Improves first-pass coding, the one lever that scales before review rates tighten.
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.
Coding happens at the end of the case, under time pressure, from whatever the record offers. When the review notice arrives later, the same record has to be opened again, the coding reconstructed and the justification written from scratch. By then, the colleague handling it may well not be the one who coded the case.
If the passage in the record that supports the secondary diagnosis is identified at first coding, half of this work disappears: that is the 50% less review effort.
The 50% is this agent's headline figure. The four-month gap and two rounds of work per review case reflect a typical case in medical controlling; they are not measurements.
The agent reads the coded diagnoses and procedures and looks for the supporting passage for each one in the case documentation. It flags every code that lacks evidence.
Case 2026-100942, 14 codes checked, 2 without a supporting passageFor each contested code, the agent sets out where a review would typically focus. It names the code, the reference and the share of case revenue at stake.
Risk list with 2 codes, secondary diagnosis E87.6 with no lab value in the recordThe agent compiles the justification with references from the record. Whether the objection is lodged in that form is up to medical controlling.
widerspruch_2026-100942_entwurf.docx, 5 referencesIt shows codes without evidence and names the reference for each. The clinical coder does the coding.
It presents the draft objection with its justification. Whether and how to object is medical controlling's decision.
It points to the places where the documentation is thin. How the Medical Service decides remains open.
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 establish the realistic effect for your organisation in the assessment before the pilot.