Xiangde Luo
Postdoctoral Scholar at Stanford University
Palo Alto, California, United States
Summary
🤩
Rockstar🎓
Top SchoolXiangde Luo is a postdoctoral scholar based in Palo Alto with a decade of experience developing annotation-efficient medical imaging AI for clinical use, particularly in radiotherapy. He builds practical tools and public resources—most notably the widely adopted SSL4MIS semi-supervised segmentation benchmark (2k+ stars) and the large-scale WORD abdominal organ dataset—and has organized community challenges like SegRap2023. His work spans self-, semi-, and weakly-supervised methods, active learning, and human-in-the-loop systems, and has garnered significant scholarly impact (≈4,800 citations, h-index 19). Comfortable bridging research and engineering, he frequently contributes code and training optimizations to open-source projects, reflecting a hands-on focus on making robust models reproducible and clinically applicable.
10 years of coding experience
Ph.D. student Medical Image Computing, Ph.D. student Medical Image Computing at University of Electronic Science and Technology of China
Github Skills (8)
mask-rcnn10
net10
faster-rcnn10
pytorch10
computer-vision10
machine-learning10
medical-image-segmentation10
semi-supervised-learning10
Programming languages (5)
Github contributions (5)
HiLab-git/SSL4MIS
Sep 2020 - Jan 2023
Semi Supervised Learning for Medical Image Segmentation, a collection of literature reviews and code implementations.
Role in this project:
ML Engineer
ML EngineerContributions:432 commits, 11 PRs, 458 pushes in 2 years 3 months
Contributions summary:Xiangde primarily contributed to the development and training of semi-supervised learning models for medical image segmentation. Their commits involve modifications to training scripts for various models (U-Net variants), indicating experimentation with different semi-supervised approaches like entropy minimization, mean teacher models, and uncertainty-aware methods. Code changes include adjustments to learning rates, patch sizes, and loss functions, showing a focus on optimizing the training process.
medical-image-segmentationsemi-supervised-learning
HiLab-git/WSL4MIS
Dec 2020 - Dec 2022
Scribbles or Points-based weakly-supervised learning for medical image segmentation, a strong baseline, and tutorial for research and application.
Contributions:52 commits, 1 PR, 52 pushes in 2 years
medical-image-segmentationweakly-supervised-learningweakly-supervised-segmentation
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.