What: Virtual lecture series on topics across machine learning in medicine, featuring extensive Q & A and panel discussions
Why: To reduce academia’s carbon footprint, accommodate the schedules of the world’s top scientists and maintain social distancing
Who: All are welcome!
Where: Zoom Webinar
Recordings of previous talks can be found here.
Miaomiao Zhang, PhD
Assistant Professor, ECE/CS, University of Virginia
When: Friday, November 14, 2025, 11am-12pm Eastern Time
Title: Learning Diffusion Models In Deformation Spaces For Medical Image and Video Generation
Abstract: Generative diffusion models have rapidly gained widespread attention for applications in image/video synthesis, classification, segmentation, and object tracking. However, existing diffusion-based image generation techniques primarily focus on manipulating intensity and texture features, often failing to preserve the topology and geometry of objects presented in images. This limitation is critical, especially in sensitive fields such as healthcare and medicine, where the accurate representation of human anatomy is required. In this talk, I will present our recent research on developing novel topology-preserving generative diffusion models that learn the underlying data distribution of latent diffeomorphic deformation features from images. Our approach enforces a geodesic flow of transformations, parameterized by a time sequence of smooth vector fields governed by nonlinear differential equations, combined with various types of conditions from text, image, or clinical variables to inform the network's learning process. I will showcase experimental results that demonstrate our model’s ability to maintain the topological structure of objects, hence improving the reliability of generative models in high-stakes domains.
Bio: Dr. Miaomiao Zhang is an Assistant Professor in Electrical and Computer Engineering and Computer Science at the University of Virginia. Her research work focuses on developing novel models at the intersection of mathematics, statistics, and computer engineering in the field of biomedical imaging. Dr. Zhang's current research projects include developing machine learning methods in image registration, segmentation, and shape analysis, with applications to cardiovascular imaging, neuroimaging, and computer-assisted surgery. Her work was recognized by numerous prestigious awards, including the MICCAI Young Scientist Award, UVA Research Innovation Award, NSF CAREER Award, and NIH Trailblazer Award. Dr. Zhang is a member of MICCAI society and has been actively serving as an area chair for MICCAI, ISBI, and MIDL.
We are currently inviting speakers for Fall 2025! If you are interested or would like to nominate a speaker, please fill out this nomination form.