Hard Negative Sample Mining for Whole Slide Image Classification

Wentao Huang, Xiaoling Hu, Shahira Abousamra, Prateek Prasanna, Chao Chen
10/3/2024
cs.CV

Official code: winston52/HNM-WSI (4 stars, last commit 2024-07-24).

Abstract

Weakly supervised whole slide image (WSI) classification is challenging due to the lack of patch-level labels and high computational costs. State-of-the-art methods use self-supervised patch-wise feature representations for multiple instance learning (MIL). Recently, methods have been proposed to fine-tune the feature representation on the downstream task using pseudo labeling, but mostly focusing on selecting high-quality positive patches. In this paper, we propose to mine hard negative samples during fine-tuning. This allows us to obtain better feature representations and reduce the training cost. Furthermore, we propose a novel patch-wise ranking loss in MIL to better exploit these hard negative samples. Experiments on two public datasets demonstrate the efficacy of these proposed ideas. Our codes are available at https://github.com/winston52/HNM-WSI

View on arXivView PDF

Code Implementations(1)

winston52/HNM-WSIOfficial100%
40PythonJul 4, 20242 years ago

Similar Papers

Task-oriented Embedding Counts: Heuristic Clustering-driven Feature Fine-tuning for Whole Slide Image Classification

89% match2024

Multiple Instance Learning Framework with Masked Hard Instance Mining for Whole Slide Image Classification

85% match1 repo982026

PreMix: Label-Efficient Multiple Instance Learning via Non-Contrastive Pre-training and Feature Mixing

85% match1 repo2024

Rethinking Pre-Trained Feature Extractor Selection in Multiple Instance Learning for Whole Slide Image Classification

84% match1 repo12024

PATHS: A Hierarchical Transformer for Efficient Whole Slide Image Analysis

84% match1 repo92024

SimMIL: A Universal Weakly Supervised Pre-Training Framework for Multi-Instance Learning in Whole Slide Pathology Images

83% match2025

Cite this paper

@article{huang2024hard,
  title  = {Hard Negative Sample Mining for Whole Slide Image Classification},
  author = {Wentao Huang and Xiaoling Hu and Shahira Abousamra and Prateek Prasanna and Chao Chen},
  year   = {2024},
  eprint = {2410.02212},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2410.02212v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2410.02212.svg)](https://codeofpaper.com/paper/2410.02212)

Discussion