[RESOLVED] ICAFIX+extract_nuisance_signal

Use Questions
1

Hi, Jure

After completing denoising with ICAFIX, I need to first use map_hcp_data, then follow the steps in the link for nuisance signal regression. BOLD task-evoked and resting-state functional connectivity preprocessing — QuNex documentation

However, I noticed that map_hcp_data appears to map BOLD images processed through steps like hcp_volume (corresponding path: MNINonLinear/Results/BOLD_1/BOLD_1.nii.gz), rather than the BOLD images denoised by ICAFIX. This makes me wonder if I executed ICAFIX and nuisance signal regression separately, rather than sequentially?

Best, Olivier

2

Hi Oliver,

You define what you want to map back through the tails parameters:

--hcp_nifti_tail 
--hcp_cifti_tail

HCP pipelines produce many outputs that underwent various processing steps. These steps are usually annotated in the tail or suffix of the file name. You need to set the above parameters so they match what you want. If your ran hcp_icafix and hcp_msmall it needs to be the cleaned up, denoised, msmal tail.

Furthremore, HCP pipelines create some outputs that are not annotated in the session.txt files. For example, all concatenated BOLDs or all concatenated rest BOLDs. To map those, you can use the --additional_bolds parameter**.**

See map_hcp_data — QuNex documentation for additional details.

Best, Jure

3

Hi, Jure

I did indeed overlook the --hcp_nifti_tail and --hcp_cifti_tail parameters earlier. Upon checking the files in the MNINonLinear/Results/1/ folder, I did find the ICAFIX-processed files—1_hp0_clean.nii.gz and 1_Atlas_hp0_clean.dtseries.nii. I believe I should be able to map correctly after filling in the --hcp_nifti_tail and --hcp_cifti_tail parameters.

However, I also have a question: Are the ICAFIX result files in the MNINonLinear/Results/1/ folder copied from the MNINonLinear/Results/fMRI_CONCAT_ALL/ folder?

Best, Olivier

4

MNINonLinear/Results/1/ will contain results for bold 1.

MNINonLinear/Results/fMRI_CONCAT_ALL will contain results for all BOLDs merged together/concatenated in the time dimension.

So, if you have only 1 bold, there will be some files with the same content in both folders, yes. If you have several bolds that is not the case.

Best, Jure

1 Like