Invited Talk: Suprosanna Shit, PhD & Chinmay Prabhakar (University of Zürich) – 3D Vessel Graph Generation Using Denoising Diffusion, Jan 31st 2025, 10 AM CET

It’s great pleasure to welcom Supro and Chinmay to our lab!

Title: 3D Vessel Graph Generation Using Denoising Diffusion
Date: Jan 31st 2025, 10 AM CET
Location: https://fau.zoom-x.de/j/63625938245?pwd=dP1KCW4At9EQRN2wudZtjVgM7B1QCv.1

Abstract
Blood vessel networks, represented as 3D graphs, help predict disease biomarkers, simulate blood flow, and aid in synthetic image generation, relevant in both clinical and pre-clinical settings. However, generating realistic vessel graphs that correspond to an anatomy of interest is challenging. Previous methods aimed at generating vessel trees mostly in an autoregressive style and could not be applied to vessel graphs with cycles such as capillaries or specific anatomical structures such as the Circle of Willis. Addressing this gap, we introduce the first application of \textit{denoising diffusion models} in 3D vessel graph generation. Our contributions include a novel, two-stage generation method that sequentially denoises node coordinates and edges. We experiment with two real-world vessel datasets, consisting of microscopic capillaries and major cerebral vessels, and demonstrate the generalizability of our method for producing diverse, novel, and anatomically plausible vessel graphs.

Suprosanna Shit, PhD:
I am currently a postdoctoral researcher at the University of Zürich (UZH) and an associate researcher at the ETH AI Center. My primary research focuses on medical imaging applications using neural implicit representation. I am also interested in extracting efficient and meaningful representations from the image domain for topology and graph domain processing. 

Chinmay Prabhakar:
I am a Ph.D. candidate supervised by Prof. Bjoern Menze at the University of Zürich (UZH) and an associate researcher at the ETH AI Center. My research interests focus on developing basic generalizable and efficient machine learning algorithms in the field of medical image analysis using graphs. My research interests include graph processing, self-supervised learning, and generative modeling.