Giter Club home page Giter Club logo

nagcn's Introduction

NAGCN: Node-aligned Graph Convolutional Network for Whole-slide Image Representation and Classification (CVPR2022)

Pre-requisites

  • NumPy
  • Pandas
  • Scikit-learn
  • Tqdm
  • PyTorch
  • PyTorch Geometric

Data Preparation

First, download TCGA data. You need to cut whole-slide images (WSIs) into patches (e.g. size $256\times 256$ at $20\times$ magnification) , and then generate the instance-level features using an encoder network. The instance-level features of each WSI are used to generate the WSI bag. Each WSI bag is stored as a .pt file.

After that you need to generate an index file (in .pth format), which stores a python list, and each item of the list is a dict, which corresponds to one WSI information.

[{'slide': 'xxx.pt', 'shape': torch.Size([N, D]), 'target': 'y'}]
  • slide: the file of WSI bag
  • shape: the size of WSI bag
  • target: the label string of WSI

Clustering Sampling

Perform hierarchical global-to-local sampling strategy to build visual codebook and generate the sampled WSI bags.

$ python codebook_generation.py

Graph Construction

Construct WSI graphs using the sampled WSI bags.

$ python graph_construction.py

Experiment on TCGA Dataset

$ CUDA_VISIBLE_DEVICES=0 python main_nsclc.py 

Citation

If you want to cite this work, please use the following BibTeX entry.

@InProceedings{Guan_2022_CVPR,
    author    = {Guan, Yonghang and Zhang, Jun and Tian, Kuan and Yang, Sen and Dong, Pei and Xiang, Jinxi and Yang, Wei and Huang, Junzhou and Zhang, Yuyao and Han, Xiao},
    title     = {Node-Aligned Graph Convolutional Network for Whole-Slide Image Representation and Classification},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month     = {June},
    year      = {2022},
    pages     = {18813-18823}
}

nagcn's People

Contributors

yohnguan avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.