Medical atlases are a staple in neuroimaging research and play a critical role in broader MR data analysis. They offer a standardized spatial and structural reference space for population-level studies, enabling researchers to investigate anatomical variability, model disease progression, and understand underlying population dynamics. In this talk, I will present current research directions in building advanced atlases within high-dimensional data spaces using generative AI. These include tackling high dimensional whole body MR images and leveraging recent advances in diffusion models to develop fast, interpretable, and conditionally generated atlases. These innovations open new avenues for scalable, data-driven understanding of anatomy and pathology
Sophie Starck is a Ph.D. student at the Institute for Artificial Intelligence and Informatics in Medicine at the Technical University of Munich (TUM). She received her Bachelor’s degree in Computer Science and Engineering, followed by a Master’s degree in image processing and computer graphics from EPITA in 2021. Her work consists in developing Machine Learning models applied to medical images and medical applications. Her research interests are focused on medical atlases, population analysis, generative AI and geometric deep learning for medical analysis and diagnosis.
