We have received funding from the German Federal Ministry of Education and Research (BMBF) for the next three years to investigate how large language models (LLMs) can be used in the context of patient triage in emergency rooms. Instead of filling out forms, patients will share their symptoms with an AI application, which will then suggest patient prioritization to medical staff. While promising, the introduction of LLMs into the healthcare context raises many privacy concerns, but also how to design such interactions is largely underexplored.

While the ILLUMINATION project focuses on developing privacy-preserving methods for LLMs, the HCC focuses on understanding user attitudes towards LLMs in healthcare. We will investigate how to provide meaningful interactive user interfaces for LLM-based applications that allow patients to provide their data for pre-prioritization. By integrating Human-Centered Explainable AI (HC-XAI), we will ensure clear communication of AI decisions to clinical staff.

For more details, please visit the project website for more information.