You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi!
The point I don't understand is that how UNet supports png.
I have more than one png file belonging to a patient. should I leave them separately, or collect all the files of a patient into nii.gz?
which one true? (one modality OCT - duke dataset)
The text was updated successfully, but these errors were encountered:
@FabianIsensee (thanks in advance, happy new years!!) for the details:
I am using Duke OCT dataset. we have one modality. Actually 61 scans are obtained for each patient. But since there is no significant difference in consecutive images, doctors created a label for every 5 slices. 11 slices are annotated for 10 patients. (so 110 images and 110 labels are shared)
We have separate png slice files for each patient.
Should I convert this to a nii.gz format (not exactly a consecutive CT though?)
or should I let nnUnet treat each slice of each patient as a separate sample
Duke Dataset Reference:
P. P. Srinivasan, L.A. Kim, P.S. Mettu, S.W. Cousins, G.M. Comer, J.A. Izatt, and S. Farsiu,
"Fully automated detection of diabetic macular edema and dry age-related macular degeneration from optical coherence tomography images",BioMedical Optics Express,5(10), pp. 3568-3577, 2014
Hi!
The point I don't understand is that how UNet supports png.
I have more than one png file belonging to a patient. should I leave them separately, or collect all the files of a patient into nii.gz?
which one true? (one modality OCT - duke dataset)
The text was updated successfully, but these errors were encountered: