FHIR R4 bundles, trial cohorts, EHR fixtures. Zero real PHI.
Deterministic HL7 FHIR R4 patient bundles with 100% referential integrity by construction. LOINC, RxNorm, and the full US ICD-10-CM catalogue shipped under free licences. Safe Harbor de-identification is a property of the generator input, not a post-hoc scrub — built for health-tech, providers, trial operators, and medical-device validators.
| clinicalStatus | icd10_code | icd10_display | id |
|---|---|---|---|
| active | I50.9 | Heart failure, unspecified | CP000000-000 |
| active | N39.0 | Urinary tract infection, site not specified | CP000000-001 |
| active | D64.9 | Anemia, unspecified | CP000001-000 |
| active | H52.4 | Presbyopia | CP000001-001 |
| active | E55.9 | Vitamin D deficiency, unspecified | CP000002-000 |
| active | Z00.00 | General adult medical exam without abnormal findings | CP000002-001 |
| active | G47.00 | Insomnia, unspecified | CP000005-000 |
| active | F41.1 | Generalized anxiety disorder | CP000005-001 |
| active | N39.0 | Urinary tract infection, site not specified | CP000005-002 |
| active | R07.9 | Chest pain, unspecified | CP000005-003 |
| active | E78.5 | Hyperlipidemia, unspecified | CP000005-004 |
Showing all 20 preview rows
Four workflows clinical + research teams buy first.
Clinical-trial digital twin
Simulate a trial's primary-endpoint population before enrolment to right-size the protocol. Deterministic seeds mean the same statistical cohort replays byte-identical in the FDA submission package as in the internal simulation.
EHR integration + conformance fixtures
Drive millions of FHIR R4 bundles against your EHR's ingest, search, and query paths. Every bundle is R4-validator-clean by construction; failures are unambiguously server-side, so regression tests diff cleanly.
ML training sets for care-delivery models
Train readmission-risk, sepsis-onset, and medication-adherence models on synthetic patient timelines generated without source patient records. Collaborators at other institutions re-run the same contract, seed and engine version and get byte-identical data. What review your institution requires remains its call.
Drug-safety signal testing
Stress-test pharmacovigilance pipelines on millions of synthetic medication-administration + adverse-event records with tunable co-occurrence rates. The sealed contract + seed lets you share the test cohort with a regulator who can reproduce it exactly.
Privacy office, ethics board, regulator.
No source patient records
The from-scratch FHIR path accepts no dataset — there is no input for patient data to arrive through. Names, medical record numbers, birth dates and addresses are minted by the generator, and MRNs are issued on a RadMah-owned identifier system that cannot collide with a real assigning authority.
Evidence for your privacy review
Every run ships an evidence bundle recording the contract, seed, validator results and artifact hashes — the input to your own classification of the output. A DPA covers the control-plane metadata RadMah processes as your processor. How the output is classified, and what agreements that implies, is your determination with your counsel.
Evidence a validated system can consume
Each run emits a determinism report, an artifact index with per-artifact hashes, and a seal binding them together — an integrity record your validated system can reference and re-check offline. Deterministic regeneration gives a stable anchor for a test plan; qualifying RadMah inside your validated environment remains your process.
Two different paths, two different analyses
From-scratch FHIR generation reads no source records. Learned synthesis from your own healthcare data does read that data, and measures membership-inference, linkage and attribute-inference indicators against a holdout on every run. They are separate capabilities with separate privacy positions — we will not blur them.
Three products. One sealed contract.
Healthcare FHIR Bundles
Eight R4 resource types. 100% referential integrity by construction. LOINC, RxNorm, and the full US ICD-10-CM catalogue shipped under free licences. SNOMED stays BYO-licence.
ExploreTabular synthesis
Measured against each customer source with per-run fidelity and privacy reports. Useful for research cohort files, claim-line data, pharmacy dispense records, and adverse-event time series.
ExploreAgentic Data Scientist
A typed tool-surface planner + executor that drives end-to-end cohort-building from a natural-language protocol description, with an explicit approval gate before any credit-spending step.
ExploreShip your EHR fixture today. No IRB, no BAA.
Free tier: 25 runs a month. Enterprise: a Hybrid SDK deployment that keeps any real reference data inside your enclave, plus a dedicated healthcare solutions engineer and a 21 CFR Part 11 validation pack under NDA.