I want to use the HCP Pipeline BIDS app to preprocess T1w, T2w, and BOLD
data collected at the CMRR. The BOLD data were acquired to apply the
"PEpolar" method, whereby one short (TR=1.5s, 10 reps) set of EPI data were
collected with AP phase-encoding, and one long (TR=1.5s, 400 reps) set of
EPI data were collected with PA phase-encoding. Per BIDS specifications and
others' suggestions (e.g.,
https://neurostars.org/t/fmriprep-docker-fieldmap-correction/623 ), I have
stored the short acquisition AP data in the "fmap" directory and the long
acquisition PA data in the "func" directory; the JSON sidecar for the
"fmap" file points to the appropriate BOLD series via the "IntendedFor" key
(see directory contents below).

I know that the HCP Pipeline BIDS app expects a fieldmap in order for the
functional data to be preprocessed. However, based on documentation, I
suspect the BIDS app in its current version does not support when the
"fmap" is a paired reverse-phase encode image? I.e., the BIDS app does not
sense that FSL's TOPUP should be used (rather than FUGUE or whatever
program is currently used for this distortion correction). Has anyone else
encountered this dilemma? Are there plans to add this functionality to the
BIDS app?

Any information would be greatly appreciated. Thank you in advance.


*##bids subject session directory contents:*
*cat ./fmap/sub-7536262_ses-00_acq-rest_dir-AP_run-01_epi.json*

*    "PhaseEncodingDirection": "j-", *
*    "IntendedFor":
"func/sub-7536262_ses-00_task-rest_run-01_bold.nii.gz", *
*    "TotalReadoutTime": 0.0527996, *

Scott J. Burwell, PhD
Postdoctoral Research Fellow
Department of Psychiatry
University of Minnesota, Minneapolis, MN

HCP-Users mailing list

Reply via email to