dwi_noddi_gpu

qx_utilities.processing.dwi.dwi_noddi_gpu(sinfo, options, overwrite=False, thread=0)

dwi_noddi_gpu [... processing options]

noddi [... processing options]

This command executes CUDIMOT's NODDI microstructure modelling. It uses precompiled CUDA (GPU) binaries and therefore requires a CUDA capable GPU to run. Currently supported CUDA version are 10.2, 11.3 and 12. The command can use two different models: Watson and Bingham. The Watson model is used by default.

Warning

To use this command, successful completion of hcp_diffusion or dwi_legacy_gpu is required.

Parameters

--batchfile (str, default ''):

The batch.txt file with all the sessions information.

--sessions (str, default ''):

A list of sessions to process.

--sessionsfolder (str, default '.'):

The path to the study/sessions folder, where the imaging data is supposed to go.

--parsessions (int, default 1):

How many sessions to run in parallel.

--overwrite (str, default 'no'):

Whether to overwrite existing data (yes) or not (no).

--logfolder (str, default ''):

The path to the folder where runlogs and comlogs are to be stored, if other than default.

--noddi_model (str, default 'Watson'):

Whether to use the Watson or the Bingham NODDI model.

--cuda_version (str, default '10.2'):

Which CUDA version to use. Supports 10.2, 11.3 and 12.

Output files

The results of this step will be present in the HCP Diffusion folder:

study
└─ sessions
   ├─ session1
   |  └─ hcp
   |    └─ session1
   |      └─ T1w
   |        └─ Diffusion.NODDI_<model>
   └─ session2
      └─ hcp
        └─ session2
          └─ T1w
            └─ Diffusion.NODDI_<model>

Examples

qunex dwi_noddi_gpu \
    --sessionsfolder="/data/qx_study/sessions" \
    --batchfile="/data/qx_study/processing/batch.txt"

qunex dwi_noddi_gpu \
    --sessionsfolder="/data/qx_study/sessions" \
    --batchfile="/data/qx_study/processing/batch.txt" \
    --noddi_model="Bingham" \
    --cuda_version="12" \
    --overwrite=no \
    --parsessions=2