AI and software for hospitals and clinics
We build AI for hospitals and clinics that takes on the paperwork around care: sorting referrals, drafting letters and moving data between the EPD and the systems around it. Clinicians and planners review every output, and patient data stays in the EU under your own controls.
Where the work gets stuck.
Admin eats clinical time
Doctors, nurses and practice assistants spend a large share of their day on letters, forms and reports. Staff shortages make every hour of that work harder to absorb.
Referrals arrive in every format
Referrals come in through ZorgDomein, Zorgmail, scanned PDFs and phone notes. Someone has to read each one, check it is complete and type the details into the EPD before a patient can be planned.
Information sits in separate systems
The EPD, lab system, planning tool and billing software each hold part of the picture. Staff copy data between them by hand, which is slow and introduces errors.
How we improve healthcare providers.
Each one is scoped around your systems and rules, and each one keeps a person in charge of the decisions that matter.
Referral intake and triage
Incoming referrals are read, the key details extracted and a triage suggestion prepared with missing information flagged. A planner or triage nurse confirms each one before it lands in the EPD.
Draft letters and discharge summaries
Drafts of GP letters and discharge summaries are assembled from the notes, results and medication list already in the record. The treating clinician edits and signs every letter before it is sent.
Protocol and guideline search
Staff ask a question and get the relevant passage from local protocols, department agreements and national guidelines, with the source linked. The answer supports their judgement and always points back to the document.
Patient questions and appointment changes
A patient portal handles rescheduling, preparation instructions and practical questions, pulling from the planning system. Anything medical or unclear is passed to the care team with a summary.
Declaration and billing checks
Care activities are checked against DBC and tariff rules before claims go to insurers through Vecozo, and likely rejections are flagged. The finance team decides what to correct and submit.
EPD and FHIR integrations
We connect the EPD to planning, lab, home care and patient apps through FHIR and HL7 interfaces. Data is entered once and the audit trail stays complete.
Built around the rules.
What we design for from the first week. Your legal and compliance people keep the final word.
GDPR (AVG) and health data
Health data is a special category under the GDPR. We minimise what reaches a model, keep processing in the EU or in your own cloud, and document the data flows for your DPIA.
NEN 7510, 7512 and 7513
We design access control, secure exchange and logging of who viewed which record around the NEN 7510 family, so your information security officer can assess the build against your existing policy.
Wegiz and data exchange standards
The Dutch law on electronic data exchange in healthcare (Wegiz) sets standards for exchanges such as medication data, patient summaries and nursing transfers. We use FHIR and the agreed national information standards where they apply.
EU AI Act and medical device rules
Our tools for providers support administration and information retrieval, and leave diagnosis and treatment to clinicians. If a tool starts to influence clinical decisions, it may fall under the MDR and the AI Act high-risk rules, and we flag that early.
Works with what you run.
If a system has an API, a database, an export or an inbox, we can build on it. These are the ones we meet most.
- HiX (ChipSoft)
- Epic
- Nexus
- Medicom and Promedico (GP systems)
- Ons (Nedap)
- ZorgDomein
- Zorgmail
- Vecozo
- FHIR and HL7 v2 interfaces
- Microsoft 365
Where to start.
Referral intake for one department
A narrow build that reads incoming referrals for a single outpatient department, extracts the details and prepares them for the planner. It proves the EPD connection and the review flow on a process with clear rules and measurable waiting-list impact.
Talk it throughWhat it includes
- Intake from ZorgDomein, Zorgmail and PDF
- Extraction and completeness checks
- Planner review screen
- Write-back to the EPD with audit log
Related work.
Guides.
AI readiness checklist: what to answer before your first agent goes live
The questions to answer before your first AI agent handles real work, from process and data to approvals, monitoring and ownership.
10 min read
AI vendor security questionnaire: what to ask before you buy or build
The questions to ask an AI vendor, or your own team, about data, model providers, access, logging, quality, incidents, GDPR and exit, and how to score the answers.
8 min read
Further reading.
Document parsing with LLMs: extracting data from invoices and forms
How to turn invoices and forms into reliable structured data with LLMs, schemas, validation and a review queue for the cases that need a person.
4 min read
Human-in-the-loop AI agents: design patterns for production
The patterns we use to run AI agents safely in production: approval checkpoints, confidence thresholds, tool permissions, audit trails and fallbacks.
4 min read
How to evaluate RAG: retrieval metrics, faithfulness and golden sets
How to measure a RAG system properly: separate retrieval from answers, check faithfulness claim by claim and build a golden set you can trust.
4 min read
Common questions.
It does not have to. We run models through EU-hosted providers or inside your own cloud tenant, and we can redact identifiers before text reaches a model. The data flow is documented so your privacy officer can check it.
Yes, through the interfaces those systems offer, usually FHIR, HL7 or the vendor's own APIs. Access often needs approval from the vendor or your application management team, so we start that conversation in the first weeks.
Your clinicians and staff remain responsible. The software drafts, extracts and suggests, and a named person reviews and approves before anything is sent or stored in the record. Every approval is logged.
We work alongside them. We build the pieces around the EPD, agree interfaces and hosting with your IT department, and hand over documentation so your own team or supplier can manage the result.
A focused build such as referral intake for one department usually runs a few months from discovery to live use. The biggest variable is how quickly EPD access and security review are arranged, which we plan for from day one.