Innse
commented
10 months ago We directly used the code of CLAM to process WSI and extracted features with the default settings. Please refer to https://github.com/mahmoodlab/CLAM .
Closed BusyJzy599 closed 10 months ago
Innse
commented
10 months ago We directly used the code of CLAM to process WSI and extracted features with the default settings. Please refer to https://github.com/mahmoodlab/CLAM .
I am using the histolab library to process the downloaded WSI dataset. The otsu threshold was followed to cut out patches containing greater than or equal to 50% of the tissue area (w/o overlap). However, during the course of the experiment it was found that the resultant metrics were not as expected. After checking, I believe it is a problem with my dataset processing code or paradigm, but due to local resource constraints there is no way to make multiple attempts. I hope to borrow the processing code from your work, and I would be more grateful if you can provide me with the feature files that have been extracted! Looking forward to your reply.