Grad_Mix : GradMix for nuclei segmentation and classification in imbalanced pathology image datasets

by Tan Nhu Nhat Doan, Kyungeun Kim, Boram Song, and Jin Tae Kwak.

Introduction

This repository is for our MICCAI 2022 paper [GradMix for nuclei segmentation and classification in imbalanced pathology image datasets] (https://link.springer.com/chapter/10.1007/978-3-031-16434-7_17).

GradMix is an data augmentation technique that is designed to improve nuclei segmentation classification in imbalanced pathology images. It takes a pair of a major-class nucleus and a rare-class nucleus, creates a customized mixing mask, and combines them using the mixing mask to generate a new rare-class nucleus. As it combines two nuclei, GradMix considers both nuclei and the neighboring environment by using the customized mixing mask. This allows us to generate realistic rare-class nuclei with varying environments.

Grad_mix Example Images

Results

Requirements

• python 3.6.10
• scikit-learn 0.23.1
• scikit-image 0.16.2
• opencv-python 4.1.2.32

Usage

Prerequisite: Dataset images, cell/nuclei instance masks and cell/nuclei centroids

1. Clone the repository and set up the folders in the following structure:

 ├── data 
 |   |── Images (Raw)
 |   |── Labels (Raw) 
 |   |── Grad_mix_Images  (output dir: New Sythesized Images)       
 |   |── Grad_mix_Labels  (output dir: New Sythesized Labels) 
 |   |── Inapinted_Images (output dir: New Sythesized Inpainted Images) 
 ├──

2. Run the jupyter file (grad_mix.ipynb) and new grad_mix images and labels will be stored in the outdirs as mentioned.
3. For Training and Testing of nuclei segmentaion and classification tasks please refer to our other repository. Repository: Sonnet:A self-guided ordinal regression neural network for segmentation and classification of nuclei in large-scale multi-tissue histology images

Citation

If Grad_mix is useful for your research, please consider citing following two papers:

@inproceedings{doan2022gradmix,
  title={GradMix for Nuclei Segmentation and Classification in Imbalanced Pathology Image Datasets},
  author={Doan, Tan Nhu Nhat and Kim, Kyungeun and Song, Boram and Kwak, Jin Tae},
  booktitle={International Conference on Medical Image Computing and Computer-Assisted Intervention},
  pages={171--180},
  year={2022},
  organization={Springer}
}
@article{doan2022sonnet,
  title={SONNET: A self-guided ordinal regression neural network for segmentation and classification of nuclei in large-scale multi-tissue histology images},
  author={Doan, Tan NN and Song, Boram and Vuong, Trinh TL and Kim, Kyungeun and Kwak, Jin T},
  journal={IEEE Journal of Biomedical and Health Informatics},
  volume={26},
  number={7},
  pages={3218--3228},
  year={2022},
  publisher={IEEE}
}