fastMRI

Website and Leaderboards | Dataset | GitHub | Publications

Accelerating Magnetic Resonance Imaging (MRI) by acquiring fewer measurements has the potential to reduce medical costs, minimize stress to patients and make MR imaging possible in applications where it is currently prohibitively slow or expensive.

fastMRI is a collaborative research project from Facebook AI Research (FAIR) and NYU Langone Health to investigate the use of AI to make MRI scans faster. NYU Langone Health has released fully anonymized knee and brain MRI datasets that can be downloaded from the fastMRI dataset page. Publications associated with the fastMRI project can be found at the end of this README.

This repository contains convenient PyTorch data loaders, subsampling functions, evaluation metrics, and reference implementations of simple baseline methods. It also contains implementations for methods in some of the publications of the fastMRI project.

Documentation

The fastMRI Dataset

There are multiple publications describing different subcomponents of the data (e.g., brain vs. knee) and associated baselines.

• Project Summary, Datasets, Baselines: fastMRI: An Open Dataset and Benchmarks for Accelerated MRI ({J. Zbontar*, F. Knoll*, A. Sriram*} et al., 2018)

• Knee Data: fastMRI: A Publicly Available Raw k-Space and DICOM Dataset of Knee Images for Accelerated MR Image Reconstruction Using Machine Learning ({F. Knoll*, J. Zbontar*} et al., 2020)

• Brain Dataset Properties: Supplemental Material of Results of the 2020 fastMRI Challenge for Machine Learning MR Image Reconstruction ({M. Muckley*, B. Riemenschneider*} et al., 2021)

Code Repository

For code documentation, most functions and classes have accompanying docstrings that you can access via the help function in IPython. For example:

from fastmri.data import SliceDataset

help(SliceDataset)

Dependencies and Installation

Note: Contributions to the code are continuously tested via GitHub actions. If you encounter an issue, the best first thing to do is to try to match the test environments in requirements.txt and dev-requirements.txt.

Note: As documented in Issue 215, there is currently a memory leak when using h5py installed from pip and converting to a torch.Tensor. To avoid the leak, you need to use h5py with a version of HDF5 before 1.12.1. As of February 16, 2022, the conda version of h5py 3.6.0 used HDF5 1.10.6, which avoids the leak.

First install PyTorch according to the directions at the PyTorch Website for your operating system and CUDA setup. Then, run

pip install fastmri

pip will handle all package dependencies. After this you should be able to run most of the code in the repository.

Installing Directly from Source

If you want to install directly from the GitHub source, clone the repository, navigate to the fastmri root directory and run

pip install -e .

Package Structure & Usage

The repository is centered around the fastmri module. The following breaks down the basic structure:

fastmri: Contains a number of basic tools for complex number math, coil combinations, etc.

• fastmri.data: Contains data utility functions from original data folder that can be used to create sampling masks and submission files.
• fastmri.models: Contains reconstruction models, such as the U-Net and VarNet.
• fastmri.pl_modules: PyTorch Lightning modules for data loading, training, and logging.

Examples and Reproducibility

The fastmri_examples and banding_removal folders include code for reproducibility. The baseline models were used in the arXiv paper.

A brief summary of implementions based on papers with links to code follows. For completeness we also mention work on active acquisition, which is hosted in another repository.

• Baseline Models

  • Zero-filled examples for saving images for leaderboard submission
  • ESPIRiT—an eigenvalue approach to autocalibrating parallel MRI: where SENSE meets GRAPPA (M. Uecker et al., 2013)
  • U-Net: Convolutional networks for biomedical image segmentation (O. Ronneberger et al., 2015)

• Sampling, Reconstruction and Artifact Correction

  • Offset Sampling Improves Deep Learning based Accelerated MRI Reconstructions by Exploiting Symmetry (A. Defazio, 2019)
  • End-to-End Variational Networks for Accelerated MRI Reconstruction ({A. Sriram*, J. Zbontar*} et al., 2020)
  • MRI Banding Removal via Adversarial Training (A. Defazio, et al., 2020)

• Active Acquisition (external repository)

  • (external repository) Reducing uncertainty in undersampled MRI reconstruction with active acquisition (Z. Zhang et al., 2019)
  • (external repository) Active MR k-space Sampling with Reinforcement Learning (L. Pineda et al., 2020)
  • On learning adaptive acquisition policies for undersampled multi-coil MRI reconstruction (T. Bakker et al., 2022)

Testing

Run pytest tests. By default integration tests that use the fastMRI data are skipped. If you would like to run these tests, set SKIP_INTEGRATIONS to False in the conftest.

Training a model

The data README has a bare-bones example for how to load data and incorporate data transforms. This jupyter notebook contains a simple tutorial explaining how to get started working with the data.

Please look at this U-Net demo script for an example of how to train a model using the PyTorch Lightning framework.

Submitting to the Leaderboard

Run your model on the provided test data and create a zip file containing your predictions. fastmri has a save_reconstructions function that saves the data in the correct format.

Upload the zip file to any publicly accessible cloud storage (e.g. Amazon S3, Dropbox etc). Submit a link to the zip file on the challenge website. You will need to create an account before submitting.

License

fastMRI is MIT licensed, as found in the LICENSE file.

Cite

If you use the fastMRI data or code in your project, please cite the arXiv paper:

@inproceedings{zbontar2018fastMRI,
    title={{fastMRI}: An Open Dataset and Benchmarks for Accelerated {MRI}},
    author={Jure Zbontar and Florian Knoll and Anuroop Sriram and Tullie Murrell and Zhengnan Huang and Matthew J. Muckley and Aaron Defazio and Ruben Stern and Patricia Johnson and Mary Bruno and Marc Parente and Krzysztof J. Geras and Joe Katsnelson and Hersh Chandarana and Zizhao Zhang and Michal Drozdzal and Adriana Romero and Michael Rabbat and Pascal Vincent and Nafissa Yakubova and James Pinkerton and Duo Wang and Erich Owens and C. Lawrence Zitnick and Michael P. Recht and Daniel K. Sodickson and Yvonne W. Lui},
    journal = {ArXiv e-prints},
    archivePrefix = "arXiv",
    eprint = {1811.08839},
    year={2018}
}

List of Papers

The following lists titles of papers from the fastMRI project. The corresponding abstracts, as well as links to preprints and code can be found here.

1. Zbontar, J.*, Knoll, F.*, Sriram, A.*, Murrell, T., Huang, Z., Muckley, M. J., ... & Lui, Y. W. (2018). fastMRI: An Open Dataset and Benchmarks for Accelerated MRI. arXiv preprint arXiv:1811.08839.
2. Zhang, Z., Romero, A., Muckley, M. J., Vincent, P., Yang, L., & Drozdzal, M. (2019). Reducing uncertainty in undersampled MRI reconstruction with active acquisition. In CVPR, pages 2049-2058.
3. Defazio, A. (2019). Offset Sampling Improves Deep Learning based Accelerated MRI Reconstructions by Exploiting Symmetry. arXiv preprint, arXiv:1912.01101.
4. Knoll, F.*, Zbontar, J.*, Sriram, A., Muckley, M. J., Bruno, M., Defazio, A., ... & Lui, Y. W. (2020). fastMRI: A Publicly Available Raw k-Space and DICOM Dataset of Knee Images for Accelerated MR Image Reconstruction Using Machine Learning. Radiology: Artificial Intelligence, 2(1), page e190007.
5. Knoll, F.*, Murrell, T.*, Sriram, A.*, Yakubova, N., Zbontar, J., Rabbat, M., ... & Recht, M. P. (2020). Advancing machine learning for MR image reconstruction with an open competition: Overview of the 2019 fastMRI challenge. Magnetic Resonance in Medicine, 84(6), pages 3054-3070.
6. Sriram, A., Zbontar, J., Murrell, T., Zitnick, C. L., Defazio, A., & Sodickson, D. K. (2020). GrappaNet: Combining parallel imaging with deep learning for multi-coil MRI reconstruction. In CVPR, pages 14315-14322.
7. Recht, M. P., Zbontar, J., Sodickson, D. K., Knoll, F., Yakubova, N., Sriram, A., ... & Zitnick, C. L. (2020). Using Deep Learning to Accelerate Knee MRI at 3T: Results of an Interchangeability Study. American Journal of Roentgenology, 215(6), pages 1421-1429.
8. Pineda, L., Basu, S., Romero, A., Calandra, R., & Drozdzal, M. (2020). Active MR k-space Sampling with Reinforcement Learning. In MICCAI, pages 23-33.
9. Sriram, A.*, Zbontar, J.*, Murrell, T., Defazio, A., Zitnick, C. L., Yakubova, N., ... & Johnson, P. (2020). End-to-End Variational Networks for Accelerated MRI Reconstruction. In MICCAI, pages 64-73.
10. Defazio, A., Murrell, T., & Recht, M. P. (2020). MRI Banding Removal via Adversarial Training. In Advances in Neural Information Processing Systems, 33, pages 7660-7670.
11. Muckley, M. J.*, Riemenschneider, B.*, Radmanesh, A., Kim, S., Jeong, G., Ko, J., ... & Knoll, F. (2021). Results of the 2020 fastMRI Challenge for Machine Learning MR Image Reconstruction. IEEE Transactions on Medical Imaging, 40(9), pages 2306-2317.
12. Johnson, P. M., Jeong, G., Hammernik, K., Schlemper, J., Qin, C., Duan, J., ..., & Knoll, F. (2021). Evaluation of the Robustness of Learned MR Image Reconstruction to Systematic Deviations Between Training and Test Data for the Models from the fastMRI Challenge. In MICCAI MLMIR Workshop, pages 25–34,
13. Bakker, T., Muckley, M.J., Romero-Soriano, A., Drozdzal, M. & Pineda, L. (2022). On learning adaptive acquisition policies for undersampled multi-coil MRI reconstruction. Accepted at MIDL, 2022