Dr. LLM

The Dr. LLM study evaluates the diagnostic potential of large language models (LLMs) in infectious disease medicine, focusing on accuracy, reliability, and effectiveness. Anonymized, transformed clinical cases are prompted into LLMs to generate a differential diagnoses list, which is then compared to those made by physicians. The study also examines how factors such as different models, prompt parameters and case text structure influence outcomes, aiming to develop practical recommendations for LLM use in clinical settings.