Yixuan Yuan
Associate Professor
Leads the AI in Medicine group, working on medical image analysis and abnormality detection with deep learning.
Yixuan Yuan's lab at CUHK works on medical imaging & healthcare AI and recognition, detection & segmentation. 103 papers since 2023, 48 at top venues, mostly TMI and IJCV. 16 PhD students. 7 alumni, most often to Meta and Tencent. Applications close .
At a glance
- papers, 2023–2026
- 27/ year
- at T1 venues
- 47%
- PhD students
- 16
- MS students
- 0
- alumni listed
- 7
Papers by yearfrom the lab homepage, 2023–2026
- 15 papers2026 · T1 9 · collecting
- 26 papers2025 · T1 14
- 35 papers2024 · T1 15
- 27 papers2023 · T1 10
T1 venuesOtherVenues: TMI, IJCV, ICCV, MICCAI, NeurIPS
Research focuspaper titles that use each topic's words
- Recognition, detection & segmentation26 papers
- Medical imaging & healthcare AI25 papers
- medical image14 papers
- image analysis1 paper
- medicine1 paper
Paper list
From the lab homepage.
202615 papers
- npj Digital MedicineAn Explainable Generative AI System for Video-to-Report Generation in Capsule Endoscopy
- Cell Reports Medicine - highlightsDeveloping and Evaluating Multimodal Large Language Model for Orthopantomography Analysis to Support Clinical Dentistry
- IJCV (T1)FedPD++: Enhanced Federated Open-Set Recognition with Parameter Disentanglement
- TMI (T1)Fiber HGNN: Heterogeneous Graph Neural Network for Fiber Tract Segmentation
- TPAMI (T1)Harnessing Lightweight Transformer with Contextual Synergic Enhancement for Efficient 3D Medical Image Segmentation
- npj Digital MedicineInterpretable Agentic AI System with Localized Reasoning for Radiology
- npj Health SystemMAP: Evaluation and Multi-Agent Enhancement of Large Language Models for Inpatient Pathways
- MICCAI (T1)MedPruner: Training-Free Hierarchical Token Pruning for Efficient 3D Medical Image Understanding in Vision-Language Models
- IJCV (T1)MRM++: Enhanced Masked Relation Modeling for Multi-Modal Medical Pre-Training
- CVPR-26OmniBrainBench: A Comprehensive Multimodal Benchmark for Brain Imaging Analysis Across Multi-stage Clinical Tasks
- TMI (T1)PIPA: Prior-Driven Prompting with Diagnosis-Oriented Retrieval-Augmentation for 3D Radiology Report Generation
- ECCV (T1)SpecV: Specification Verification for Robust Unified Multimodal Evaluation
202526 papers
- ACL-25A Survey of LLM-based Agents in Medicine: How far are we from Baymax?
- NeurIPS (T1)EndoBench: A Comprehensive Evaluation of Multi-Modal Large Language Models for Endoscopy Analysis
- MICCAI (T1)EndoGen: Conditional Autoregressive Endoscopic Video Generation
- MedIAFedBM: Stealing knowledge from pre-trained language models for heterogeneous federated learning
- NeurIPS (T1)FedGPS: Goal-Path Synergy for Robust Federated Training Against Data Heterogeneity
- TMI (T1)Foundation Model-guided Gaussian Splatting for 4D Reconstruction of Deformable Tissues
- MedIAGAGM: Geometry-aware graph matching framework for weakly supervised gyral hinge correspondence
- ICCV (T1)GaussianReg: Rapid 2D/3D Registration for Emergency Surgery via Explicit 3D Modeling with Gaussian Primitives
- IPMI-25GeoT: Geometry-guided Instance-dependent Transition Matrix for Semi-supervised Tooth Point Cloud Segmentation
- ICRA-25Hide-in-Motion: Embedding Steganographic Copyright Information into 4D Gaussian Splatting Assets
- NeurIPS (T1)HumanCrafter: Synergizing Generalizable Human Reconstruction and Semantic 3D Segmentation
- ICCV (T1)InfoBridge: Balanced Multimodal Integration through Conditional Dependency Modeling
202435 papers
- ECCV (T1)Alternate Diverse Teaching for Semi-supervised Medical Image Segmentation
- IJCV (T1)Bi-VLGM : Bi-Level Class-Severity-Aware Vision-Language Graph Matching for Text Guided Medical Image Segmentation
- ECCV (T1)CLIFF: Continual Latent Diffusion for Open-Vocabulary Object Detection
- MedIAComprehensive Learning and Adaptive Teaching: Distilling Multi-modal Knowledge for Pathological Glioma Grading
- TMI (T1)DEeR: Deviation Eliminating and Noise Regulating for Privacy-preserving Federated Low-rank Adaptation
- MICCAI-24DiffRect: Latent Diffusion Label Rectification for Semi-supervised Medical Image Segmentation
- TMI (T1)Disentangle Then Calibrate with Gradient Guidance: A Unified Framework for Common and Rare Disease Diagnosis
- MICCAI-24Endora: Video Generation Models as Endoscopy Simulators
- MICCAI-24EndoSparse: Real-Time Sparse View Synthesis of Endoscopic Scenes using Gaussian Splatting
- BIBMEnhancing Clinical Information for Zero-Shot Medical Diagnosis by Prompting Large Language Model
- MICCAI-24F2TNet: FMRI to T1w MRI Knowledge Transfer Network for Brain Multi-phenotype Prediction
- TMI (T1)FedOSS: Federated Open Set Recognition via Inter-client Discrepancy and Collaboration
202327 papers
- IJCV (T1)A Comprehensive Study of The Robustness for LiDAR-Based 3D Object Detectors Against Adversarial Attacks
- CVPR-23Adjustment and Alignment for Unbiased Open Set Domain Adaptation
- ICCV-23CLIP-Driven Universal Model for Organ Segmentation and Tumor Detection
- TASECMCNet: Colorization-Aware Mix-Uncertainty-Adaptive Consistency Network for Semi-Supervised Fruit Counting
- MICCAI-23Conditional Physics-Informed Graph Neural Network for Fractional Flow Reserve Assessment
- TNNLS (T1)Decoupled Unbiased Teacher for Source-free Domain Adaptive Medical Object Detection
- TMI (T1)Disentangle First, Then Distill: A Unified Framework for Missing Modality Imputation and Alzheimer's Disease Diagnosis
- MedIADynamic Feature Splicing for Few-Shot Rare Disease Diagnosis
- CVPR-23EfficientViT: Memory Efficient Vision Transformer with Cascaded Group Attention
- TMI (T1)FedDM: Federated Weakly Supervised Segmentation via Annotation Calibration and Gradient De-conflicting
- ICCV-23FedPD: Federated Open Set Recognition with Parameter Disentanglement
- IJCV (T1)Generalized Gradient Flow Based Saliency for Pruning Deep Convolutional Neural Networks
Where alumni went7 listed, 50% to industry
Industry 3Academia 3
Most often to
- Meta
- Tencent
Also to
- Imperial College London
- Stanford University
- The University of Hong Kong
Apply to CUHKFall 2027
Computer Science and Engineering MPhil/PhD
- Round
- HKPFS / main round
- Opens
- —
- GRE
- See program page
- English
- Interview in English
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- GBUCL
Peter Bentley
- Genomics & computational biology
- USUC Berkeley
Michael Lustig
- Medical imaging & healthcare AI
- CNSJTU
Shaoting Zhang
- Medical imaging & healthcare AI
- USUCLA
Yuchen Cui
- Efficient training & inference
- CNPeking
Tao Wang
- Computational neuroscience & cognition
- CNSJTU
Xiaofan Zhang
- Medical imaging & healthcare AI
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
CUHKcloses