writing of extra file t.seg.nrrd about segmentationreview HOT 7 CLOSED

Hi Deepa!
Thanks for your suggestion! I was not able to quickly find a way to directly save nii files. I see the script you attached also uses an intermediate file "-label.nrrd". Do you know if there is a way to directly save niftys?

from segmentationreview.

Hi Anna,

So I believe that slicer.util.saveNode works for both nrrd and nii.gz, so instead of:
filepath = outputPath + "/" + referenceVolumeNode.GetName() + "-label.nrrd"
You can instead do:
filepath = outputPath + "/" + referenceVolumeNode.GetName() + "-label.nii.gz"

from segmentationreview.

Yeah so that does not work, I have tried it awhile ago and just now, here is the log:
[Python] Failed to save node to file
[Python] Error: No writer found to write file .nii.gz of type SegmentationFile.

from segmentationreview.

Is there a functional reason why you convert the nrrd/save the nifti? Why not just save the segmentations as .seg.nrrd? I can think of a few benefits of using .seg.nrrd instead of nifti, such as preserving header/metadata, segmentation color

from segmentationreview.

Good point, the only reason I can think of is the consistency between input and output. Also, nifty also saves headers with metadata (not the color tho)

from segmentationreview.

The inputs can be .nrrd files too (with the new mappings feature). Personally, I think the extension would benefit from saving the segmentations as .seg.nrrd. If nothing else, saves the overhead of re-saving the segmentations as niftis.

from segmentationreview.

i have removed the t.seg.nrrd in the latest commit, saving only nifty files.

from segmentationreview.