For t1 structural data, there is an option in VBM 8 to preprocess longitudinal data- whereby each image is registered to a subject's mean image across time points. I cant find any similar or equivalent option for FSL TBSS. I wonder if anyone has figured out a way to do this?
Have U tried use TRACULA? In the FreeSurfer package?
https://surfer.nmr.mgh.harvard.edu/fswiki/Tracula
Fabricio, thanks that looks like one possible way to do it. Found this "https://ftp.nmr.mgh.harvard.edu/pub/docs/TraculaNov2013/tracula.workshop.v.pdf" which might be relevant
Anyway, i'm trying out a unbiased tensor based longitudinal registration (Article An unbiased longitudinal analysis framework for tracking whi...
) which seems superior to the traditional FA registration approach. There are a few scripts for this method here https://groups.google.com/forum/#!topic/dtitk/ciF3INwnGt8 if anyone's interested
I would not suggest using Tracula as I have found that it often leads to false positives, and it only focuses on a certain number of tracts.
TBSS can definitely conduct longitudinal analyses; just google 'TBSS longitudinal'.
Another option is using MRtrix3 which can conduct not only the conventional FA approach but a much more valid registration approach i.e. registration of FODs.
- Badges
- Science topic
- Similar topics
- Indexes