Boli-trainee / MHFAN

Cell localization and counting: 1) Exponential Distance Transform Maps for Cell Localization; 2) Multi-scale Hypergraph-based Feature Alignment Network for Cell Localization; 3) Lite-UNet: A lightweight and efficient network for cell localization

Cell Localization and Counting

This repository includes three papers about cell:

1. Exponential Distance Transform Maps for Cell Localization; Paper
2. Multi-scale Hypergraph-based Feature Alignment Network for Cell Localization; Paper
3. Lite-UNet: A lightweight and efficient network for cell localization. Paper

Overview

Visualizations

Datasets

• Download all dataset from Google-Drive

Installation

Download MHFAN:

git clone https://github.com/Boli-trainee/MHFAN

Environment

python >=3.6 
pytorch >=1.4
opencv-python >=4.0
scipy >=1.4.0
h5py >=2.10
pillow >=7.0.0
imageio >=1.18
nni >=2.0 (python3 -m pip install --upgrade nni)
and so on

Generate EDT Map (Ground Truth)

cd data
python CoNIC_process.py
# Generate all datasets by this way

Generate image file list: python make_npydata.py

Training and Testing

python train.py --dataset BCData
python test.py

Thanks

This code is based on FIDTM (https://github.com/dk-liang/FIDTM). Many thanks for your code implementation.

Reference

If you find this project is useful for your research, please cite:

@article{li2024multi,
  title={Multi-scale hypergraph-based feature alignment network for cell localization},
  author={Li, Bo and Yong, Zhang and Zhang, Chengyang and Piao, Xinglin and Hu, Yongli and Yin, Baocai},
  journal={Pattern Recognition},
  pages={110260},
  year={2024},
  publisher={Elsevier}
}

@article{li2024exponential,
  title={Exponential distance transform maps for cell localization},
  author={Li, Bo and Chen, Jie and Yi, Hang and Feng, Min and Yang, Yongquan and Zhu, Qikui and Bu, Hong},
  journal={Engineering Applications of Artificial Intelligence},
  volume={132},
  pages={107948},
  year={2024},
  publisher={Elsevier}
}

@article{li2024lite,
  title={Lite-UNet: A lightweight and efficient network for cell localization},
  author={Li, Bo and Zhang, Yong and Ren, Yunhan and Zhang, Chengyang and Yin, Baocai},
  journal={Engineering Applications of Artificial Intelligence},
  volume={129},
  pages={107634},
  year={2024},
  publisher={Elsevier}
}

@article{liang2022focal,
  title={Focal inverse distance transform maps for crowd localization},
  author={Liang, Dingkang and Xu, Wei and Zhu, Yingying and Zhou, Yu},
  journal={IEEE Transactions on Multimedia},
  year={2022},
  publisher={IEEE}
}