Dear experts,
I'm trying to run NODDI on my multi-shell data for the first time.
I want to ask if there is an optimal pre- and post-processing pipeline before and after fitting NODDI. Until now I run FSL Eddy and BET for the extraction of the brain mask.
Are there some specific options for eddy or other steps useful for a better DWI correction? Moreover, do I have to use as NODDI input a skull-stripped image or with the scalp?
Also, when providing the bvecs for NODDI, do I have to put the original bvecs or the rotated ones produced by eddy?
Last question: I read about tissue-weighted correction after ROI extraction as a procedure to correct ODI, NDI and other measures for the fraction representing the amount of tissue contained in each voxel. In particular, generate the tissue fraction (TF) map = 1-ISOI; then voxel-wise multiply both the NDI and ODI maps (or FA) by the TF map; finally extract the ROI mean of NDI * TF, ODI * TF and TF and divide mean NDI * TF and mean ODI * TF measures by the mean TF (https://github.com/tdveale/TissueWeightedMean). I was wondering if this could be performed also on voxel-wise data and not for the scalar extracted values, in order to use FA, ODI, (ect...) maps corrected for CSF as a sort of free water correction (so to stop at the second point were ODI (or whatever) map is multiplied by the TF map. Could be feasible or correct?
Many thanks
Sincerely
Beatrice
########################################################################
To unsubscribe from the DIFFUSION list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=DIFFUSION&A=1
This message was issued to members of www.jiscmail.ac.uk/DIFFUSION, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
|