Hyper-adaC

Hyper-AdaC: Adaptive clustering-based hypergraph representation of whole slide images for survival analysis

Hakim Benkirane (hakim.benkirane@centralesupelec.fr)

Oncostat Team, U1018 Inserm, CESP Laboratory of mathematics and informatics of CentraleSupelec

Hyper-AdaC: Adaptive clustering-based hypergraph representation of whole slide images for survival analysis

Hyper-AdaC: Adaptive clustering-based hypergraph representation of whole slide images for survival analysis, ML4H 2022. [HTML]
Hakim Benkirane, Maria Vakalopoulou, Stergios Christodoulidis, Ingrid-Judith Garberis, Stefan Michiels, Paul-Henry Cournède

@InProceedings{pmlr-v193-benkirane22a,
  title = 	 {Hyper-AdaC: Adaptive clustering-based hypergraph representation of whole slide images for survival analysis},
  author =       {Benkirane, Hakim and Vakalopoulou, Maria and Christodoulidis, Stergios and Garberis, Ingrid-Judith and Michiels, Stefan and Courn{\`e}de, Paul-Henry},
  booktitle = 	 {Proceedings of the 2nd Machine Learning for Health symposium},
  pages = 	 {405--418},
  year = 	 {2022},
  editor = 	 {Parziale, Antonio and Agrawal, Monica and Joshi, Shalmali and Chen, Irene Y. and Tang, Shengpu and Oala, Luis and Subbaswamy, Adarsh},
  volume = 	 {193},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {28 Nov},
  publisher =    {PMLR},
  pdf = 	 {https://proceedings.mlr.press/v193/benkirane22a/benkirane22a.pdf},
  url = 	 {https://proceedings.mlr.press/v193/benkirane22a.html},
  abstract = 	 {The emergence of deep learning in the medical field has popularized the development of models to predict survival outcomes from histopathology images in precision oncology.  Graph-based formalism has opened interesting perspectives for generating informative representations, as they can be context-aware and model local and global topological structures in the tumor’s microenvironment. However, the critical issue in using graph representations lies in their generalizability. They can suffer from overfitting due to their large sizes or high discrepancies between nodes due to random sampling from WSI. In addition, standard graph formulations are limited to pairwise interactions, which can sometimes fail to represent the reality observed in histopathology and hinder the interpretability of those interactions. In this work, we present Hyper-AdaC, an adaptive clustering-based hypergraph representation to model high-order correlations among different regions of the WSIs while being compact enough to help graph neural networks generalize in the case of survival prediction. We evaluate our approach on $5$ different public available cancer datasets. Our method outperforms most state-of-the-art graph-based methods for survival prediction with WSIs, creating a more efficient and robust alternative to other graph representations. Moreover, due to our formulation, attention maps are depicted at different resolutions depending on the tissue characteristics of each WSI. The code is available at: https://github.com/HakimBenkirane/Hyper-adaC.}
}

Introduction

Hyper-adaC is a new method to represent efficiently Whole WSIs in a way that captures high-level interactions between patches while still being compact enough to generalize the GNNs.

Paper Link: Link to the published paper

Downloading TCGA Data

To download whole slide images (formatted as .svs files) and other clinical metadata, please refer to the NIH Genomic Data Commons Data Portal and the cBioPortal.

Running Experiments

Experiments can be executed through the script main.py, the basic usage to run a tumor type classification on the Pancancer dataset is as follows:

python main.py --cohorts TCGA-BRCA

License

This source code is licensed under the MIT license.

Cite us

If you find our work useful in your research, please consider citing our paper at:

@InProceedings{pmlr-v193-benkirane22a,
  title = 	 {Hyper-AdaC: Adaptive clustering-based hypergraph representation of whole slide images for survival analysis},
  author =       {Benkirane, Hakim and Vakalopoulou, Maria and Christodoulidis, Stergios and Garberis, Ingrid-Judith and Michiels, Stefan and Courn{\`e}de, Paul-Henry},
  booktitle = 	 {Proceedings of the 2nd Machine Learning for Health symposium},
  pages = 	 {405--418},
  year = 	 {2022},
  editor = 	 {Parziale, Antonio and Agrawal, Monica and Joshi, Shalmali and Chen, Irene Y. and Tang, Shengpu and Oala, Luis and Subbaswamy, Adarsh},
  volume = 	 {193},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {28 Nov},
  publisher =    {PMLR},
}