MosaicMRI title

Last update: Aug 31, 2026

A large-scale and diverse dataset of raw musculoskeletal MRI

3,096 volumes
88,416 slices
854 patients
10 anatomies
Multi-contrast + multi-coil
Request Access Paper (PDF) Code + Baselines Benchmark

About the Dataset

MosaicMRI is the largest and most diverse open-source raw musculoskeletal MRI dataset to date.
The 1.5T public release spans 10 anatomy labels, 3,096 volumes, and 88,416 stored frames/slices.

Volume Distributions

Coarse anatomy distribution across MosaicMRI

A diverse multi-anatomy dataset

MosaicMRI brings realistic clinical variability across anatomy, contrast, orientation, and coil configuration far beyond the existing brain- and knee-focused datasets.

Raw data for reconstruction research

MosaicMRI facilitates research on accelerated MRI, low-field reconstruction, motion suppression, and other real-world reconstruction challenges.

A new benchmark for stress-testing AI

Explore Benchmark

This initial release goes beyond typical accelerated reconstruction to probe anatomical and contrast generalization under real-world variability.

Enabling foundation model research in MRI

MosaicMRI provides a testbed for studying key foundation model challenges, including scaling laws, data synergies, continual learning, data mixtures, reliability, and out-of-distribution generalization.

Example Visuals

RSS reconstruction examples from the updated 1.5T release, grouped by anatomy and orientation.

Coronal Shoulder coronal RSS sample
Sagittal Shoulder sagittal RSS sample
Axial Shoulder axial RSS sample

Dataset Details

MosaicMRI is designed for learning-based MRI under realistic clinical variability in anatomy, contrast, orientation, and coil configuration.

Volumes
3,096
Patients
854
Slices
88,416
Public H5 Size
2.89 TiB

Constructing the Dataset

The release was built from Siemens raw musculoskeletal MRI acquisitions.

Retained acquisitions were converted to ISMRMRD and reviewed for consistent geometry, standardized anatomy/contrast labels, and fully sampled Cartesian phase-encoding support.

Protocols and Labels

  • •Contrast labels are standardized to PD, PD_FS, STIR, T1, T1_FS, T2, and T2_FS.
  • •Each volume includes orientation, legacy anatomy, finer anatomy, and coarse anatomy attributes.

Dataset Partitioning

Splits are patient-disjoint to avoid leakage, with target ratios 70% train, 15% validation, and 15% test. The selected proposal preserves exact patient targets and balances files, stored frames/slices, anatomy labels, and contrast labels.

Split Scans Patients Slices Size
train 2,166 598 61,903 2,074.79 GiB
val 465 128 13,265 443.11 GiB
test 465 128 13,248 438.52 GiB

Data Format and Quickstart

File organization and baseline usage for reconstruction experiments.

File structure

Directory layout (current release statistics):

MosaicMRI_1p5T/
  multicoil_train/                                (2,166 files, 2,074.79 GiB)
    *.h5
  multicoil_val/                                  (465 files, 443.11 GiB)
    *.h5
  multicoil_test/                                 (465 files, 438.52 GiB)
    *.h5
  • Core splits: multicoil_train, multicoil_val, and multicoil_test are the public reconstruction splits.
  • Split policy: files are grouped by a private patient-level key before assignment; public H5 files do not expose those identifiers.
  • Release size: the public H5 release is 2,956.42 GiB across 3,096 files.
See Benchmark Challenges

Inside each H5

  • kspace: complex64 raw k-space stored as [slice, coil, readout, phase].
  • reconstruction_rss: float32 root-sum-of-squares target stored as [slice, recon_readout, recon_phase].
  • ismrmrd_header: sanitized ISMRMRD XML string with public geometry and acquisition metadata.
  • Labels: acquisition, contrast, anatomy, finer_anatomy, coarse_anatomy, and orientation.
  • Geometry: encoding_size, recon_size, num_slices, num_coils, padding_left, padding_right, original/released matrix attributes, and removed phase-column counts.
  • Acquisition metadata: scanner vendor/model/field strength, receiver channels, trajectory, sequence type, TR/TE/TI, flip angle, and positioning metadata.

Quickstart

GitHub repo

Minimal steps to download a file, apply a mask, and run a baseline reconstruction.

1) Install
git clone https://github.com/AIF4S/mosaicmri
cd mosaicmri
conda env create -f varnet/environment.yml
conda activate mosaic_mri_varnet
2) Run demo reconstruction
python varnet/run_pretrained_varnet_inference.py   --state_dict_file /path/to/weights.ckpt   --data_path /path/to/MosaicMRI_1p5T/multicoil_test   --output_path /path/to/recons   --accelerations 8   --center_fractions 0.04

Request Access to MosaicMRI

To obtain access, submit the request form below. Approved users receive access through the gated Hugging Face dataset.

Hugging Face logo

Hugging Face account required

Create your Hugging Face account before requesting access.

Create Hugging Face account

Please use an institutional email address when possible. Personal emails may take longer to verify.

Create your Hugging Face account first, then enter that account username here.

Data Sharing Agreement for Research and Educational Use

Open formal agreement page

By downloading or accessing the dataset entitled “Testing the performance and reliability of AI-based image reconstruction” (the “Dataset” or “Data”), you (“Researcher”) acknowledge that the Data are proprietary to and owned by the University of Southern California (“USC” or “Licensor”).

  • USC grants the Researcher a non-exclusive, royalty-free license to access and use the Data solely for internal, non-commercial research or educational purposes. No other rights are conveyed.
  • The Researcher will receive a download link to access the Data without charge and will not share the download link with any other individual. Each user must register separately and agree to these Terms.
  • The Data may not be sold, licensed, monetized, or otherwise commercially exploited. Those seeking commercial use must contact the USC Stevens Center for Innovation, 3720 S Flower Street, Floor 3, Los Angeles, CA 90089, at mta@stevens.usc.edu.
  • The Researcher shall not distribute, publish, reproduce, retransmit, copy, or transfer any portion of the Data, or variables derived from it, to anyone outside their direct supervision, except as necessary for academic publications or presentations that properly cite the Dataset.
  • The Researcher shall ensure that any colleagues or students within the same institution who access the Data first agree to be bound by these Terms. Anyone under the Researcher’s supervision must follow the same restrictions and obligations.
  • If the Researcher is employed by a for-profit or commercial entity, the employer is also bound by these Terms, and the Researcher represents that they are authorized to enter into this Agreement on behalf of the employer.

Compliance, Use Restrictions, and Safeguards

  • The Data will not include personally identifiable information. If the Data are coded, USC will not release, and the Researcher will not request, the key to the code.
  • The Researcher agrees not to attempt to re-identify individuals, link the Data with other datasets for identification purposes, or contact any individual subjects.
  • The Researcher will comply with all applicable laws, regulations, and institutional policies, including obtaining any required ethics approvals.
  • The Researcher will maintain the security and confidentiality of the Data and use appropriate safeguards to prevent unauthorized access, use, or disclosure.
  • If the Researcher suspects that any portion of the Data contains protected health information (PHI), they will not use it, will immediately destroy it, and will promptly notify USC.

Acknowledgment and Attribution

The Researcher agrees to acknowledge USC as the source of the Data in all written, visual, or oral public disclosures, using citation language consistent with scholarly standards.

“Data used in the preparation of this article were obtained from the University of Southern California MosaicMRI database. USC investigators provided data but did not participate in analysis or writing of this report.”

Please cite: Arguello, P., Tinaz, B., Mohammad, S. S., Soltanolkotabi, Maryam, and Soltanolkotabi, Mahdi. "MosaicMRI: A Diverse Dataset and Benchmark for Raw Musculoskeletal MRI." arXiv (2026). https://arxiv.org/abs/2604.11762.

Warranty Disclaimer and Limitation of Liability

  • THE DATA ARE PROVIDED “AS IS.” USC has no obligation to provide maintenance, support, updates, enhancements, or modifications.
  • USC MAKES NO REPRESENTATIONS OR WARRANTIES, express or implied, including but not limited to warranties of merchantability, fitness for a particular purpose, or non-infringement.
  • The Researcher accepts full responsibility for their use of the Data and agrees to defend and indemnify USC and its employees, trustees, officers, and agents against all claims arising from such use, including claims related to copyrighted images created from the Data.
  • USC shall not be liable for any loss, claim, or demand made by or against the Researcher arising from use of the Data.
  • IN NO EVENT SHALL USC BE LIABLE for incidental, consequential, exemplary, indirect, or economic damages of any kind, including lost profits, lost business, or lost goodwill, regardless of legal theory or prior notice of potential damages.

Termination and Post-Termination Obligations

  • USC reserves the right to terminate the Researcher’s access to the Data at any time.
  • Upon termination or completion of the Researcher’s work, all copies of the Data must be destroyed.

Governing Law

These Terms and any disputes arising from the use of the Data shall be governed by the laws of the State of California.

By submitting, you agree to comply with the dataset license and access policy.

MosaicMRI Benchmark

A three-track, 8x-acceleration benchmark: Mixed Anatomy Reconstruction, Anatomy Generalization (held-out ankle), and Contrast Generalization (held-out T1-FS). Participants upload reconstructed H5 files and are evaluated against hidden ground truth with PSNR, SSIM, and NMSE leaderboards.

Go to Benchmark

License, Access Policy, and Ethics

Access is granted for research use after manual review.

License

MosaicMRI is released for non-commercial research and method development under the posted license terms.

Scope
Research-only / non-commercial use

De-identification

Metadata is de-identified before release.

  • •PHI removed from headers/metadata
  • •Research-only use

Citation

Please cite the dataset paper if you use MosaicMRI.

BibTeX

@article{mosaicmri_2026,
  title   = {MosaicMRI: A Diverse Dataset and Benchmark for Raw Musculoskeletal MRI},
  author  = {Arguello, Paula and Tinaz, Berk and Mohammad, Shahab Sepehri and Soltanolkotabi, Maryam and Soltanolkotabi, Mahdi},
  journal = {arXiv},
  year    = {2026},
  doi     = {10.48550/arXiv.2604.11762}
}

Citation metadata will be updated if publication details change.

Paper (PDF)

Collaborators

AIF4S Research Group
DISC Research Group
University of Utah Logo

University of Utah

USC Logo

University of Southern California (USC)

UC Irvine Logo

University of California, Irvine (UCI)