Bio
I am a postdoctoral researcher investigating interpretable machine learning (ML) and large language model (LLM) applications in clinical radiology. My current research focuses on two complementary areas: understanding what human-interpretable concepts self-supervised vision foundation models learn through mechanistic interpretability techniques like sparse autoencoders, and developing LLM-based systems, including agentic workflows and retrieval augmented generation (RAG) architectures, that leverage unstructured hospital data to improve radiological workflows. I earned my PhD from LMU Munich, where I focused on clinically motivated machine learning applications in medical imaging in the Department of Radiology.
My work is partially funded by a Walter Benjamin Fellowship from the DFG (German Research Foundation).
Professional Education
-
Doctorate (PhD / Dr. rer. nat.), LMU Munich, Munich, Germany, Machine Learning in Medical Imaging
-
Master of Science, LMU Munich, Munich, Germany, Physics
-
Bachelor of Science, Heidelberg University, Heidelberg, Germany, Physics