
- 3D Slicer: Open platform for visualization, segmentation and image computing
- HD-BET: Deep-learning brain extraction (skull stripping) for MRI
- FastSurfer: Deep-learning whole-brain segmentation and cortical surface reconstruction, a fast alternative to FreeSurfer recon-all
- FreeSurfer: Neuroimaging analysis and visualization suite
More msk datasets
- LERA: DX, CR · Stanford AIMI research use agreement (non-commercial; commercial license sold separately)
- MRNet: MR · Stanford AIMI research use agreement (non-commercial; commercial license sold separately)
- MURA: DX, CR · Stanford AIMI research use agreement (non-commercial; commercial license sold separately)
- RSNA 2022 Cervical Spine Fracture Detection: CT · Kaggle competition rules (non-commercial)
- RSNA 2024 Lumbar Spine Degenerative Classification: MR · Kaggle competition rules (non-commercial)
- RSNA Knee Abnormality Detection (2026): MR · Kaggle competition rules (non-commercial)
Hounsfield Lab links to the source and does not host images. The license and access summary reflects the source page on the last-checked date; terms can change, so read the source before you download. Not legal advice.