WSI Stain Classifier / WSI 染色分类器

An automated whole slide image (WSI) stain type classification tool.
全切片图像（WSI）染色类型自动分类工具。

The model is trained on 1000+ WSIs from ACROBAT2023 dataset (5 stains). We use a liner prob based on the GPFM.

Features / 功能特性

✅ Multi-format Support
✅ 多格式支持

+ Supported formats: .svs, .ndpi, .tif, .mrxs
+ 支持格式：.svs, .ndpi, .tif, .mrxs

Python Environment / Python环

sudo apt-get install openslide-tools
conda create -n pathology python=3.10
conda activate pathology
pip install -r requirements.txt 

How to use / 如何使用

# download the checkpoint and put it in the checkpoints/ folder
export CUDA_VISIBLE_DEVICES='1'

# single file
# python inference.py input.svs --output result.csv

# files under a dir
python inference.py --input /jhcnas3/Pathology/original_data/MIDOG_2022/images --output batch_results.csv

If you find this tool useful, please cite our paper:

@article{ma2024towards,
  title={Towards a generalizable pathology foundation model via unified knowledge distillation},
  author={Ma, Jiabo and Guo, Zhengrui and Zhou, Fengtao and Wang, Yihui and Xu, Yingxue and Cai, Yu and Zhu, Zhengjie and Jin, Cheng and Lin, Yi and Jiang, Xinrui and others},
  journal={arXiv preprint arXiv:2407.18449},
  year={2024}
}

@article{xu2024multimodal,
  title={A multimodal knowledge-enhanced whole-slide pathology foundation model},
  author={Xu, Yingxue and Wang, Yihui and Zhou, Fengtao and Ma, Jiabo and Yang, Shu and Lin, Huangjing and Wang, Xin and Wang, Jiguang and Liang, Li and Han, Anjia and others},
  journal={arXiv preprint arXiv:2407.15362},
  year={2024}
}