Skip to content

Code for our BVM workshop submission "Attention-based Multiple Instance Learning for Survival Prediction on Lung Cancer Tissue Microarrays"

License

Notifications You must be signed in to change notification settings

DeepMicroscopy/Cox_AMIL

Repository files navigation

Cox-AMIL

Attention-based Multiple Instance Learning for Survival Prediction on Lung Cancer Tissue Microarrays

This repository contains the code that was used for our BVM workshop sumbission Attention-based Multiple Instance Learning for Survival Prediction on Lung Cancer Tissue Microarrays available on ArXiv.

The code is based on the CLAM framework for weakly-supervised classification on whole-slide images. The images were preprocessed as described in the original documentation. The model was trained using main_survival.py. Some evaluations are availabe in tma_notebooks\09_evaluation_Ostercappeln.ipynb.

License

The code is released under the GPLv3 License following the original code base here.

Acknowledgements

J.A. acknowleges funding by the Bavarian Institute for Digital Transformation from the Project ReGInA.

About

Code for our BVM workshop submission "Attention-based Multiple Instance Learning for Survival Prediction on Lung Cancer Tissue Microarrays"

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published