About me [CV]

• I am a postdoctoral researcher in the Department of Radiation Oncology at Stanford Medicine, focusing on AI for precision oncology, supervised by Prof. Ruijiang Li.
• Previously, I received my PhD (2024) and Bachelor’s degree (2018) from the School of Mechanical and Electrical Engineering at the University of Electronic Science and Technology of China, advised by Prof. Shaoting Zhang. My doctoral dissertation (in Chinese) is available here. I also completed research internships at Shanghai AI Lab (3 years), SenseTime (1 year), and West China Hospital (4 months).

Recent News

• November 2025, I transferred to computational pathology for precision oncology. Here, we developed a codebase nnMIL for efficient computational pathology research.
• September 2025, I was listed in the Stanford/Elsevier World’s Top 2% Scientists 2025.
• May 2025, served as an Area Chair for MICCAI 2025.
• December 23, 2024, the challenge report of SegRap2023 organized in conjunction with MICCAI2023 was accepted to Medical Image Analysis thanks to all co-authors and participants. BTW, we will organize SegRap2025 in conjunction with MICCAI2025, see you in Daejeon, Republic of Korea.
• November 3, 2024, A clinical paper about source-free active domain adaptation for NPC GTVp segmentation across multiple hospitals and multiple raters was accepted by International Journal of Radiation Oncology*Biology*Physics (also named RedJournal, one of the top-tier journals in the radiation and oncology field, double-blind review), thanks to all co-authors.
• On June 17, 2024, Three papers were accepted to MICCAI2024 (two early acceptances), one about the robustness segmentation of abdominal organs (RAOS) and two about source-free active domain adaptation (SFADA-UWF-SLO and UGTST), congrats to Hongqiu, Zihao, Zihan.
• June 6, 2024, A paper about source-free active domain adaptation for NPC GTVp segmentation across multiple hospitals (ArXiv) was accepted by IEEE Transaction on Medical Imaging, congrats to Mr. Hongqiu Wang.
• I passed my PhD defence on May 28, 2024!
• May 19, 2023, A clinical paper about automatic organs at risk segmentation on heterogeneous computed tomography images for abdominal radiotherapy was accepted by International Journal of Radiation Oncology*Biology*Physics (also named RedJournal, one of the top-tier journals in the radiation and oncology field, double-blind review), congratulate M.D. Wenjun Liao.
• Apr. 10, 2023, We will host an automatic segmentation challenge SegRap2023 at MICCAI2023 focusing on GTVnx, GTVnd, and 45 OARs segmentation for nasopharyngeal carcinoma radiotherapy planning. Welcome to participate in it.
• Mar. 23, 2023, Our paper Towards Multi-Center Cross Tissues Histopathological Cell Segmentation via Target-Specific Finetuning was accepted by IEEE Transaction on Medical Imaging, congrats to Dr. Zhongyu Li.
• Jan. 9, 2023, Our paper Deep learning-based accurate and robust delineation of primary gross tumor volumes of nasopharyngeal carcinoma on heterogeneous magnetic resonance imaging: a large-scale and multi-center study was accepted by Radiation and Oncology (also named GreenJournal, one of the top-tier journals in the radiation and oncology field), thanks to all co-authors, code and paper are available.
• Sep. 20, 2022, Our paper WORD: A large scale dataset, benchmark and clinical applicable study for abdominal organ segmentation from CT image was accepted by Medical Image Analysis, thanks to all co-authors, dataset and paper are available.
• July 12, 2022, Our project SSL4MIS received the 1000th star, a milestone moment for us; keep going!!!
• Jun 30, 2022, Our MICCAI2022 paper (WSL4MIS) was selected to receive a MICCAI2022 Student Travel Award.
• Jun 10, 2022, Our paper Semi-Supervised Medical Image Segmentation via Uncertainty Rectified Pyramid Consistency was accepted by Medical Image Analysis, thanks to all co-authors, code and paper are available.
• May 5, 2022, Our paper Scribble-Supervised Medical Image Segmentation via Dual-Branch Network and Dynamically Mixed Pseudo Labels Supervision was provisionally accepted by MICCAI2022 (top 13% in total 1825 submissions), thanks to all co-authors, code and paper are available.
• Mar. 25, 2022, A clinical paper about semi-supervised Gross Target Volume (GTVnx and GTVnd) of Nasopharyngeal Carcinoma (NPC) segmentation was accepted by International Journal of Radiation Oncology*Biology*Physics (also named RedJournal, one of the top-tier journals in the radiation and oncology field, double-blind review), this work is a clinical applicable study of our work MICCAI2021, congratulate M.D. Wenjun Liao.
• Mar. 8, 2022, We have done a comprehensive study about scribble-supervised medical image segmentation based on the ACDC dataset, where more than ten weakly-/semi-supervised methods are tested on the same setting (five-fold cross-validation). The tech report and Code are available.
• Mar. 1, 2022, Our paper Semi-Supervised Medical Image Segmentation via Cross Teaching between CNN and Transformer was accepted by MIDL2022, thanks to all co-authors, code and paper are available.
• Oct. 21, 2021, Our paper SCPM-Net: An Anchor-free 3D Lung Nodule Detection Network using Sphere Representation and Center Points Matching was accepted by Medical Image Analysis, thanks to all co-authors, code and paper are available.
• Aug. 29, 2021, We released a 2D inference code and GUI of MIDeepSeg (published in MedIA2021), the repo at MIDeepSeg.
• Jun. 12, 2021, Two co-authors’ papers were accepted by MICCAI 2021 (~33% acceptance rate) after the rebuttal, congratulate all collaborators.
• May 14, 2021, Our paper Efficient Semi-Supervised Gross Target Volume of Nasopharyngeal Carcinoma Segmentation via Uncertainty Rectified Pyramid Consistency was early accepted by MICCAI2021 (top 13% in total 1630 submissions), thanks to all co-authors.
• May 06, 2021, Our paper MIDeepSeg: Minimally Interactive Segmentation of Unseen Objects from Medical Images Using Deep Learning was accepted by Medical Image Analysis, thanks to all co-authors, code and paper.
• April 29, 2021, One co-author paper Medical Image Segmentation using Squeeze-and-Expansion Transformers was accepted by IJCAI2021 (~13% acceptance rate), congratulate Dr. Shaohua Li.
• Jan. 25, 2021, We released a code base and some 2D examples for weakly-supervised medical image segmentation research, the repo at WSL4MIS, any advice and suggestions are welcomed.
• Jan. 08, 2021, Our paper Deep Elastica for Image Segmentation was accepted by ISBI 2021.
• Dec. 01, 2020. Our paper Semi-supervised Medical Image Segmentation through Dual-task Consistency was accepted by AAAI 2021 (~21% acceptance rate).
• Oct. 07, 2020. We released a code base and some examples (both 2D and 3D) for semi-supervised medical image segmentation research, the repo at SSL4MIS, any advice and suggestions are welcomed.