Han Wu 吴瀚
5th-year Ph.D. Student· ShanghaiTech University · Advised by Prof. Dinggang Shen
wuhan2022@shanghaitech.edu.cn
Shanghai, China
About
I am a Ph.D. student in Computer Science at the School of Biomedical Engineering, ShanghaiTech University, advised by Prof. Dinggang Shen. I earned my B.Eng. from Wuhan University of Technology in 2022.
My research focuses on multimodal large language models (MLLMs) for medical image analysis and reasoning. Since June 2025, I have been a research intern at United Imaging Intelligence (UII), where I work with Dr. Dijia Wu and Dr. Xi Ouyang on MLLMs and contribute to the YuanZhi Foundation Model.
Beyond my main research focus, I'm also interested in a wide range of topics, including digital dentistry, cardiovascular disease, and EEG analysis, where I've worked on projects myself, supervised junior graduate students, or collaborated with others.
Research Interests
Routine clinical care produces images, reports, and signals at scale, while expert annotations stay scarce and costly. My work runs along two lines that start from opposite sides of that gap:
Multimodal LLMs & Foundation Models
Medical image analysis and reasoning through vision-language pretraining and MLLM post-training, alongside self-supervised pretraining of foundation models on large-scale unlabeled clinical data.
Label-Efficient Learning
Segmentation and landmark detection with limited, weak, or partial annotations, using semi-supervised, weakly supervised, and partially supervised learning.
Education
Wuhan University of Technology
B.Eng. in Information Engineering
Outstanding Graduate; Postgraduate Recommendation
Experience
United Imaging Intelligence (UII)
Research Intern, R&D
Multimodal LLMs with Dr. Dijia Wu and Dr. Xi Ouyang
News
2026
- 09/2026One paper (cephalometric landmark detection across ages) is accepted by MedIA!
- 08/2026Our MVAA-UniSeg won 1st place in the MICCAI 2026 MVAA Challenge (Mitral Valve Anatomy Analysis)!
- 06/2026One paper (gingiva segmentation in CBCT) is accepted by MICCAI 2026 after rebuttal!
- 06/2026One paper (CTA–DSA vascular registration) is accepted by MedIA!
- 05/2026One paper (weakly-supervised coronary artery segmentation) is early accepted by MICCAI 2026!
2025
- 12/2025One paper (semi-supervised echocardiography landmark tracking) is accepted by IEEE TMI, a nice ending for 2025!
- 07/2025One paper (semi-supervised medical image segmentation) is accepted by IEEE TMI!
- 07/2025One paper (knowledge-informed diffusion model for tooth alignment) is accepted by MedIA!
- 06/2025One paper (semi-supervised echocardiogram segmentation) is accepted by MICCAI 2025 after rebuttal!
- 06/2025Started my research internship at United Imaging Intelligence (UII).
- 05/2025One clinical paper (condylar evaluation for orthognathic surgery) is accepted by Journal of Dentistry!
- 05/2025One paper (foundation-model-based dental caries detection) is early accepted by MICCAI 2025!
- 05/2025Google Scholar citations crossed 100.
- 03/2025One paper (CLIP in medical imaging survey) is accepted by MedIA!
2024
- 10/2024One paper (sparse-view CBCT reconstruction) is accepted by IEEE TMI!
- 06/2024One paper (cephalometric landmark detection across ages) is accepted by MICCAI 2024!
- 06/2024Passed the Ph.D. Qualification Exam (PQE) and officially became a Ph.D. candidate!
2023
- 12/2023Our survey on CLIP in medical imaging is online!
- 06/2023One paper (multi-view vertebra localization and identification) is accepted by MICCAI 2023!
2022
- 09/2022Started a new journey at ShanghaiTech!
- 06/2022Graduated from WUT with the “Outstanding Graduate” honor.
Publications
* indicates equal contribution and + indicates corresponding author.
Foundation & Multimodal Models
Taming Foundation Model with Invariance-Oriented Pre-training for Broad-Spectrum EEG Analysis across Signal-Level, Brain-State, and Brain-Health Tasks
arXiv preprint, 2026 · Under Review
Cardiovascular
Weakly-Supervised Coronary Artery Segmentation from DSA Sequence via Motion-Aware Modeling
MICCAI 2026 Early Accept · Top 9%
Two-Stage Robust 3D CTA–2D DSA Alignment via Vascular-Aware Rigid and Pyramid-Based Hierarchical Non-Rigid Registration
Medical Image Analysis (MedIA), 2026
A Semi-Supervised Knowledge Distillation Framework for Left Ventricle Segmentation and Landmark Detection in Echocardiograms
MICCAI 2025
Digital Dentistry
CeLDA+: Cephalometric Landmark Detection across Ages with Learnable Prototypical Network
Medical Image Analysis (MedIA), 2026
Cross-surface Representation Learning for Accurate Gingiva Prediction in CBCT
MICCAI 2026
Awards & Achievements
- 1st Place, MICCAI 2026 MVAA Challenge, 2026
- Outstanding Student of ShanghaiTech (Top 10%), 2023
- Outstanding Graduate of WUT, 2022
- Outstanding Graduation Thesis of WUT (Top 4%), 2022
- 2nd National Prize, Engineering Practice & Innovation Competition, 2021
- 3rd National Prize, “China Software Cup”, 2021
- 1st Prize, Hubei Engineering Practice & Innovation, 2021
- 3rd Prize, Intelligent Car Race (Hubei), 2021
- 3rd Prize, Computer Design Competition (Hubei), 2021
- Ranked 27/169, China Software Cup Online, 2021
- 3rd Prize, “Internet+” Innovation Competition (Hubei), 2020
- Ranked 3/218, Intelligent Car Race: Crowd Counting, 2020
- Ranked 15/332, Baidu AIStudio Bugs Detection, 2020
Academic Services
Conference Reviewer
Journal Reviewer
- IEEE Transactions on Medical Imaging (TMI)
- IEEE Transactions on Biomedical Engineering (TBME)
- IEEE Transactions on Geoscience and Remote Sensing (TGRS)
- ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM)
- Neural Networks
- Signal Processing
- Digital Signal Processing
Teaching & Mentorship
Teaching Assistant
BME2113 — Algorithms Design and Analysis (Python), ShanghaiTech, Fall 2023
Mentored Students
During my Ph.D. study, I also help supervise junior postgraduate and undergraduate students.
Tao Luo
- Adapting Foundation Model for Dental Caries Detection with Dual-View Co-Training MICCAI 2025 · Early Accept
Wei Jia
- Fully Automated Evaluation of Condylar Remodeling After Orthognathic Surgery JoD 2025
- National College Biomedical Engineering Innovation Design Competition 1st Prize · Top 5%
Invited Talks
MVAA-UniSeg: A Query-Conditioned Universal Model with Modality-Specific Experts for Mitral Valve Segmentation
Artificial Intelligence in Digital Dentistry
Knowledge-driven Landmark Detection in Medical Images
Miscellaneous
A big fan of Taylor Swift, Jay Chou, and JJ Lin.
Love playing guitar — Masaaki Kishibe is my favorite guitarist.
Enjoy League of Legends with friends — reached Master tier in LOLM.
Contact
Office
IDEA Lab, Room 405, BME Building, ShanghaiTech
Address
393 Middle Huaxia Road, Pudong, Shanghai, China