Closed brandon-gallas closed 6 years ago
I confirmed that I get the same kind of (bad) result. It’s a bug. We should have the result you are looking for. We will fix this bug soon.
Qi has compiled a new version and shared it with our colleague.
Response from our colleague:
"Tried it a moment ago. Yes the table looks right now. Thank you!"
I am logging in an issue submitted by a colleague via email.
See the snapshot below. In the table “AUCs for each reader and modality”, the values in the column “AUC difference” seem not right to me because they are simply a copy of the column “AUC ffdm_dense”. Is the software designed to be like this, or I didn’t use it correctly? Thanks!