Doctor's/discharge-letter generation at large volume
Today
Discharge letters are assembled from many individual documents at the end of the stay.
With local AI
A draft letter is built from approved progress data. Medical review and approval remain.
Local AI for hospitals: discharge letters, coding and hospital knowledge in your own data centre
In hospitals, many teams access patient data under medical confidentiality. We therefore plan AI for your own data centre, with roles per department and documented maintenance paths.
- Own data centre / air-gapped possible
- HIS connection after an interface and security review
- GPU servers instead of an appliance for hospital scale
Customers from mid-sized business, industry and the public sector, in Germany and internationally
The situation
Patient data and AI in the hospital: the starting point
- Discharge letters, tumour-board preparation and coding at large volume tie up clinical time.
- External AI services extend the data and responsibility path. A local deployment keeps processing within the hospital's controlled infrastructure.
- Many departments, many professionals: confidentiality under Section 203 StGB runs through the entire institution, so permissions must apply per role and department.
Use cases
AI in the hospital: use cases and what changes
These examples describe workflows. Data sources, permissions and professional review are defined for the concrete use case; review and approval stay with you.
Tumour-board and case-conference preparation
Today
Findings and progress notes are gathered from several systems for case conferences.
With local AI
The AI prepares a structured case summary from the approved sources.
RAG over internal guidelines, SOPs and hospital knowledge
Today
Guidelines, SOPs and house standards are spread across intranet and drives.
With local AI
Staff query hospital knowledge and receive answers with source references, limited by role.
DRG coding support
Findings and documentation summarisation
Our approach
Advise. Build. Operate. From one team.
Sovereign AI is a lifecycle, not a device purchase: advice, build and operations come from one team.
- 01
Advise & design
Workshop, sizing, data classification and §203 contract framework. We understand your stack, professional software and compliance requirements before we recommend.
- 02
Build & integrate
On-premise build on your hardware, RAG on your documents with access control, integration into your professional software. Plus the DPA and, on request, a written secrecy obligation under Section 203 StGB.
- 03
Operate & maintain
Updates, monitoring, model upgrades and RAG upkeep in ongoing operation. For the AI Cube, AI Cube Care runs with the device for the first twelve months and is cancellable monthly after that; handover to your team and knowledge transfer included.
Legal framework
Local AI for GDPR and Section 203 requirements
We configure and operate the local AI platform for the agreed GDPR and Section 203 scope, as an AI Cube or on your own infrastructure. This includes controlled data paths, limited permissions, documented maintenance, the appropriate DPA and, on request, written secrecy obligations for the people involved. Legal evaluation of the specific use remains with the controller and its advisers.
What §203 StGB prohibits
§203 StGB protects confidential information entrusted by clients, patients and other parties. The necessary involvement of contributing persons is permitted. What matters is that access is limited to what is necessary, the people involved are bound to secrecy and the complete data and maintenance path is controlled.Why cloud AI is the problem
With public AI services, inputs and documents leave your own infrastructure. Whether a service is suitable in a specific case therefore depends not only on a DPA, but also on data flows, subprocessors, access permissions and the actual use.Why local processing helps
On-premises inference, local vector search and internally operated interfaces keep technical processing in your controlled environment. This reduces external data paths and subprocessors and allows access to be limited precisely.Managed service with a secrecy obligation
When we provide setup or operations as a contributing person, the people involved accept a written secrecy obligation and are instructed about the criminal consequences. Where personal data is involved, this is combined with a DPA, limited administrative rights and documented maintenance paths.
What we deliver for it: AVV (data-protection layer) · Written secrecy obligation (secrecy and instruction on criminal consequences) · Maintenance and access concept (necessary rights, logging and controlled paths).
This content is general information and not legal or tax advice. The specific implementation under §203 must be reviewed professionally on a case-by-case basis.
Frequently asked questions about AI and data protection in hospitals
Answers to the most important questions
- For extensive processing of health data, a data protection impact assessment under Art. 35 GDPR should generally be considered; the hospital decides with its data protection officer. We provide the technical description for it: data flows, roles, external connections and maintenance paths.
- Yes, if models, updates, identity, knowledge sources and maintenance processes are designed for offline operation. We plan the required scope as a dedicated architecture.
- For this, we size dedicated GPU servers or clusters based on models, context, user count and availability requirements. The AI Cube is not the default hospital path.
- After reviewing the available interfaces, permissions and actual workflow. A HIS connection is a separate integration project.
- A deployment on dedicated GPU servers starts with sizing and architecture alignment; we state the schedule in the proposal. Professional-software connections and automated RAG pipelines are planned as separate steps.
- We document data flows, roles, technical safeguards and operational responsibility. Where personal data is processed, we conclude the appropriate DPA; on request, the people involved also accept a written secrecy obligation under Section 203 StGB, including instruction on the criminal consequences.
- Yes. We hand over with documentation and train your team. Maintenance, updates, model upgrades and RAG upkeep are agreed as needed or handled by us in managed operations.
Further reading
Related tutorials & guides
Reviews & projects
What clients say about working with us
From a knowledge base with an AI assistant and local AI integration to data sovereignty and ongoing operations.
AI Platform
We were looking for a secure yet easy-to-use alternative to the large American providers and came across WZ-IT. WZ-IT advised us on choosing the right models and implemented the integration for us. Beforehand, we were able to try everything out calmly on a test instance. Today WZ-IT runs Open WebUI for us on our own domain, with a convenient Microsoft login and open-source models hosted by a German provider. Our team can now work productively with AI without resorting to external tools or shadow AI. Working together is straightforward, and questions are answered quickly and clearly. With WZ-IT we also have a partner at our side for further AI projects. (translated)
Carl-Frederic KornHead of E-Commerce & Online Marketing, Krüger Dirndl GmbH
Knowledge Management & AI
The knowledge from many years in the solar industry lived in the head of our former managing director, and it still lives in the heads of our employees today. We wanted to secure it, not as a folder full of notes, but in a way that lets us ask questions in daily work and get reliable answers. WZ-IT advised us from the start and weighed different approaches and knowledge management systems against each other with us before we decided. What was built is a custom knowledge base with an AI assistant that brings together conversations, notes, documentation and manuals: from the platform through onboarding our team to support in operation. Conversations are transcribed and facts are extracted; we review and correct everything ourselves before it is stored, because we do not want to trust an AI blindly. The system runs in our own environment, and the data stays with us. What surprised us was how efficient it was: WZ-IT used existing open-source software and extended it specifically for our requirements, instead of selling an expensive custom build from scratch, as is often usual with IT service providers. Timo and Robin understood our actual problem, preserving knowledge during the generational handover, and implemented it step by step: from the proof of concept through setup and validation with our own data to productive operation, explaining everything clearly along the way. Next, we want to extend this to the knowledge of our employees, so that this does not depend on individual people either.
Joshua von AmelnManaging Director, solarwerkstatt berlin GmbH
Read client reviews
International
Selected projects
- LiveRemote access platform for distributed water treatment plantsABCO Water Systems · Australia · Secure Access & OT
- 81%AWS migration with substantially lower operating costsEVA Real Estate · UAE · Cloud exit & infrastructure
- 98%Provision new sites in minutes instead of hoursnextGYM GmbH · Germany · Automation & IoT
WZ-IT designs, develops and operates infrastructure and software for clients in Germany and internationally. We deliver projects remotely and continue supporting them in ongoing operations after go-live.
Enquiry
Assess local AI for Hospital & Clinic
Describe the workflow and data that should be processed locally or in a controlled environment.
- Straight with Timo and Robin - no sales team, no pitch
- A professional assessment of your situation
- Concrete next steps for infrastructure, software or AI
The conversation is free and non-binding.
WZ-IT's advice on our Azure migration was technically sound and completely non-binding right from the intro call - we took away a great deal.
Jakob ÖschlbergerInno7 GmbH
WZ-IT's advice on our Azure migration was technically sound and completely non-binding right from the intro call - we took away a great deal.