Deletions are marked like this. | Additions are marked like this. |
Line 15: | Line 15: |
tkmedit bert norm.mgz -segmentation aseg.mgz $FREESURFER_HOME/FreeSurferColorLUT.txt | tkmedit bert norm.mgz -segmentation mri/aseg.mgz $FREESURFER_HOME/FreeSurferColorLUT.txt |
[wiki:DownloadAndInstall top] | [wiki:SetupConfiguration previous]
Testing your FreeSurfer installation
tkmedit
There is a sample subject, called 'bert', found in the freesurfer/subjects directory, which contains pre-processed data stored in .mgz files (not in COR formatted files, which is a deprecated format).
To view a volume, type:
tkmedit bert orig.mgz
To view the subcortical segmentation results, type:
tkmedit bert norm.mgz -segmentation mri/aseg.mgz $FREESURFER_HOME/FreeSurferColorLUT.txt
As you move the mouse over areas of the brain, the label info is found in the 'Sgmtn label' box.
tksurfer
To view a surface, type:
tksurfer bert rh pial
To increase brightness on that surface, in tksurfer:
View->Configure->Lighting->Brightness
To view cortical parcellation labels, in tksurfer:
File->Label->Import Annotation Browse to file rh.aparc.annot and select OK Redraw View icon is the one with the up/down arrows
recon-all -help
The Freesurfer command:
recon-all --help | less
will print text on the overall functionality of Freesurfer.
recon-all -all
To re-create all volume and surface files, on the command-line type:
recon-all -s bert -all
Note that this operation may take 24 hours to complete. Consult recon-all --help for instructions on how to conduct multi-stage processing (eg. -autorecon1, -autorecon2 and -autorecon3). The -autorecon1 stage completes in less than a half an hour, and is worth running to test your installation, as it exercises most of the libraries. To run, type:
recon-all -s bert -autorecon1
Getting started with your own data
Given a subject having two sets of DICOMS (two runs of structural scans), do the following to setup a subject, where <subjid> is replaced with whatever name or id you want to give to that subject:
cd $SUBJECTS_DIR mksubjdirs <subjid> cd <subjid> cd mri/orig mkdir 002 mkdir 003 cp -r <path_to_dicoms>/95555-2-*.dcm 002/ cp -r <path_to_dicoms>/95555-3-*.dcm 003/ mri_convert -it dicom -ot mgz 002/ 002.mgz mri_convert -it dicom -ot mgz 003/ 003.mgz
Now you will have two .mgz files in mri/orig, and recon-all will automatically take these two structural scans and average/motion-correct them. Note that in your own case, the name of the dicom files will be different (in this example, 95555-2-*.dcm consisted of 128 files).
Questions?
Send email to: freesurfer AT nmr DOT mgh DOT harvard DOT edu
Bug reports: analysis-bugs AT nmr DOT mgh DOT harvard DOT edu