search

mahmoodlab / Patch-GCN

Context-Aware Survival Prediction using Patch-based Graph Convolutional Networks - MICCAI 2021
http://mahmoodlab.org
GNU General Public License v3.0
123 stars 29 forks source link

main.py #1

Closed szc19990412 closed 3 years ago

szc19990412 commented 3 years ago

Will the code of main.py be announced? And the compressed files in datasets_csv cannot be decompressed. Looking forward to your reply！

Richarizardd commented 3 years ago

Hi @szc19990412 - thank you for interest in our repository! We are still cleaning up the code + data preprocessing, and will let you know when it is ready!

chengzs123 commented 3 years ago

I want to know how your data changed from 1024-dimensional features to graphs? Can this part of the code be made public?

Richarizardd commented 3 years ago

@chengzs123 - this part of the code will be made public. For the time being, you can walk through the Inference Benchmark .ipynb to see how loading graph features would work.

Richarizardd commented 3 years ago

@szc19990412 @szc19990412 please see the most recent update, in which:

RRG29 commented 3 years ago

@Richarizardd Hello，where is the script for graph creation ?

Richarizardd commented 3 years ago

Hi @RRG29 - Thank you for following-up. Script for WSI-Graph Construction is uploaded here via this Jupyter Notebook. Will update README on instructions on how to use it, but should be self-explanatory if you are familiar with the CLAM feature extraction pipeline. Given the h5 files (which contains the x,y-coordinates for each patch), we can do fast K-NN on these coordinates via HNSW.