Pei Liu

I am a 3rd year Ph.D. Candidate advised by Prof. Luping Ji at University of Electronic Science and Technology of China (UESTC). Prior to starting my Ph.D., I obtained my B.E./M.S. in School of Computer Science and Engineering at University of Electronic Science and Technology of China.

My research interests mainly lie in computational pathology (CPATH), multiple instance learning (MIL), survival analysis, predictive uncertainty, and multi-modal learning. Most of my research papers have been published in top-tier journals and conferences like ICML, IEEE TMI, MedIA, etc. I also serve as a regular reviewer in IEEE JBHI, Computers in Biology and Medicine, ESWA, etc.

news

May 2, 2024 :fire: I am thrilled to announce that one paper (arXiv, OpenReview) is accepted at ICML 2024. Please see more in our Github repo (on updating).
Mar 31, 2024 :sparkles: One paper is accepted by Computer Methods and Programs in Biomedicine.
Jan 25, 2024 :fire: Our lab and related works are reported by UESTC Official News.
Jan 3, 2024 :sparkles: One paper is accepted by IEEE Trans. on Medical Imaging.
Oct 31, 2023 :sparkles: One paper is accepted by Medical Image Analysis.

selected publications

  1. 2024-icml-mirel.png
    Weakly-Supervised Residual Evidential Learning for Multi-Instance Uncertainty Estimation
    Pei Liu, and Luping Ji
    In Proceedings of the 41st International Conference on Machine Learning, 2024
  2. conceptual-psemix.png
    Pseudo-Bag Mixup Augmentation for Multiple Instance Learning-Based Whole Slide Image Classification
    Pei Liu, Luping Ji, Xinyu Zhang, and Feng Ye
    IEEE Transactions on Medical Imaging, 2024
  3. arch-advmil.png
    AdvMIL: Adversarial multiple instance learning for the survival analysis on whole-slide images
    Pei Liu, Luping Ji, Feng Ye, and Bo Fu
    Medical Image Analysis, 2024
  4. dsca-arch.png
    DSCA: A dual-stream network with cross-attention on whole-slide image pyramids for cancer prognosis
    Pei Liu, Bo Fu, Feng Ye, Rui Yang, and Luping Ji
    Expert Systems with Applications, 2023
  5. arch-prodiv.png
    ProDiv: Prototype-driven Consistent Pseudo-bag Division for Whole-slide Image Classification
    Rui Yang, Pei Liu, and Luping Ji
    Computer Methods and Programs in Biomedicine, 2024