Produces structured clinical letters from treatment data.
Developed in Germany · Hosted in the EU or on-premises
Delivered together with
Sovereign hosting in Germany, single-tenant from the first user.
Legal, tax and compliance expertise for legally sound deployment.
Advisory and implementation for the public sector, starting with the AI readiness assessment.
Diagnoses sit in the hospital information system, lab values in the results system, the clinical course in the chart, the medication in the prescribing module. Whoever writes the letter opens each one in turn and copies it into the in-house template. Because every author works in a different order, two letters from the same ward look different.
Before sign-off, all the doctor has left to do is read and correct the three text blocks; writing effort falls by 60%.
The 60% reduction in writing effort is this agent's headline figure. The number of sources, writing time and blocks describe a typical discharge letter on a ward; they are illustrative, not counted.
The agent reads the diagnoses, findings, clinical course and medication for the stay from the hospital information system. Every detail it carries over keeps a record of its source.
Case ST3-2026-0812, 5 sources read, 31 details carried overThe agent fills in history, course, findings and recommendation in the order of the in-house template. It uses medical terms exactly as documented.
Draft with 6 sections, every line with a source referenceThe agent files the draft, with references and open passages, in the patient record. The treating doctor reads, amends and signs.
arztbrief_ST3-2026-0812_entwurf.docx, 2 pages, 3 open passagesIt carries over the coded diagnoses, each with its reference in the hospital information system. The treating doctor is responsible for the medical content.
It files the draft in the patient record. The medical records team sends it to the GP or the patient after sign-off.
If a day's clinical course is left blank, the draft shows an open passage with the date. The doctor fills it in.
Connected systems
Operations
A pilot typically runs on your own data within 4 to 6 weeks. The agent connects to your existing systems rather than running as a separate, parallel environment.
EU cloud, private cloud or on-premises. Processing is GDPR-compliant and models are never trained on your data. Details in the Trust Center.
It is based on experience from comparable projects and process benchmarks, not a commitment. We work out the realistic effect for your organisation during the assessment before the pilot.