dangnh0611 / kaggle_rsna_breast_cancer

1st place solution of RSNA Screening Mammography Breast Cancer Detection competition on Kaggle: https://www.kaggle.com/competitions/rsna-breast-cancer-detection
MIT License
79 stars 26 forks source link

About cmmd_cleaned_label.csv #2

Closed jessicachen01 closed 1 year ago

jessicachen01 commented 1 year ago

Hi I am also working on a project using CMMD dataset, the original dataset just provides the label of one side which causing confusion especially for D2 data.

The file structure is like: D2-0032

Could you let me know how can you get the cmmd_cleaned_label.csv? and how do you distinguish the left and right image from the above four file?

Thanks in advance!

dangnh0611 commented 1 year ago

Hi, The laterality (left-right) can be extracted from dicom metadata associated with each dicom file (*.dcm). For D2 subset, the missing side (laterality) usually means that side is normal (non-cancer). For details to get the cmmd_cleaned_label.csv, refer to the function stage1_process_cmmd() defined here

jessicachen01 commented 1 year ago

Thank you! it helps a lot.