No PHI submitted
Patient records and confidential datasets are never collected through this website.
Healthcare facilities trust us to license de-identified data to frontier AI labs. Security around that data is our biggest priority: legal counsel, PhD-led de-identification, independent audits, and expert determination — each step before you approve a deal.
How it works
De-identification and re-identification risk are not afterthoughts. They are the spine of every engagement. Each layer must clear before the next begins.
Nothing is collected, shared, or licensed without your written approval. You set the terms, the buyers, and what is included.
Our senior research team applies Shfa's proprietary technology to remove all 18 HIPAA identifiers.
Shfa AI research team
John Snow Labs works with us to complete de-identification and verify that protected health information has been removed.
John Snow Labs
Independent expert determination vendors review the final dataset and assess re-identification risk before it can move forward.
Third-party vendors
Pillsbury Law — a top-tier national firm — advises on structure, compliance, and the contractual framework around every licensing engagement.
Pillsbury Law
Only after every step above — and only within the scope you approved — does de-identified data reach a frontier AI lab under contract.
See it in action
Patient names, ages, facilities, dates, phone numbers, MRNs, and providers — identified and replaced while the clinical meaning stays intact.
Original text
De-identified text
Elena Vasquez is a 58-year-old patient admitted to Riverside Community Hospital on 08/22/2025 for chest pain. Contact: (602) 555-0142 MRN: RG-441982. Attending: Dr. Thomas Reed
Patricia Huang is a 61-year-old patient admitted to Valley Medical Center on 06/10/2025 for chest pain. Contact: (602) 555-8831 MRN: RG-773204. Attending: Dr. Naomi Keller
01 — Legal counsel
Pillsbury Law — consistently ranked among the largest and most respected firms in the country — advises Shfa on the legal structure of healthcare data licensing. That includes healthcare data privacy compliance, the contractual framework with buyers, and the terms you approve before anything moves forward.

02 — Research team
Our proprietary de-identification technology is developed and operated by a senior team of AI researchers with doctoral training in healthcare data, clinical NLP, and privacy-preserving machine learning.

03 — De-identification partner
We work with John Snow Labs to complete de-identification and verify the output. Their platform is the most widely deployed de-identification solution in large academic medical centers — regulatory-grade accuracy across clinical notes, structured data, PDFs, DICOM, and FHIR.
Independently validated at two billion patient notes for Providence Health — 99%+ accuracy, zero successful re-identifications in adversarial testing, and peer-reviewed results that outperform every major cloud API. State of the art in healthcare de-identification.

04 — Expert determination
Expert determination is the last gate: independent vendors, qualified statisticians and privacy experts, assess whether the final dataset presents a very small risk of re-identification, consistent with HIPAA's de-identification standard.
This step exists because de-identification alone is not always enough. Re-identification risk depends on what remains in the dataset, what external data exists, and how a buyer might combine sources. Expert determination answers that question with an independent opinion, not ours.
What we never do
Hard limits on how we operate, not fine print buried in a policy page.
Patient records and confidential datasets are never collected through this website.
AES-256 at rest and TLS in transit, from your systems to ours.
You can change your terms or withdraw access if your priorities shift.
Questions
The appraisal is free, there is no commitment, and you approve every deal. One call tells you whether licensing makes sense for your organization.
Book your free appraisalOne call · No obligation · You approve every deal