Sandbox related extensions installing and running. Could not find Stitcher and could not run total segmentator

Salamo alekom (hello everybody) I am trying to use sandbox extension; I installed SANDBOX on (ver5.2.1). I searched (CTRL+F) for stitcher. I did not find it. Also, when trying to run Total segmentator. it prompts me to install PYtorch( I instilled and updated it).

I can confirm that after installing the Sandbox extension in Slicer-5.2.1, the Stitch Volumes module is available:

You can try reinstalling Slicer from scratch and if you still have problems then upload the application log somewhere and post the link here.

Please delete your entire Slicer installation folder and follow the TotalSegmentator setup instructions word-by-word. If you find any difference in what you see and what you expect, stop immediately, take a screenshot and copy the logs and create a new topic on this forum to ask for help.

1 Like

Thanks a lot for your reply, it WORKED!!. In total segementator i understand there is a fast mode, What the difference? would it affect the statistics(quantification as volume, area, and size) driven from the segment ?

Thanks in advance

1 Like

Fast mode uses 3mm cubes for segmenting objects (instead of 1.5mm of the full-resolution segmentation). Probably size differences are negligible for most use cases, but you need to test if it makes a difference for you. You can run the full-resolution segmentation on CPU, you just need to wait more for the result (about 40 minutes).

1 Like

I have CUDA 11.7 but i do not think it is running
I enabled it through NEVDA control panel, but still

I am not sure if running on GPU

You can paste

import torch

into 3D Slicers Python console to check this.

1 Like

I was trying it again it comes back with this error msg
Traceback (most recent call last):
File “C:\ProgramData\NA-MIC\Slicer 5.2.1\bin\Python\slicer\”, line 2961, in tryWithErrorDisplay
File “C:/ProgramData/NA-MIC/Slicer 5.2.1/NA-MIC/Extensions-31317/TotalSegmentator/lib/Slicer-5.2/qt-scripted-modules/”, line 264, in onApplyButton
self.logic.process(self.ui.inputVolumeSelector.currentNode(), self.ui.outputSegmentationSelector.currentNode(),
File “C:/ProgramData/NA-MIC/Slicer 5.2.1/NA-MIC/Extensions-31317/TotalSegmentator/lib/Slicer-5.2/qt-scripted-modules/”, line 749, in process
File “C:/ProgramData/NA-MIC/Slicer 5.2.1/NA-MIC/Extensions-31317/TotalSegmentator/lib/Slicer-5.2/qt-scripted-modules/”, line 656, in logProcessOutput
raise CalledProcessError(retcode, proc.args, output=proc.stdout, stderr=proc.stderr)
subprocess.CalledProcessError: Command ‘[‘C:/ProgramData/NA-MIC/Slicer 5.2.1/bin/…/bin\PythonSlicer.EXE’, ‘C:\ProgramData\NA-MIC\Slicer 5.2.1\lib\Python\Scripts\TotalSegmentator’, ‘-i’, ‘C:/Users/M2/AppData/Local/Temp/Slicer/__SlicerTemp__2023-02-13_11+22+31.867/total-segmentator-input.nii’, ‘-o’, ‘C:/Users/M278837/AppData/Local/Temp/Slicer/__SlicerTemp__2023-02-13_11+22+31.867/segmentation’, ‘–ml’, ‘–task’, ‘total’, ‘–fast’]’ returned non-zero exit status 120.

came back true. Still slow

If you don’t have a powerful GPU or CUDA version is not supported by pytorch then you have to force using the CPU as described here.

If you have Intel Integrated Graphics + discrete NVIDIA GPU in your system then make sure that in NVIDIA settings you set SlicerApp-real.exe to use the NVIDIA GPU.

1 Like

I checked the box with fast option and I had this previous mentioned error message