These terms govern your use of the OpenMSK web service at openmsk.com. By uploading data, you accept them. OpenMSK is an independent academic project, not a university or hospital service.
OpenMSK is also free and open-source. You may install it from github.com/gattia/OpenMSK and run the same analyses on your own computer, in which case these terms do not apply and no data leaves your machine.
OpenMSK is for non-commercial research and education. It is not a medical device, and has not been reviewed or cleared by the FDA or any other regulator. You must not use it for clinical diagnosis, treatment planning, or any decision about a patient. Results are generated automatically, are not reviewed by a clinician, and may be wrong.
Segmentation and shape fitting use neural networks, but produce measurements only. BScore and all other outputs are research measures that have not been clinically validated.
By uploading data, you confirm that:
If you upload data in error, contact us and we will delete it. You may not resell access, submit volumes that degrade the service for other users, or attempt to access another user's jobs. We may decline any submission.
Uploads are processed on our server, hosted on commercial cloud infrastructure in the United States; if you are outside the US, your data is transferred here. Uploads and downloads use HTTPS, and the disks are encrypted at rest.
Once your image is read it is converted to NRRD, a format that holds voxel data and geometry only and cannot carry DICOM metadata. Conversion therefore strips the DICOM headers, and the image is renamed, so your original filename is not kept either. Where an analysis requires more than one echo (qDESS T2 mapping uses two), each echo is stored separately so that results remain reproducible.
From the headers we keep a fixed list of scan acquisition parameters and discard everything else:
Kept: echo times, repetition time, inversion time, flip angle,
echo train length, number of averages, pixel bandwidth, sequence name and variant,
field strength, scanner manufacturer and model, receive coil, slice thickness,
voxel spacing, image dimensions, patient position and orientation, body part,
laterality, sex, age in years (90 and over recorded as 90), and height and weight
where the headers carry them. Also two vendor private fields,
(0019,10B6) and (0019,10B7), which encode gradient
spoiling and are required to compute T2 from qDESS; they describe the pulse
sequence, not the person.
Discarded: everything else — patient name, patient and accession numbers, all dates, institution, referring physician, device serial numbers, and all other private tags. Study and series identifiers are kept only as a one-way hash.
Subject information you provide. The upload form optionally requests the subject's sex, age, height and weight. All four fields are optional, and leaving them blank does not affect your results. What you enter is kept alongside the acquisition parameters above, on the same terms as everything else here, and may be published in de-identified form (§5). It is stored separately from any value read from the headers, and neither overwrites the other. Age is recorded in whole years and capped at 90; a recorded age of 90 means 90 or older. Sex, age under 90, height and weight are not among the eighteen HIPAA Safe Harbor identifiers, so recording them remains within the de-identification standard required by §3. Do not enter any other information about the subject, in these fields or in a filename.
The file you uploaded, and anything extracted from it, is deleted within 24 hours.
Your results are provided as a zip at a download link, which we also email to you. The link is not password-protected, so anyone who has it can download your results; do not forward it. The zip is deleted 24 hours after your job finishes.
We keep the converted image, the derived results, the acquisition parameters above, and any subject information you entered, and use them to improve OpenMSK and for musculoskeletal imaging research. Retention is a condition of using the free hosted service; to keep your data entirely in-house, run OpenMSK on your own computer instead (§1).
You agree that what we keep may be included in our training and evaluation data, shared with other researchers, and published in de-identified form; we periodically release OpenMSK training and benchmark data so that results can be reproduced. Published material contains only converted images, derived results, acquisition parameters, and subject information you entered — never original files, filenames, or email addresses. Published data cannot be recalled. We do not sell any of it.
We also store a SHA-256 checksum of the image data, so that we can identify the same scan being uploaded twice and avoid double-counting it. A checksum is one-way and cannot reproduce or reveal the image.
We use your email address to send your results, and occasionally to contact you about OpenMSK — significant changes, related papers, or collaboration. We keep it indefinitely, do not sell it, and will stop emailing you or delete it on request.
To have your data deleted, email info@openmsk.com with your job ID (shown on the results page and in your notification email) or the email address you used. This includes the derived results. We will respond within 90 days. We may retain the duplicate-detection checksum, which identifies no one, and we cannot recall anything already published.
OpenMSK builds on components licensed for non-commercial academic use, and the neural shape modeling method it uses is the subject of a pending patent application filed by Stanford University. Nothing in these terms grants you any license — patent, copyright, or otherwise — beyond use of this service for non-commercial research. If you wish to use OpenMSK or its methods commercially, contact us first.
The example scan and results published on the home page are a maintainer's own, contributed for that purpose. You may use them for research and teaching, including in publications, provided you cite the paper below. §2 applies to them as it does to everything else here.
Results we produce for you are yours to use in your research and to publish. If OpenMSK contributes to your work, please cite it:
Goyal A, Belibi F, Sahani V, Pedersen R, Vainberg Y, Williams A, Chu C, Haddock B, Gold G, Chaudhari A, Kogan F, Gatti A. Automating Imaging Biomarker Analysis for Knee Osteoarthritis Using an Open-Source MRI-Based Deep Learning Pipeline. medRxiv 2025. doi:10.1101/2025.02.21.25322094
This section is guidance for investigators. It does not form part of these terms and is not legal advice.
If your board requires something this does not cover, contact us.
The service runs on a single server. It may be slow or unavailable without notice, jobs may be lost, and we may modify or discontinue it at any time. Do not rely on it as the only copy of any data, and download your results before they expire.
THE SERVICE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, ACCURACY, AND NON-INFRINGEMENT. TO THE FULLEST EXTENT PERMITTED BY LAW, WE ARE NOT LIABLE FOR ANY DAMAGES OR LOSSES ARISING FROM YOUR USE OF THE SERVICE, INCLUDING LOST DATA OR LOST RESEARCH TIME.
If data you upload in breach of §3 gives rise to a claim against us, you agree to indemnify us for it.
We may update these terms; the version and date above will change, and the version current at the time governs uploads made while it is in effect. A copy of the version that applied to your job is included in your results zip, so you retain the terms you accepted. We will not retroactively change how we use data already uploaded under an earlier version. Superseded versions are available on request.
These terms are governed by the law of the State of California, and any dispute will be heard in the courts of California. Questions: info@openmsk.com.