Dealing with phase reverse encoding EPI

,
1

Hi everyone,

I’m currently in the process of converting my raw fMRI data to BIDS format and I’ve run into a bit of a snag with how to handle my phase reverse encoding files.

After the subject finishes watching a task during the fMRI scan, I immediately start the EPI phase reverse encoding sequence. I couldn’t find clear guidance in the BIDS documentation on how to incorporate these specific files into the BIDS structure.

I’m using HeuDiConv to convert everything to BIDS, and here is an example of what I usually have for the first run in my .tsv table:

series_description sequence_name
task-backtothefuture_run-1_SBRef epfid2d1_106
task-backtothefuture_run-1 epfid2d1_106
task-backtothefuture_run-1_PErev_SBRef epfid2d1_106
task-backtothefuture_run-1_PErev epfid2d1_106

I’ve found the ‘dir-’ label in the BIDS documentation and I guess something like that would be BIDS valid, wouldn’t it?

sub-01/
...
└── func/
    ├── sub-01_task-moviewatching_run-1_dir-PA_bold.nii.gz
    ├── sub-01_task-moviewatching_run-1_dir-PA_bold.json
    ├── sub-01_task-aftermovie_run-1_dir-AP_bold.nii.gz
    ├── sub-01_task-aftermovie_run-1_dir-AP_bold.json
...

So the general question is:

How should I name and organize the phase reverse encoding files in the BIDS directory? Should I use the same task name as in my actual task and should I use the dir- label?

Any detailed advice or examples from those who have encountered and solved this issue would be greatly appreciated!

Thanks in advance for your help!

2

Hi @egor.levchenko,

Is this reverse phase encoded file a full fMRI late or just a single volume used for distortion correction purposes?

Best,
Steven

3

Hey @Steven!

Thank you for your response! This is just a single volume for distortion correction purposes.

4

Hi @egor.levchenko,

Then it goes into the fmap folder with an epi prefix, and must use either the IntendedFor or B0FieldSource/Identifier method for associating back to your BOLD file.

Best,
Steven

1 Like
5

Hey @Steven,

Thank you! If I understood you correctly, it should look something like this:

...
├── anat
│   ├── sub-07_ses-001_T1w.json
│   └── sub-07_ses-001_T1w.nii.gz
├── fmap
│   ├── sub-07_ses-001_run-001_epi.json
│   ├── sub-07_ses-001_run-001_epi.nii.gz
│   ├── sub-07_ses-001_run-002_epi.json
│   ├── sub-07_ses-001_run-002_epi.nii.gz
│   ├── sub-07_ses-001_run-003_epi.json
│   └── sub-07_ses-001_run-003_epi.nii.gz
├── func
│   ├── sub-07_ses-001_task-backtothefuture_run-001_bold.json
│   ├── sub-07_ses-001_task-backtothefuture_run-001_bold.nii.gz
│   ├── sub-07_ses-001_task-backtothefuture_run-001_events.tsv
│   ├── sub-07_ses-001_task-backtothefuture_run-001_sbref.json
│   ├── sub-07_ses-001_task-backtothefuture_run-001_sbref.nii.gz
│   ├── sub-07_ses-001_task-backtothefuture_run-002_bold.json
│   ├── sub-07_ses-001_task-backtothefuture_run-002_bold.nii.gz
│   ├── sub-07_ses-001_task-backtothefuture_run-002_events.tsv
│   ├── sub-07_ses-001_task-backtothefuture_run-002_sbref.json
│   ├── sub-07_ses-001_task-backtothefuture_run-002_sbref.nii.gz
│   ├── sub-07_ses-001_task-backtothefuture_run-003_bold.json
│   ├── sub-07_ses-001_task-backtothefuture_run-003_bold.nii.gz
│   ├── sub-07_ses-001_task-backtothefuture_run-003_events.tsv
│   ├── sub-07_ses-001_task-backtothefuture_run-003_sbref.json
│   └── sub-07_ses-001_task-backtothefuture_run-003_sbref.nii.gz
└── sub-07_ses-001_scans.tsv
...

where each .json file in the fmap folder has an IntendedFor field. In my case, for example, in sub-07_ses-001_run-001_epi.json I should be able to find:

"IntendedFor": [
    "ses-001/func/sub-07_ses-001_task-backtothefuture_run-001_bold.nii.gz"
]

Should I add dir- and acq- prefixes to the fmap filenames just to make it more informative?

Correct me if I am wrong. Thank you for your help!

Best,
Egor

6

All looks good (but remember to use validator to confirm)! I would add the dir label, and also I personally like to use the acq label to denote which scan it is for, for example acq-fmri or acq-dwi.

1 Like