Using motion-corrected, MNI-warped EPI in (for EPINorm-ed data)

AFNI version info (afni -ver):
Precompiled binary linux_openmp_64: Jun 16 2023 (Version AFNI_23.0.01 'Commodus')

I would like to use the EPINorm approach (Calhoun et al. 2017, on a functional dataset that is lacking high-resolution T1ws. (testing with @auto_tlrc and MNI_EPI produced reasonably aligned output).

Since does not directly support EPINorm, I need to perform motion-correction and registration prior. Is there a way to inject 3dvolreg output directly into the pipeline (i.e. for outlier detection and nuisance regressors), or would I need to create a custom processing script to do so?


I am not familiar with EPINorm, thanks for the citation.

So, you currently have EPI data but no T1w for those subjects, and you want to map the EPI to standard space (specifically, an MNI space), and EPINorm provides a nonlinear warp from an EPI reference volume to template space? And you want to have an command for the rest of your processing (EPI motion correction, regression modeling), and inject the EPI->MNI warp, as well?


For reference this is bringing discussion over from this thread on NeuroStars:

In a sense, yes. However, @pmolfese provides an excellent answer clarifying that EPINorm (i.e. subject EPI / fmri --> MNI EPI template) can be used with some of the parameters (e.g. -tlrc_base MNI_EPI+tlrc) need to be tweaked.