hcp_apply_auto_reclean

qx_utilities.hcp.process_hcp.hcp_apply_auto_reclean(sinfo, options, overwrite=False, thread=0)

hcp_apply_auto_reclean [... processing options]

Runs the ApplyAutoRecleanPipeline step of HCP Pipeline (ApplyAutoRecleanPipeline.sh).

Warning

The code expects the input images to be named and present in the QuNex folder structure. The function will look into folder:

<session id>/hcp/<session id>

for files:

MNINonLinear/Results/<boldname>/<boldname>.nii.gz

Parameters

--batchfile (str, default ''):

The batch.txt file with all the sessions information.

--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.

--parelements (int, default 1):

How many elements (e.g. bolds) to run in parallel.

--hcp_suffix (str, default ''):

Specifies a suffix to the session id if multiple variants are run, empty otherwise.

--logfolder (str, default ''):

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

--hcp_icafix_bolds (str, default ''):

Specify a list of bolds for ICAFix. You should specify how to group/concatenate bolds together along with bolds, e.g. "<group1>:<boldname1>,<boldname2>| <group2>:<boldname3>,<boldname4>", in this case multi-run HCP ICAFix will be executed, which is the default. Instead of full bold names, you can also use bold tags from the batch file. If this parameter is not provided ICAFix will bundle all bolds together and execute multi-run HCP ICAFix, the concatenated file will be named fMRI_CONCAT_ALL. Alternatively, you can specify a comma separated list of bolds without groups, e.g. "<boldname1>,<boldname2>", in this case single-run HCP ICAFix will be executed over specified bolds. This is a legacy option and not recommended.

--hcp_icafix_highpass (int, default 0):

Value for the highpass filter, [0] for multi-run HCP ICAFix and [2000] for single-run HCP ICAFix.

--hcp_bold_res (str, default '2'):

Resolution of data.

--hcp_lowresmesh (int, default 32):

Mesh resolution.

--hcp_grayordinatesres (int, default 2):

The size of voxels for the subcortical and cerebellar data in grayordinate space in mm.

--hcp_bold_smoothFWHM (int, default 2):

Smoothing FWHM that matches what was used in the fMRISurface pipeline.

--hcp_autoreclean_model_folder (str, default '<$HCPPIPEDIR/ICAFIX/rclean_models>'):

The folder path of the trained models. Will use the HCP's model folder by default.

--hcp_autoreclean_model_to_use (str, default 'MLP,RandomForest'):

A comma separeted list of models to use. HCP available models are: MLP, RandomForest, Xgboost and XgboostEnsemble. Will use MLP and RandomForest by default.

--hcp_autoreclean_vote_threshold (int):

A decision threshold for determing reclassifications, should be less than to equal to the number of models to use.

--hcp_matlab_mode (str, default default detailed below):

Specifies the Matlab version, can be 'interpreted', 'compiled' or 'octave'. Inside the container 'compiled' will be used, outside 'interpreted' is the default.

Output files

The results of this step will be generated and populated in the MNINonLinear folder inside the same sessions's root hcp folder.

Notes

hcp_apply_auto_reclean parameter mapping:

QuNex parameter	HCPpipelines parameter
hcp_icafix_bolds	fmri-names
hcp_icafix_bolds	mrfix-concat-name
hcp_icafix_highpass	bandpass
hcp_bold_res	fmri-resolution
hcp_lowresmesh	low-res-mesh
hcp_grayordinatesres	grayordinatesres
hcp_bold_smoothFWHM	smoothingFWHM
hcp_autoreclean_model_folder	model-folder
hcp_autoreclean_model_to_use	model-to-use
hcp_autoreclean_vote_threshold	vote-threshold
hcp_matlab_mode	matlabrunmode

Examples

qunex hcp_apply_auto_reclean \
    --batchfile=processing/batch.txt \
    --sessionsfolder=sessions

qunex hcp_apply_auto_reclean \
    --batchfile=processing/batch.txt \
    --sessionsfolder=sessions \
    --hcp_icafix_bolds="GROUP_1:BOLD_1,BOLD_2|GROUP_2:BOLD_3,BOLD_4" \
    --hcp_matlab_mode="interpreted"