We are sharing a collection of voluminous, expertly curated, and rigorously deidentified data sets for machine learning research on accelerating MRI.
The fastMRI dataset includes:
- raw data from 1,500 fully sampled knee MRIs
- DICOM images from 10,000 clinical knee MRIs
- raw data from nearly 7,000 fully sampled brain MRIs
- NEW raw data from 312 T2-weighted and diffusion-weighted prostate MRIs
The fastMRI project began as a collaborative partnership between CAI2R and Meta AI Research (formerly Facebook AI Research) to accelerate research on machine learning image reconstruction.
Related Publications
fastMRI: A Publicly Available Raw k-Space and DICOM Dataset of Knee Images for Accelerated MR Image Reconstruction Using Machine Learning.
Radiol Artif Intell. 2020 Jan 29;2(1):e190007. doi: 10.1148/ryai.2020190007
FastMRI Prostate: A public, biparametric MRI dataset to advance machine learning for prostate cancer imaging.
Sci Data. 2024 Apr 20;11(1):404. doi: 10.1038/s41597-024-03252-w
fastMRI: An Open Dataset and Benchmarks for Accelerated MRI.
arXiv:1811.08839 21 Nov 2018.
Please cite these works if you are using fastMRI data in your research (see section 6 of the data sharing agreement on the fastMRI Dataset download request page, linked below).
Related Stories
Radhika Tibrewala, graduate student in biomedical imaging, talks about the new fastMRI prostate dataset, deep learning in MRI, and how she started a PhD remotely in 2020.
Patricia Johnson, who researches machine learning image reconstruction, talks about faster MRI, visual preferences, and diagnostic interchangeability.
Florian Knoll, incoming chair of imaging at University of Erlangen–Nuremberg, talks about his background, not being greedy, and why he does what he does.