-
Notifications
You must be signed in to change notification settings - Fork 5
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
outputPixelSpacing of resampled volume is hard coded at 1x1x1 #2
Comments
The reason it was hard-coded is that the original use case was an MRI-TRUS registration, and the module that SegmentRegistration uses internally poses very concrete constraints about the inputs that if not met, then the registration is unsuccessful. If you open SegmentRegistration.py in the installed (or built) extension and change the output pixel spacing then can you achieve a good registration? |
My initial thought would be to dynamically change the spacing so that the number of voxel in the resample volume is similar, regardless of the size of the volume. I haven't looked at the inner works of the other module, so not sure if this would work. |
This might work. My main concern is that we have ongoing clinical projects using these modules, and changing parameter initialization may considerably change the output on those types of inputs. However I think this is worth a try. |
I'm happy to contribute and can look into updating the code and making a PR. |
Awesome! Looking forward to it! |
Me again!
The harden transform bug was fixed in the latest nightly 3D Slicer build, but I've found that I can't run this module with my data as the 'outputPixelSpacing' for the resampled volume is hard coded to 1x1x1. The volumes I'm using a whole body PET/CT, so 1x1x1 mm is perhaps a bit small and it takes 10 minutes to resample the volume on my computer.
Is there a reason this spacing was chosen? Should the spacing be the same as the moving volume?
The text was updated successfully, but these errors were encountered: