Deploy into the hospital. Leave nothing behind.
Your model runs on the hospital's own infrastructure and is never decrypted there. Patient data never leaves the hospital and your model never arrives in the clear.
Where the deal stalls today
01The hospital that keeps everything inside
It will not export data and will not accept your cloud, so the model has to be installed on premises.
02The device OEM embedding deal
An OEM wants your model inside its device, and the talks stall on IP terms.
03The integrator after the contract
An integrator with access to the installation could re-host your model once the contract ends.
Where it runs
- On premisesGPU appliance or VM
- PackageDocker container
- Imaging workflowSits beside your PACS and VNA integration
This method is not a regulatory approval and does not change your device classification.
How it fits the technical file
Built to sit inside your release and documentation process.
- One encrypted artefact per release, versioned with the model it protects.
- The accuracy delta of encrypted inference is measured and documented for your model in the pilot.
- A technical-file annex template is part of the pilot deliverables.
Pilot
€20K to €50K
Indicative range, depending on model and target. The fixed fee is confirmed in writing after the architecture review.
- Your model converted and running encrypted on your target hardware
- Benchmark report on your hardware: latency, accuracy delta, memory
- Deployment recipe
Bring one model and one target.
We tell you what it costs in latency and what it buys in protection.