DTI preparation for TBSS analysis

Before anything else you need to be sure that the data you need (diffusion and T1 images) have all been retrieved from Dougal into /data/blinded and contain no duplicates.

1) LOGIN IN TO ONE OF THE LINUX SERVERS (nanlnx1, nanlnx2,...)

% ssh kxxxxxxx@nanlnx1.iop.kcl.ac.uk

2) ADD THE NBL MODULE (THIS WILL SET UP YOU ENVIRONMENT SO YOU CAN RUN THE PIPELINE COMMANDS)

%  module add /home/k1339392/software/nbl /etc/nblmodule

3) CREATE THE FOLDERS FOR YOUR PROJECT (i.e. DRUM_PROJECT)

% mkdir ~/DRUM_PROJECT

% mkdir ~/DRUM_PROJECT/nandata

% mkdir ~/DRUM_PROJECT/dtidata

% mkdir ~/DRUM_PROJECT/expdtidata

% cd ~/DRUM_PROJECT

% ls

4) GET THE DATA FROM DOUGAL (add new folders with NIFTI data to ./nandata with NIFTI data)

% getnandata /data/blinded/CNSCNSD ./nandata DRUMPILOT01

% ls ./nandata

5) PREPARE DTI DATA FOR PREPROCESSING ( adds *.nii.gz *.bvec *.bval _mask.nii.gz files to ./dtidata). Use -C option if your data was acquired in the old CNS scanner using the 32 directions pulse sequence (i.e. BRCATLAS and MIAMRC projects).

% prepdti -C  ./nandata ./dtidata DRUMPILOT01 DTI_2mm_60B1500_FDA

% ls ./dtidata

6) PREPARE T1 DATA REQUIRED FOR EPI CORRECTION (adds new _T1.nii.gz _T1_mask.nii.gz files to ./dtidata)

% prept  ./nandata ./dtidata DRUMPILOT01 MPRAGE

7) RUN EXPLORE DTI PREPROCESSING SCRIPT (creates new *.mat *_EPI.mat files in ./expdtidata )

SM/EC CORRECTION ONLY

% expdticorr -R  ./dtidata ./expdtidata DRUMPILOT01 _SMEC

% ls ./expdtidata

SM/EC/EPI CORRECTION

% expdticorr -E  -R  ./dtidata ./expdtidata DRUMPILOT01 _EPI

% ls ./expdtidata

The final result will be a new file in ./expdtidata called DRUMPILOT01_[EPI/SMEC].mat. This file can only be opened with ExploreDTI. Using Explore DTI you will be able to generate any kind of diffusion images/maps you need, including of course tractography datasets.

See DTI pipeline document for more detail

Random freesurfer commands and links

Table of surface areas (Desikan atlas)

aparc2statstable --hemi lh --subjects <list subjects> --tablefile lh.aparc.area.table

Table of thickness values (Desikan Atlas)

aparcstats2table --hemi lh --subjects <list subjects> -meas thickness --parc aparc --tablefile  lh.aparc.stats.thickness.table

Table of cortical volumes (Desikan Atlas)

aparcstats2table --subjects bert ernie fred margaret --parc aparc.a2005s --hemi rh --meas volume --tablefile aparc_stats_rh.txt

Table of subcortical volumes

asegstats2table --subjects bert ernie fred --meas volume --tablefile aseg_stats.txt

or create text file with path of all subjects and 

asegstats2table --subjectsfile=subject_to_write --meas volume --tablefile aseg_stats.txt

Change background of figure to white:

convert -fill white -opaque black <tksurfer_image_black_background>

<new_image_with_white_background>

http://ftp.nmr.mgh.harvard.edu/fswiki/FsTutorial/MorphAndRecon

Command line Group Analysis

https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/GroupAnalysis

Smoothing/qcache command

http://freesurfer.net/fswiki/FsTutorial/QdecGroupAnalysis_freeview

Methods citation

https://surfer.nmr.mgh.harvard.edu/fswiki/FreeSurferMethodsCitation

Check surfaces

tkmedit your_subject_name wm.mgz rh.white -aux brain.mgz \

tkmedit your_subject_name wm.mgz lh.white -aux brain.mgz

https://surfer.nmr.mgh.harvard.edu/fswiki/BasicReconstruction

Check subcortical segmentation
tkmedit SUBJECT orig.mgz -aux-segmentation aseg.mgz

Explanation of steps of reconstruction
http://ftp.nmr.mgh.harvard.edu/fswiki/FsTutorial/MorphAndRecon

Recon-all descriptions
https://surfer.nmr.mgh.harvard.edu/fswiki/recon-all