MHKD: Multi-Step Hybrid Knowledge Distillation for Low-Resolution Whole Slide Images Glomerulus Detection

We pioneer glomerulus detection on low-resolution kidney pathology images, proposing a multi-step hybrid knowledge distillation method that successively transfers global features and semantic information from high-resolution models through offline and online distillation, achieving AP0.5:0.95 improvements of 23.1% (LN) and 15.9% (HUBMAP).

摘要

Glomerulus detection is a critical component of renal histopathology assessment, essential for diagnosing glomerulonephritis. To mitigate the increasing workload on pathologists, AI-assisted diagnostic methods based on high-resolution digital pathology whole slide images have been developed. However, these current AI-assisted approaches are limited to high-resolution whole slide images, necessitating expensive digital scanner equipment, high image storage costs, and significant computational complexity. To address this limitation, this paper pioneers a method for facilitating glomerulus detection in low-resolution human kidney pathology images. Specifically, we propose a novel multi-step hybrid knowledge distillation method. Our method distills both the global features and the semantic information through a hybrid knowledge distillation strategy that integrates offline and online knowledge distillation, where the information from high-resolution pathological images is successively transferred to student model from the global features in the shallow network layers to the semantic information of the back-end through a multi-step training strategy. Experimental results on two datasets show that the proposed method achieves effective detection outcomes for low-resolution kidney pathology images. Compared to other state-of-the-art detection techniques, our method achieves an AP 0.5:0.95 improvement of 23.1% on the private LN dataset and 15.9% on the public HUBMAP dataset.

主要结果

BibTeX

@article{zhang2025mhkd,
  author  = {Zhang, Xiangsen and Han, Longfei and Xu, Chenchu and Zheng, Zhaohui and Ding, Jin and Fu, Xianghui and Zhang, Dingwen and Han, Junwei},
  title   = {MHKD: Multi-Step Hybrid Knowledge Distillation for Low-Resolution Whole Slide Images Glomerulus Detection},
  journal = {IEEE Journal of Biomedical and Health Informatics},
  volume  = {29},
  number  = {2},
  pages   = {767--774},
  year    = {2025},
  doi     = {10.1109/JBHI.2024.3513716}
}