Healthcare
The discharge summary gets written at nine at night, and episode coding runs months behind while the hospital's funding depends on it. Inferana puts open models to work on that text without it leaving the hospital.
What each regulation in your sector requires, and which part of that the deployment solves.
Health data is a special category: explicit legal basis and reinforced safeguards to process it.
Prompts and responses are never stored or used to train models, and every access is logged per user and team.
Primary and secondary use of electronic health data, with processing kept inside the European Union.
Inference on European infrastructure or in your own data centre, with clinical text staying inside the hospital's perimeter.
Traceability, data governance and human oversight for high-risk systems in healthcare settings.
Open models with published model cards and licences, access controls and an auditable record of every query.
Hospitals are essential entities: incident reporting and supply chain security obligations.
A single European provider with deployment documentation for your supplier inventory, and an on-premise option on your own infrastructure.
The same ones you already use, specific to your sector.
Ten minutes per patient, a waiting list that keeps growing and reports written after hours. The approved tool has been sitting with the committee for months, so the same thing keeps happening:
My consultants write their discharge summaries at nine at night.
Clinical service leadCoding is five months behind and the funding follows the coding.
Clinical coding officeTen minutes per patient and half of that goes into typing.
Outpatient clinic coordinationClinical notes end up in a public chatbot and I cannot see it.
Data protection officerThe HIS is twenty years old and nobody wants to open it.
Health information systemsTo start a study I need thousands of reports anonymised.
Clinical researchI cannot sign off a cost that grows with every new user.
Operations managementInferana does not carry a CE mark as a medical device under the MDR (Regulation (EU) 2017/745). It is built for documentation, administrative and knowledge work: drafting, summarising, classifying and searching. If your organisation puts it to work supporting clinical decisions, that medical purpose is set by you, and you are the one entering the MDR perimeter with the obligations that come with it.
Wherever you decide: on European infrastructure we manage, or in the hospital's own data centre, on-premise. With an on-premise deployment clinical text never leaves your infrastructure. In both cases prompts and responses are never stored or used to train models.
The model reads the episode and proposes the codes with the sentence in the report that justifies each one. The coder confirms or corrects, so what is typing from scratch today becomes reviewing a proposal with the evidence next to it. The dataset you submit stays your responsibility, and the record of every query is there to audit the process.
Processing stays inside the European Union, and on-premise it stays on your own machines. The access application to the health data access body, the legal basis and the purpose of the secondary use remain yours. We provide the record of who queried what, and anonymisation before the analysis starts.
Access control per user and per team, with a record of every query. The DPO and internal audit can review who asked what and when, and export it for an investigation or a breach notification.
Yes, with the same key. There the query does go out to the model provider, so clinical text and any patient data are handled with the open models inside your perimeter. Access is configured per team.
No. The API is OpenAI-compatible: swap the key and the endpoint and what you already run stays as it is. Anything new connects over the API, with no need to open up the record system.
The models are open and the API is OpenAI-compatible, so your exit plan does not depend on us: you can take the weights with you and point the code elsewhere by changing a key. Your data and your indexes are exportable.
In the demo we go through your case: which models you need, where they run and what it takes to meet the regulation that applies to you.
We reply within one business day.
