ROI Analyses in FreeSurfer

Note: This is a short post jotting down my thoughts on FreeSurfer ROI analysis. This will be developed into a longer post later; however, this is a topic I've been looking into for some time, and I want to keep track of my sources. 

Regions of interest (ROIs) are groups of voxels that parameter estimates (or other data) are extracted from. With volumetric data, you specify the x-, y-, and z-coordinates of the origin for your ROI, and then build a sphere around it. (Spheres aren't the only method, but they are common.) You can do this with, for example: AFNI's 3dUndump; SPM's Marsbar toolbox; FSL's fslmaths.

I haven't been able to find out how to do a similar procedure with FreeSurfer. FS is 2D, so the ROI would probably be a circle instead of a sphere. Once your ROI is defined, then you could extract thickness, volume, and other measurements from the ROI.

See this post by Doug Greve about using a vertex as the base for an ROI: https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/QuestionAnswers

Also keep this:

#!/bin/tcsh

setenv SUBJECTS_DIR `pwd`

#Create 5mm sphere ROI with 3dUndump
3dUndump -srad 5 -prefix S2.nii -master MNI_caez*+tlrc.HEAD -orient LPI -xyz ROI_file.txt
#Convert to NIFTI
#3dAFNItoNIFTI S2+tlrc

#View in tkmedit
tkmedit -f MNI_caez_N27.nii -overlay S2.nii -fthresh 0.5

#Register anatomical template to fsaverage (FreeSurfer template)
fslregister --s fsaverage --mov MNI_caez_N27.nii --reg tmp.dat

#View ROI on fsaverage
tkmedit fsaverage T1.mgz -overlay S2.nii -overlay-reg tmp.dat -fthresh 0.5 -surface lh.white -aux-surface rh.white


#Map ROI to fsaverage surface
mri_vol2surf --mov S2.nii \
        --reg tmp.dat \
        --projdist-max 0 1 0.1 \
        --interp nearest \
        --hemi lh \
        --out lh.fsaverage.S2.mgh \
        --noreshape

#Check how well the ROI maps onto the inflated surface
tksurfer fsaverage lh inflated -overlay lh.fsaverage.S2.mgh -fthresh 0.5

If you want to extract data from an ROI label, use the mri_segstats option. For example, let's say that I have all of my subjects stacked together in a single dataset using the mris_preproc option, and I have the right hemisphere volume maps in a single dataset. If my ROI is called ROI.mgh, I would use this command:


mri_segstats --i ../../rh.volume.StudyName.10.mgh --seg ROI.mgh --excludeid 0 --avgwf ROI_results.dat


Which would print the average volume over the ROI into the file ROI_results.dat.

Best,


-Andy

FreeSurfer Workshop, Day #1: Overview and Installation

This post is for the FreeSurfer Workshop given at Haskins Laboratories on Wednesday, December 6th, at 3:00pm. Here you will find links to the programs that were discussed during the workshop. Workshop attendees can also post comments in the Disqus section below.

FreeSurfer Download:

FreeSurfer can be downloaded from this link: https://surfer.nmr.mgh.harvard.edu/fswiki/DownloadAndInstall

Prerequisites for installing parallel:

Apple Xcode: https://itunes.apple.com/us/app/xcode/id497799835?mt=12

Homebrew: https://brew.sh/

Follow the instructions on the Homebrew site for installation, then open a terminal and type "brew install parallel." Then go into your directory containing your anatomical images (in NIFTI format) and type the following:

ls *.nii | parallel --jobs 8 recon-all -s {.} -i {} -qcache -3T -all

You may want to set your SUBJECTS_DIR variable to your current working directory with the following command:

In t-shell: setenv SUBJECTS_DIR `pwd`

In bash: export SUBJETS_DIR=`pwd`

Open Science Grid

Join OSG by going here: https://www.opensciencegrid.org/join/

To deidentify fMRI data, convert your DICOM files with dcm2niix: https://github.com/rordenlab/dcm2niix

MRIs can be anonymized with mri_deface: https://surfer.nmr.mgh.harvard.edu/fswiki/mri_deface

Next, get fsurf, OSG's FreeSurfer command: https://support.opensciencegrid.org/support/solutions/articles/12000008488-setting-up-fsurf-on-your-computer

Downloading the sample data

Anatomical data from openfmri.org can be found here: https://openfmri.org/dataset/ds000174/

Running fsurf

Create a text file called "run_fsurf.sh" and use this text:

#!/bin/tcsh

#This script submits each zipped directory structure to fsurf
#You can check the status of the workflow by typing "fsurf list"
#First create the file "subjList.txt" by going into the ds174 directory containing all of the subject directories, and typing the following:
# ls | grep sub* > subjList.txt

foreach subj (`cat subjList.txt`)
cd $subj/ses-BL/anat
fsurf submit --subject=$subj --input=$subj.zip --defaced --deidentified --version 6.0.0 --freesurfer-options='-all -3T -qcache'
cd ../../..
end

As the file states, first create a textfile called "subjList.txt" by typing this from the terminal:

ls | grep sub* > subjList.txt

Then run the script using the t-shell:

tcsh run_fsurf.sh

You can check on your workflows on the OSG by typing "fsurf list"

FreeSurfer Tutorials on YouTube

This page will be updated with more links and videos; check back soon!