-
Notifications
You must be signed in to change notification settings - Fork 27
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
Help wanted: SyN registration of anatomical T1 to diffusion b0 image fails after printing 'Negative spacing is not supported and may result in undefined behavior' #296
Comments
What are the orientations of your input images, including the template? We very likely assume that everything but the EPI image is in RAS orientation. the "Negative spacing is not supported" suggests to me that you probably have an LAS image somewhere. |
Hi @effigies ! Thanks for your comment, this is likely the issue. MRTrix Thanks again and best regards, |
Because BOLD EPIs are often very long, we prefer not to reorient them unless we must, so we have been intentionally more lenient there. I would expect it to be fine to have your EPI in LAS or RAS, but please let us know if that's incorrect. |
Hi again ! Changing just the anatomical scan to RAS didn't solve the issue, but converting both the EPI and anatomical images allowed the workflow to run without errors. Haven't checked the output yet, but at least the error is gone, thanks for your comment ! |
Commenting from the peanut gallery, I will just mention that we have had much more luck registering the dMRI anisotropic power map to T1-weighted templates. There's an implementation of APM in DIPY. |
Thanks for the comment and the reference. Is the full abstract available somewhere ? I can only see a low-resolution, cropped snapshot of Figure 1, which seems to mention registration of T1 to AP. |
This is another relevant reference: https://www.biorxiv.org/content/10.1101/864108v2 |
Hi !
I'm trying to register anatomical T1 to diffusion b0 image using sdcflows version 2.1.1, with the Syn option as I don't have any fieldmap nor b0 with opposite phase encoding.
I affine registered the
mni_icbm152_t1_tal_nlin_asym_09c.nii
template, which I think corresponds to theMNI152NLin2009cAsym
template used by sdcflows, to my anatomical T1w scan using AntsPy. The output 'fwdtransforms' is used as std2anat_xfm in sdcflows. I created anatomical and b0 brain masks using FSL bet with the -m option. I took the epi metadata from the original dwi.json file, as the b0 was extracted from the dwi.nii.gz file.I have the following code, where I create a
init_syn_preprocessing_wf()
andinit_syn_sdc_wf()
, linked together via aestimate_sdc
workflow:You can have a look at the terminal log below. The log ends after a warning from ITK saying that 'Negative spacing is not supported and may result in undefined behavior', and 'suspicious return value (2) from SVDC'. There's also a stderr printed during the
epi2anat
node, saying that 'file NULL does not exist '.Is there an error that strikes you in the code above ? Am I missing something obvious ? Might not be the best way to use sdcflows, any other suggestion would be welcome.
Thanks in advance and best regards !
David
Terminal log can be found here
The text was updated successfully, but these errors were encountered: