Last active
July 22, 2021 19:53
-
-
Save jcreinhold/8787667df85839be66355089eb148c43 to your computer and use it in GitHub Desktop.
Example CSV for lesion-predict CLI in tiramisu-brulee
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
subject | flair | pd | t1 | out | |
---|---|---|---|---|---|
pred_subject_1 | /path/to/flair_1.nii.gz | /path/to/pd_1.nii.gz | /path/to/t1_1.nii.gz | /path/to/save/predicted_segmentation_1.nii.gz | |
pred_subject_2 | /path/to/flair_2.nii.gz | /path/to/pd_2.nii.gz | /path/to/t1_2.nii.gz | /path/to/save/predicted_segmentation_2.nii.gz |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
The names in the
subject
column are arbitrary designators the row and do not affect data loading.The non-
label
columns in the example training CSV file are the same ones used here.The
out
column holds the paths to save each predicted segmentation image volume.The order of the columns doesn't matter. The non-
label
/non-out
columns are sorted for input to the network. For example, in this case, the first channel will be a FLAIR image, the second channel will be a PD-w image, and the third channel will be a T1-w image (if you are using a 3D network).