search

bianhao123 / Mixed_supervision

MICCAI2022: Multiple Instance Learning with Mixed Supervision in Gleason Grading.
GNU General Public License v3.0
53 stars 13 forks source link

Some questions about data processing #1

Open Springfield233 opened 1 year ago

Springfield233 commented 1 year ago

Good work! I would like to ask the author that the data you provide is only. bin files. Please tell me how to do data preprocessing and could you provide the corresponding code？

bianhao123 commented 1 year ago

Thanks! Our data preprocessing pipeline mainly follows the https://github.com/histocartography/seg-gini/blob/main/bin/create_sicap_data.py to create the whole data. And then follow the https://github.com/histocartography/seg-gini/blob/main/bin/preprocess.py to extract the superpixel region feature. The data preprocessing code of our framework version will be released soon.

Springfield233 commented 1 year ago

Thanks. I'll try it first.

wxyz-wjx commented 10 months ago

谢谢。我先尝试一下。 请问你解决了数据预处理以及可视化的操作了吗

zhanghaoyue commented 6 months ago

I used the code to extracted patches and the node number does not match yours. It seems you have 5x more nodes than if use segGini's approach...

SuooL commented 1 week ago

The code at https://github.com/histocartography/seg-gini/blob/1c90f832d7c872f8d9eafa72aa6afcdab7808b7f/bin/create_sicap_data.py#L76 can not run successfully. All the mask files were in JPG format, not PNG format.

Can you provide the dataset with PNG masks?