Hi tractography experts,
I'm currently attempting to run probabilistic tractography between ROIs associated with the mentalizing network (e.g., VMPFC, DMPFC, TPJ, IFG, ATL, amygdala, and so on). However, across all subjects (N=~100), I'm getting almost all zeroes for waytotal values for certain tractography maps. Some of these tracts include but are not limited to:
AMG-DMPFC
AMG-IFG
DMPFC-TPJ
IFG-PCC
IFG-VMPFC
TPJ-VMPFC
The tractography maps listed above are ones in which there are waytotal values of 0 for >~90% of subjects. The number of zeroes for waytotal varies in degree across all the tractography maps, but the ones listed above suffer from this the most. Interestingly, even though nearly all subjects will have a waytotal of 0, some (for example, fewer than 5) will have seemingly 'normal' waytotals. The corresponding fdt_paths look good to me. This tells me that probtrackx2 is running properly, but there may be something wrong with some subject data. Upon looking at bedpostx's output, however, the images look fine to me. My registrations and transformations look fine as well. I registered each subject's FA image to the FMRIB58 template using FLIRT and FNIRT (via tbss commands), and I inverted these transforms to project standard-space ROIs into subject-specific diffusion space. Upon visual inspection, all of these elements look fine to me.
Here is a sample of what I'm calling:
probtrackx2
--network -x R_AMG_R_DMPFC.txt -l --onewaycondition -c 0.2 -S 2000 --steplength=0.5 -P 5000 --fibthresh=0.01 --distthresh=0.0 --sampvox=0.0 --forcedir --opd -s subj_bedpostX/merged \ -m nodif_brain_mask.nii.gz --dir=probtrack_results
Does anyone have any insight as to why probtrack is failing for some of these tractography maps, for only some subjects?
Thanks so much in advance!
|