The advancing digitalisation in the medical field opens up exciting new perspectives to support doctors in effectively treating their patients with the help of computer-based systems. This important undertaking is even considered a central mission by the Federal Government as part of the High-Tech Strategy 2025 under the title “Data helps heal” (Federal Government, 2020).
The starting point for this digital transformation was the introduction of the electronic patient record (ePA) in 2021, which enabled convenient digital availability of large amounts of medical data. Previously, this information was often only available in analogue form in doctors’ offices and hospitals. Although digital access to this data now exists, it is still mostly in the form of unstructured texts. The analysis of this data still requires considerable time from medical professionals. A visual analysis system that clearly displays and compresses the patient history is missing so far.
The aim of this research project is to extract, structure and visualise information about therapy, diagnosis, medication and symptoms from the ePA using machine learning techniques. The international standard for data exchange in the medical field, “Fast Healthcare Interoperability Resources” (FHIR), will be used to ensure smooth cooperation with existing systems in the healthcare sector. This is intended to facilitate and accelerate access to the information in the primarily text-based documents of the ePA.