User Tools

Site Tools


mirecc:mireccanat

DTI processing

  • convert the raw DTI data into nifti format
  • skull strip the DTI
  • run eddy current correction
  • run dtifit to calculate FA,eigenvalues,etc
    • these steps are handled through the make_dti script
  • run bedpostx on the cluster to estimate crossing fibers
    • running bedpost requires copying the dtifit results to a temporary directory within the MIRECC.03/Analysis/dti_temp folder and running each slice individually through the cluster. If you do not run the slices individually they are run sequentially, which can take 1 week to finish.
    • runBEDPOST_PRE.sh – sets up the bedpostx directory
    • submit_CROSS.sh – submits the individual slices through runBEDPOST_CROSS.sh (this does not need to be run)
    • once all 60+ slices are completed, they need to be combined into one volume and moved back to the individual subject's directory.
    • navigate to the subject's folder (on Selye) and run:
bedpostx_postproc.sh dti
  • copy the dti.bedpostx folder to the subject's folder on Cohen
  • now dti can be normalized and dyads can be masked: post_dti_steps.sh
  • navigate to /data/code and run it as bash -x post_dti_steps.sh SUBJNUM

First, FreeSurfer, Manual Comparison

flowchart showing steps of getting first/freesurfer volumes into the same space

:mirecc:mireccanat:mirecc_free_first_chart1.jpg

mirecc/mireccanat.txt · Last modified: 2014/08/04 16:03 (external edit)