Tianyu Han

Dr. Tianyu Han is a postdoc researcher at the University Hospital RWTH Aachen’s Department of Diagnostic and Interventional Radiology. With a PhD in Physics from RWTH Aachen University, his doctoral study explored generative modeling for medical image analysis, resulting in multiple high-impact publications in Nature Machine Intelligence, Nature Communications, Radiology, and Science Advances. Since completing his PhD, Dr. Han has extended his research to developing and deploying advanced large language models (LLMs) for medical applications. His notable contributions include the creation of MedAlpaca, an open-source medical LLM, and the compilation of Medical Meadow, a substantial medical dataset designed for LLM training. His research has demonstrated the practical utility of multimodal LLMs in clinical settings, contributing to their adoption for general medical image interpretation. Dr. Han’s work has also led to significant advancements in using adversarially trained models for pathology detection and the anonymous sharing of medical data through generative models. His innovative methodologies have not only enhanced diagnostic accuracies but also facilitated the understanding of radiological confounders on an individual level.