In collaboration with the Fraunhofer Institute and the Berlin Institute for Health Research at Charité, AI pioneer Insiders is developing “FHIR-Starter,” a solution for structuring medical data.
Insiders is participating as a key technology partner in the “FHIR-Starter” research project. The project is led by the Fraunhofer Institute for Experimental Software Engineering (IESE) and is being carried out in collaboration with the Berlin Institute for Health Research at Charité.
Until now, medical information has often been exchanged via traditional doctor’s letters in PDF format. The manual transfer of relevant data into practice management systems (PVS) or hospital information systems (HIS) is time-consuming and prone to errors. The goal of the “FHIR-Starter” project is to develop an innovative software service that uses Artificial
(AI). By using Large Language Models (LLMs) and Natural Language Processing (NLP), unstructured full-text documents are to be analyzed and converted into standardized data structures. This enables seamless integration into existing ITsystems at hospitals and medical practices, thereby representing a decisive step toward digitization and automation in the healthcare sector.
Key requirements for the solution include to guarantee the quality and security of the structured data and, furthermore, to ensure GDPR-compliant processing of sensitive medical data.
“FHIR-Starter” is receiving 1.64 million euros in funding from the Federal Ministry for Economic Affairs and Energy as part of the innovation competition “Generative AI for Small and Medium-Sized Enterprises.”
Insiders Takes on Development of the Large Language Model
As an expert in AI-powered document processing, Insiders is participating in the project, specifically in the development of the Large Language Model (LLM), which, as a core component of the software, automatically extracts relevant information from medical documents—such as doctors’ letters —automatically extracts the relevant information from medical documents. This structured information is then converted into the standardized FHIR format
, which was developed specifically for the exchange of health data and gives the project its name.
“We’re leveraging our many years of expertise in AI-powered data extraction to automate the processing of medical documents,” explains Christian Schön, Product Owner for OvAition at Insiders. “By developing a high-performance LLM, we’re making it possible to efficiently structure medical data and make it usable. Automating administrative tasks reduces the workload on
medical staff and reduces errors. This allows doctors to refocus on their core mission—providing the best possible care for patients.”
In addition to developing the LLM, Insiders is also involved in other key project tasks, including the identification of relevant data, developing a prototype, and presenting scientific findings at professional conferences.
Wide-ranging application possibilities for a digital future
The structured processing of medical data opens up far-reaching possibilities: Hospitals and medical practices can optimize their processes for managing medical reports and patient documentation, medical research institutes gain easier access to relevant information, and patients benefit from more efficient and better care.
As part of the three-year project, which launched in February 2025, Insiders is making an important contribution to the digitization of the healthcare system in Germany.
