Yaoyu Liu
Ph.D. Student
Institute of Medical Robotics (IMR), Shanghai Jiao Tong University
Research Interests
- Medical Image Analysis
- Vessel & Airway Segmentation
- Topology-Aware Modeling
- Foundation Models
About Me
I am a Ph.D. student at the Institute of Medical Robotics (IMR), Shanghai Jiao Tong University, advised by Prof. Yun Gu. My research focuses on topology-aware medical image analysis — including segmentation and centerline tracking of vessel-like tubular structures, and foundation models for topological knowledge of vascular anatomy.
ATM26 Challenge Organizer. I am a member of the organizing team for the ATM26 challenge (Airway Tree Modeling for Endobronchial Surgery). See the challenge website and the public leaderboard.
Research Overview
"Building topology-aware models that explicitly understand connectivity and branching of vessel-like anatomy for reliable medical image analysis."
Topology-Aware Foundation Models
Developing multimodal foundation models that unify topological knowledge understanding and segmentation generation for vessel-like tubular anatomy (TubeMLLM).
Vessel Skeleton Tracking
Designing graph-based centerline tracking methods that preserve topological consistency for vessel modeling and navigation (TopoVST).
Latest News
View AllSelected Publications
Set `showOnHomepage` and `featuredOrder` in `data/publications.json` to control what appears here.
Honors & Awards
View AllAcademic Service
MICCAI 2026 ATM26 Challenge Organizer
Organizing Team Member
Task design, data & annotation specification, evaluation metrics, and Grand-Challenge platform deployment & maintenance.
Education
Ph.D. in Control Science and Engineering
Sep 2023 - Jun 2028 (expected)
School of Automation and Perception, Shanghai Jiao Tong University
B.E. in Automation
Sep 2019 - Jun 2023
School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University
Contact
Office Address
Institute of Medical Robotics (IMR), Shanghai Jiao Tong University, Shanghai, China.
lyyu19@sjtu.edu.cn
ATM26 Challenge
Visit the public leaderboard for the Validation and Final Test phases.