Released code for the paper 'Mixing Histopathology Prototypes into Robust Slide-Level Representations for Cancer Subtyping' by Joshua Butke, Noriaki Hashimoto, Ichiro Takeuchi, Hiroaki Miyoshi, Koichi Ohshima, and Jun Sakuma.
Accepted to and presented at the 14th INTERNATIONAL CONFERENCE ON MACHINE LEARNING IN MEDICAL IMAGING (MLMI 2023).
This repo contains the Python code for the formulation of a MLP-Mixer-like architecture for application to the domain of computational pathology.
Please cite our paper, if this work is of use to you or you use the code in your research:
@inproceedings{butke2023mixing,
title={Mixing Histopathology Prototypes into Robust Slide-Level Representations for Cancer Subtyping},
author={Butke, Joshua and Hashimoto, Noriaki and Takeuchi, Ichiro and Miyoshi, Hiroaki and Ohshima, Koichi and Sakuma, Jun},
booktitle={International Workshop on Machine Learning in Medical Imaging},
pages={114--123},
year={2023},
organization={Springer}
}
If you have any questions you can contact me at joshua.butke@riken.jp, however we do not gurantee any support for this software.
This work was supported by JST CREST JPMJCR21D3 and Grant-in-Aid for Scientific Research (A) 23H00483. J.B. was supported by the Gateway Fellowship program of Research School, Ruhr-University Bochum, Bochum, Germany.