[RESOLVED] Dwi_legacy_gpu support for muitlple dwi files in each subject

Dear Jure,

How to run dwi_legacy_gpu command for a dataset with multiple dwi nii files in each subject as shown below?

qunex_pipeline_run1…
sessions
xyz
Diffusion
sub-xyz_ses-01_run-01_dwi.bval
sub-xyz_ses-01_run-01_dwi.bvec
sub-xyz_ses-01_run-01_dwi.json
sub-xyz_ses-01_run-01_dwi.nii.gz
sub-xyz_ses-01_run-02_dwi.bval
sub-xyz_ses-01_run-02_dwi.bvec
sub-xyz_ses-01_run-02_dwi.json
sub-xyz_ses-01_run-02_dwi.nii.gz

I am unable to use the below dwi_legacy_gpu command at subject level as the diffdatasuffix param is a string and is not allowing wildcard like “run.*_dwi” .

Please let me know if I am missing anything or you have any other alternative solution.

qunex dwi_legacy_gpu \
    --sessionsfolder='qunex_pipeline_run1/sessions' \
    --sessions='xyx' \
    --pedir='1' \
    --echospacing='0.69' \
    --te='2.46' \
    --unwarpdir='x-' \
    --diffdatasuffix='run-01_dwi' \
    --usefieldmap='yes' \
    --overwrite='yes'

Thanks
Suhas Reddy

Hi Suhas,

In this case, we usually merge the data manually into a single image and then process it, see [RESOLVED] Hcp_diffusion on a diffusion dataset with a single phase encoding direction - #9 by ehui for details.

I will think about whether we could somehow easily integrate this into the command itself.

Jure

In the next version, we will add the option of specifying multiple input images. If multiple images are provided then QuNex will automatically merge the data and continue with processing.

Jure

We are close to the QuNex 0.98.1 release, in this release you can provide a comma separated list for the diffdatasuffix parameter. QuNex will then use the provided suffixes to merge the images behind them into a single one and run dwi_legacy_gpu over the merged image.