Freesurfer
https://surfer.nmr.mgh.harvard.edu/fswiki/rel7downloads
-See below in install-
## If mamba is not installed ##
conda install --channel=conda-forge --name=base mamba
mamba create --override-channels --channel=conda-forge --name=enigma_meg mne
conda activate enigma_meg
pip install git+https://github.com/nih-megcore/enigma_anonymization_lite
This is a suite of tools for anonymization and bidsification of MEG and structural MRI data. While it was developed for the ENIGMA MEG Working Group, it is useful for anyone wanting to share their data. There is a specific function for each processing pipeline that you may be using. Currently, a fully functional MNE python is implemented, with an additional tool for Brainstorm users. Other functions will be added at user request.
usage: enigma_anonymization_mne.py [-h] [-topdir TOPDIR]
[-csvfile CSVFILE] [-njobs NJOBS]
[-linefreq LINEFREQ] [-bidsonly]
This function takes a csv file containing a list of datasets to convert to BIDS format. A sample file (sample.csv) is distributed with this package and contains the following fields:
subjid: This is the subject ID, will be appended with sub- in the BIDS output
full_mri_path: The full path to the T1 weighted MRI
full_meg_path: The full path to the raw MEG dataset
session: Session, to permit multiple images per participant
trans_fname: The full path to the .fif transform file produced by MNE python
Upon execution, this function places the mri files into a staging directory, then does basic freesurfer processing (not the full segmentation) to obtain a surface of the head and perform defacing. Next, the BIDS structure is created and populated with the MRI and MEG data. Finally, a QA HTML report document is created, so that you can easily view all the resultant MRI images to ensure that the anonymization is adequate and that the coregistration is accurate. The resultant BIDS tree is placed in topdir/bids_out, and the freesurfer subjects directory can be found in topdir/bids_out/derivatives/freesurfer/subjects. The .html QA report will be in topdir/bids_out/derivatives/BIDS_ANON_QA/Coreg_QA_report.html. In addition, individual QA images will be stored in the same QA directory and can be rapidly assessed with the Run_QA.py tool.
optional arguments:
-h, --help show this help message and exit
-topdir TOPDIR The directory for the outputs
-csvfile CSVFILE The name of the CSV file described above
-njobs NJOBS Optional, number of jobs for Freesurfer processing
-linefreq LINEFREQ Optional, powerline frequency, defaults to 60s
-bidsonly Perform bidsification only, with no anonymization
Brainstorm users should use the "Export BIDS dataset [experimental]" process, which will create the BIDS data structure for both the MEG and MRI datasets. Brainstorm will not, however, update the BIDS structure with the location of the MEG fiducials in the space of the anatomical MRI. For that, we have provided and additional tool.
usage: update_json_bstorm.py [-h] [-bids_root BIDS_ROOT] [-bids_id BIDS_ID]
[-bst_id BST_ID] [-bst_datapath BST_DATAPATH]
This function will take the location of the BIDS tree produced by Brainstorm, the BIDS ID in the bids_root (which should have the form sub-SUBJID), the Brainstorm subject ID, and the datapath to the Brainstorm Protocol data folder.
usage: Run_QA.py [-h] [-bids_root BIDS_ROOT] [-rows ROWS]
[-columns COLUMNS] [-imgsize IMGSIZE]
This function runs an easy to use QA browser to assess .png images returned from any of the process_anonimization_X.py routines. By clicking on each image, users can toggle through states including "GOOD" (green outline), "BAD" (red outline) and "Unchecked" (grey outline). A "Save" button will save this infomation in a log file. You can change the size of the matrix of images displayed, as well as the size of each image. This routine can be run with no arguments, in which case the default bids_root is bids_out, 4 rows, 2 columns, and an image size of 400. The log file will be stored in:
bids_out/derivatives/BIDS_ANON_QA/Coreg_QA_logfile.txt
The logfile will state if an image has been marked "GOOD", "BAD", or unchecked. If you would like this logfile in an easier to use CSV format, you can run an additional tool:
usage: enigma_anon_parseQAlogs.py [-h] [-bids_root BIDS_ROOT]
This commmand will produce a file:
bids_out/derivatives/BIDS_ANON_QA/Coreg_QA_summary.csv
This file can easily be opened with any spreadsheet program.