Each colour maps a published value to the exact source passage used to support it. Only bounded excerpts are public; administrators can inspect the complete captured source.
Title
Polish university hospital runs an on-premise Gemma-based AI platform for documentation, medication recommendations and demand forecasting
…se GDPR Article 9 special category data could not leave its own infrastructure. InTechHouse delivered an on-premise platform running Gemma on local GPU servers, with MLOps tooling and a validated reference database that production systems…
Description
A university clinical hospital in Poland with 32 clinical departments needed AI under GDPR Article 9 constraints, so InTechHouse delivered an on-premise platform running Gemma on local GPU servers with MLOps tooling. The LLM is used only in a content-preparation layer; validated outputs are published to a local reference database that the hospital information system queries. Source reports precision@k of 0.80 against a 0.70 contractual requirement for medication recommendations, and 320 documentation schemas across 32 departments.
…Sovereignty Production AI for a Regulated Hospital, With Full Data Sovereignty A university clinical hospital in Poland with 32 clinical departments needed production AI for documentation, medication recommendations, and demand forecasting because GDPR Article 9 special category data could not leave its own infrastructure. InTechHouse delivered an on-premise platform running Gemma on local GPU servers…
Company
Unnamed Polish university hospital (delivered by InTechHouse)
…Sovereignty Production AI for a Regulated Hospital, With Full Data Sovereignty A university clinical hospital in Poland with 32 clinical departments needed production AI for documentation, medication recommendations, and demand…
Country
Poland
…precision@k 0.80 against a 0.70 requirement. Author Tomasz Andrysiak, DSc, PhD Country Poland Industry Life Sciences & Pharma Solution Industrial Data Platforms & OT/IT Inte…
Industry
Healthcare
…y Life Sciences & Pharma Solution Industrial Data Platforms & OT/IT Integration University hospital with 32 departments under GDPR Article 9 A university clinical hospital in Poland, one of the largest in its region, wit…
Problem
A university clinical hospital in Poland with 32 clinical departments needed production AI for documentation, medication recommendations and demand forecasting, but GDPR Article 9 special category data could not leave its own infrastructure.
…Sovereignty Production AI for a Regulated Hospital, With Full Data Sovereignty A university clinical hospital in Poland with 32 clinical departments needed production AI for documentation, medication recommendations, and demand forecasting because GDPR Article 9 special category data could not leave its own infrastructure. InTechHouse delivered an on-premise platform running Gemma on local GPU servers…
Solution
InTechHouse delivered an on-premise platform running Gemma on local GPU servers, with MLOps tooling and a validated reference database that production systems query instead of the LLM. The LLM is used only in the content-preparation layer.
…ustrial environments. The architecture principle that made it production-grade: the LLM is used exclusively in the content-preparation layer. All generated outputs are validated, versioned, and published to a local reference database, and it is this database, not the model, that production systems query. Live op…
Business value
At handover, medication recommendations reached precision@k of 0.80 against a contractual requirement of 0.70, and 320 documentation schemas were generated across all 32 departments. Forecasting achieved hospital-wide WAPE in the 15 to 17 percent range. Results are reported by the delivery firm.
…pipeline, and version control on every generation cycle. Recommendation module: precision@k of 0.80 achieved against a contractual requirement of 0.70 (0.85 on the validation sample, 0.94 micro-averaged), covering 100 medications…
AI capabilities
Generative AI, Predictive Analytics
…expert-reviewed test sets. 3. Demand forecasting for medications and supplies. Forecasting models run once a day and deliver seasonality-aware order proposals to pharmacy and logistics before a fixed morning cut-off, replacing manual esti…
Business functions
Supply Chain & Logistics
…Nurses gain material recommendations consistent with local department practice. Pharmacy and logistics gain daily, seasonality-aware order proposals in place of manual estimates. Because the production path is decoupled from the generative layer, the system…
Technology
Gemma
…were selected against EU AI Act alignment and provenance from trusted vendors. The platform runs on Gemma, Google's open model family. Measured results at handover: Documentation m…
Deployment options
On-premise
…grade AI, not a pilot, inside its own walls. The requirements were unambiguous: all data processing and model training had to run entirely on-premise, in the client's server room clinical and operational staff needed AI supp…
Applicable regulations
General Data Protection Regulation (GDPR), EU AI Act
…l-wide WAPE (Weighted Absolute Percentage Error) in the 15 to 17 percent range. The deployment addresses regulatory requirements at two levels. Under GDPR, all personal data processing occurs within the hospital's own…
Headline outcome
…validated reference database that production systems query instead of the LLM. Medication recommendations reached precision@k 0.80 against a 0.70 requirement. Author Tomasz Andrysiak, DSc, PhD Country Poland Industry Life Sciences & Pharm…
Use case type
Decision support
…dedicated intermediate database HIS integration via API and database interfaces Daily decision support for physicians, nurses, and pharmacy The platform turns years of dormant historical data into a structured foundatio…
