Hi Yong – Your 6 directions are on the low end for tractography. I’d recommend using TRACULA with 30 directions or more. TRACULA doesn’t use the tensor model, which you can fit with just 6 directions, but a model that can capture multiple bundles per voxel, so you need a few more directions for that (and it’s the part that takes long).
You can use the wmparc, which is a version of the freesurfer cortical parcellation where the cortical labels have been grown into the white matter. If you do that, you should transform the wmparc from structural space to diffusion space and compute the ROI stats in diffusion space. Check the multimodal integration tutorial for more info on how to do that.
Hope this helps,
a.y
From: <freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of Yong Li <yongli86@gmail.commailto:yongli86@gmail.com> Reply-To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu> Date: Friday, June 16, 2017 at 8:16 AM To: "freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu" <freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu> Subject: [Freesurfer] FA ROI analysis using 6-gradient-dir DWIs
Dear Bruce, Dear Anastasia,
I would like to investigate white matter changes in a group of elderly patients. The acquired DWI data are 6-gradient-direction+1-non-DWI images by a clinic routine (not by me unfortunately). The meaningful analyses may be ROI-based DTI model considering FA, MD etc. measures to perform group comparisons. Subcortical ROIs like thalamus, putamen in aseg as well as white matter bundles like *h.cst_AS etc. in TRACULA are of my research interests. I would like to kindly ask the following questions:
1. To compute FA/MD of thalamus for instance, shall I simply apply individual thalamus mask extracted from aseg to wmparc then output the results in .stats files?
2. TRACULA analysis works with my DWI data (6 gradient-direction) so I could extract masks of the 18 tracts, which serve as ROIs in the FA/MD group analysis. I wonder whether there exists a better alternative to extract fiber bundle masks? The practical issue is that it takes >20 hours to process single patient and I have hundreds and it's accumulating.
3. Is there a new development to enable adding new labels to the TRACULA atlas? I would like to include Corona Radiata, Inferior Fronto-Occipital Fasciculus, Olfactory Tract etc. So far I only found this post to do it manually: http://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg31955.html.
Thank you all very much in advance!
Kind regards,
Yong