Graph Learning for Medical Image Analysis

This tutorial aims to highlight the latest advancements in graph learning for medical image analysis. We will discuss state-of-the-art techniques, current challenges, and opportunities in this field. Our goal is to provide the audience with a solid foundation in graph learning, along with insights into its current developments. Through real-world case studies—such as graphs for medical image classification and reconstruction—we will demonstrate the practical applications of graph learning in medical imaging.

Organizing Team

Angelica I Aviles-Rivero (Assistant Professor, Tsinghua University) centres on graph-based techniques for (bio-)medical applications, focused on novel functionals (PDEs) with carefully designed priors, allowing intertwining deep learning models. Recognitions include an outstanding paper award (ICML 2020) and elected officer (SIAM SIAG/IS 2022).
Chun-Wun (Sam) Cheng (Graduate Student, University of Cambridge) has an interest in the interaction between machine learning, implicit deep learning techniques, Diffusion Model and their application in the field of medical imaging and inverse problems.
Yanqi Cheng (Graduate Student, University of Cambridge) is interested in applying applied mathematics and machine learning for large-scale real-world problems. In particular, Yanqi is interested in combining knowledge-driven and data-driven modelling, and has been working on surgical data science and traffic flow analysis, which are large-scale problems.
Yue Gao (Associate Professor, Tsinghua University) is an Associate Professor in School of Software, Tsinghua University, Beijing, China. His research falls in the field of Artificial Intelligence and Graph Learning. He leads the iMoon laboratory. He has large experience in developing Hypergraph Neural Networks and complex interaction networks with a wide range of real-world applications including medical image analysis.
Xiangmin Han (Postdoctoral Researcher, Tsinghua University) is a postdoctoral researcher at Tsinghua University. His research interests include hypergraph computation and medical image analysis, particularly brain network analysis and pathological image analysis.
Jiahao Huang (Graduate Student, Imperial College London) is interested in deep learning-based techniques for inverse problems in magnetic resonance imaging (MRI) like reconstruction and denoising, and relative application for diffusion tensor MRI. He focuses on the graph-based model for MRI reconstruction.
Carola-Bibiane Schonlieb (Professor, University of Cambridge) is currently a Professor of applied mathematics with DAMTP, the Head of the Cambridge Image Analysis Group and the Director of the Cantab Capital Institute for Mathematics of Information. Her research interests include variational methods, partial differential equations, machine learning for image analysis, image processing, and inverse imaging problems.
Emma Shujun Wang (Assistant Professor, The Hong Kong Polytechnic University) specializes in applications on digital health and computational precision medicine designing AI-driven computational methods to enable reliable medical decision-making for precision medicine, covering from disease diagnosis to prognosis, and from medical image computing to multi-modal biomedical data integration.
Guang Yang (Associate Professor, Imperial College London) has a research group interested in developing novel and translational techniques for imaging and biomedical data analysis. His group focuses on the research and development of data-driven fast imaging, data harmonisation, image segmentation, image synthesis, federated learning, explainable AI etc. He is currently working on a wide range of clinical applications in cardiovascular disease, lung disease and oncology.
Lequan Yu (Assistant Professor, The University of Hong Kong) is an Assistant Professor at University of Hong Kong. He was a postdoctoral research fellow at Stanford Medical School and obtained his Ph.D. from CUHK in 2019. His research interests are designing advanced machine learning algorithms for biomedical data analysis, primarily focusing on medical images. He has published more than 60 top-tier papers and has been cited over 12000 times with a 46 h-index. He ranked top 2% of Scientists on Stanford List (2022, 2023) and was on the World’s First List of Top 150 Chinese Young Scholars in AI (2022). He has co-organized the ICCV 2021&2023 workshop on Computer Vision for Automated Medical Diagnosis. He served as a senior program committee for IJCAI 2021 and AAAI 2022, and an area chair for MICCAI 2022 & 2023.
Lipei Zhang (Graduate Student, University of Cambridge) is interested in physics-informed deep learning for image processing. Specialising in reconstruction, denoising, and segmentation of medical and natural images, he combines robust computational skills with a profound understanding of PDE equations, driving innovation in advanced image analysis techniques.

Tutorial Content

Lecture 1: Semi-Supervised Graph Learning for Medical Image Analysis
Speaker: Angelica I Aviles-Rivero
Lecture 2: Graph Neural Networks
Speaker: Yanqi Cheng
Lecture 3: Graph Transformers
Speaker: Chun-Wun (Sam) Cheng
Lecture 4: Higher-Order Learning—Hypergraph Computation for Medical Data
Speaker: Yue Gao
Lecture 5: Graph Learning for MRI Reconstruction
Speaker: Jiahao Huang
Lecture 6: Whole Slide Pathological Image Analysis Based on Hypergraph Computation
Speaker: Xiangmin Han
Lecture 7: Graph Transformers for Whole Slide Image Analysis
Speaker: Lequan Yu

Time Slot

To be determined (TBD)

MICCAI 2025

Visit the MICCAI 2025 Official Website

Join us to explore the cutting-edge of graph learning and its transformative applications in medical image analysis. Through real-world case studies, gain insights into the challenges and opportunities in this exciting field!