fabianheinemann / deep_learning_for_liver_nas_and_fibrosis_scoring Goto Github PK

Dear Author,

I'm trying to predict pathologist-like NAS score on slide level. I would appreciate it if you could share the link to download the original whole slide image used in the paper (pathologist_scores.csv and pathologist_scores_test.csv).
Should I provide any additional information for the data request?

@FabianHeinemann

As mentioned in the paper, ' Liver sections were assigned to discrete scores quantifying the histopathological features of ballooning (0–2), inflammation (0–3) '.
But the dataset only contain ballooning (0-1, ignore), inflammation (0–2, ignore) . It is not consistent.

Could you please further explain how each file is named? What does the number in the file name mean？

1. for the image tiles in ballooning folder, can we know whether it has other features (fibrosis, inflammation, steatosis) based on the name?
2. can we group the tiles to get the whole slide image based on the file name?

Thanks. @FabianHeinemann