Wir überwinden Grenzen

The aim of the three-year collaboration between the University Medical Center Freiburg and Dedalus HealthCare is to integrate AI applications even more deeply into radiological processes and make data more usable for research

Radiology already has one of the highest levels of digitalization in medicine and relies heavily on artificial intelligence (AI). However, to simplify the integration of AI evaluations into routine procedures and to improve the evaluation of existing data sets by AI, findings must be collected in a clearly structured manner. In particular, the automatic transfer of AI results into existing findings templates and their interactive display in a radiological image system are seen as key innovations. This will allow the growing volume of radiological examination data to be managed more effectively and increase its added value for doctors and patients. To achieve this goal, the Department of Diagnostic and Interventional Radiology at the Freiburg University Medical Center and Dedalus HealthCare began a close collaboration on 22 January 2025, which will initially run for three years.

Structured reporting: Basis for data-driven medicine

Structured reporting makes it possible to record radiological data in a standardized way and use it more effectively for clinical and scientific purposes. Dedalus has developed a module for this purpose, the functionalities of which are to be evaluated and further developed in order to support data-driven medicine and increase the quality of radiological findings.

"Structured reporting is a key component in not only managing the rapidly increasing volume of examination data, but also using it specifically for better diagnostics and research," explains Prof. Dr. Elmar Kotter, Managing Senior Physician at the Department of Diagnostic and Interventional Radiology at the Freiburg University Medical Center.

AI integration for more efficient workflows

Another focus of the project is the integration of AI into routine radiology workflows. Various AI systems are already being used in radiology at the University Medical Center Freiburg. However, these often deliver results that have to be processed manually. The collaboration aims to display AI results directly in the image data system, known as the PACS viewer, and make them interactively usable.

In the future, these results are to be automatically transferred to the findings templates, which will optimize workflows and make the findings more precise. Work is also underway to prioritize cases based on AI results in order to process critical findings more quickly.

Tangible results in the first year

The first improvements for routine radiology and patient care should already be visible in the first year. In addition to accelerated reporting, readability is to be specifically improved through structuring in order to achieve added clinical value in interdisciplinary communication. In the long term, the project aims to establish new standards for the use of AI and structured reporting.

"Our cooperation with the University Medical Center Freiburg is a significant step towards making radiology more efficient, data-driven and patient-centric," summarizes Michael Strüter, Managing Director of Dedalus HealthCare GmbH.