I think you need to have another look at the FSL probtrackx2 instructions. You
need to be producing a GIFTI file that contains both surface coordinates and
topology and binary labels. FSL provides tools to generate this file (as it is
a non-standard GIFTI file).
I would not recommend using the volume files in place of surface files as they
will have some biases in streamline counts (being more likely to catch fibers
that are passing by, rather than terminating). You will want to use white
surfaces from ${StudyFolder}/${Subject}/T1w/fsaverage_LR32k folder (and pial
surfaces for stopping). If you do that, you would not need to use
transformation matrices as all of the data will be in the subject’s physical
space (T1w folder).
Peace,
Matt.
From:
<hcp-users-boun...@humanconnectome.org<mailto:hcp-users-boun...@humanconnectome.org>>
on behalf of Xinyang Liu <xinyang_ie...@163.com<mailto:xinyang_ie...@163.com>>
Date: Thursday, March 22, 2018 at 3:04 AM
To: HCP 讨论组
<hcp-users@humanconnectome.org<mailto:hcp-users@humanconnectome.org>>
Subject: [HCP-Users] problems about fibre tracking between surface ROIs
Dear HCP experts,
Hi. We just started to learn tractography between functional ROIs and met some
problems. Our purpose is to do fibre-tracing between face-related functional
ROIs. We got the surface ROIs (.func.gii) with HCP fMRI data and Glasser MMP
atlas. However, in our initial try of fibre tracing, there was no streamline
coming out from the seed ROI. Our seed and target ROIs were separately the FFA
and OFA surface regions in the right hemisphere.
Our command was:
probtrackx2 -x /.../100307_thickness_analysis_ROI_FFA/R.2stat.func.gii
--seedref=/.../100307/T1w/Diffusion/B0_brain.nii.gz -l -c 0.2 -S 2000
--steplength=0.5 -P 5000 --meshspace=caret --forcedir --opd -s
/.../100307/T1w/Diffusion.bedpostX/merged -m
/.../100307/T1w/Diffusion.bedpostX/nodif_brain_mask
--dir=/.../face_network_tracking/output/output1
--waypoints=/.../100307_thickness_analysis_ROI_OFA_R/R.2stat.func.gii -o
FFAtoOFA
The output waytotal is 0 and the fdt_path image is empty.
My questions are as follows:
1. Why there was no streamline coming out from the seed surface ROI?
2. Is it better that we change the surface ROI to volume ROI? If so, how to
acquire the cortical functional volume ROI with HCP data?
3. The FSL says that it now accepts surface data, does it mean we don't need to
do transfer between surface and diffusion space and then set parameters like
--xmf, --meshspace and --seedref?
Any guidance would be very appreciated.
Best regards,
Xinyang Liu
_______________________________________________
HCP-Users mailing list
HCP-Users@humanconnectome.org<mailto:HCP-Users@humanconnectome.org>
http://lists.humanconnectome.org/mailman/listinfo/hcp-users
_______________________________________________
HCP-Users mailing list
HCP-Users@humanconnectome.org
http://lists.humanconnectome.org/mailman/listinfo/hcp-users
