All datasets
Application requiredMRI

NYU fastMRI Knee Dataset

The NYU fastMRI Knee Dataset contains de-identified knee MRI data developed for research on accelerated MR image reconstruction. It includes raw k-space acquisitions and a separate collection of clinical DICOM knee MRI examinations.

Licence and access

Access requires an individual application and acceptance of the NYU fastMRI Dataset Sharing Agreement; use is limited to internal research or educational purposes and redistribution is prohibited.

application requiredcommercial use restrictedredistribution restricted
Visit official sourcelast verified 2026-09-03
Access status: Access requires an application, data-use agreement, registration workflow or institutional approval.
Coverage
kneeno labelsin vivo
subjects
not verified
images
not verified
formats
HDF5, DICOM
origin
USA · 2020
Citation
Knoll F, Zbontar J, Sriram A, et al. fastMRI: A Publicly Available Raw k-Space and DICOM Dataset of Knee Images for Accelerated MR Image Reconstruction Using Machine Learning. Radiology: Artificial Intelligence. 2020;2(1):e190007. DOI: 10.1148/ryai.2020190007.
Notes

The original publication reports 1,594 raw k-space measurement datasets and an additional 10,012 clinical DICOM image datasets from 9,290 patients. The current dataset page describes more than 1,500 fully sampled knee MRIs and lists 1,398 raw scans by contrast because its table excludes challenge data. The total number of unique subjects across both components and the total number of individual images could not be verified.

This record is a pointer. Licence and access terms are recorded as they stood on the verification date above; what binds you is the dataset's own licence text. Confirm at the source before use.

Orthopedic AI Research Updates

Monthly research digest, product updates, and clinical AI insights.

Unsubscribe anytime.

NYU fastMRI Knee Dataset, dataset record