Sounds ok to me. You may also want to have some sort of a control tract.
Saad
On 12 Dec 2012, at 06:08, YX Li wrote:
> hello all,
> I was wondering what is the best way to analyse the fiber tracking results when running the probabilistic tractography. I have two group (trainings and conrols). Probabilistic tractography were performed. The seed region was defined from the FSLVBM analysis. I want to test if the training affect the FA values in the fiber tracts connecting the seed region. I can't sure if my statistic processing was right as following:
>
> 1. performed probabilistic tractography from seed mask to all other voxels in the brain
>
> 2. tractography in individual subjects was threshold (at least 50 streamlines had passed out of 5000)
>
> 3. individual tracts was then binarized, transformed to standand space, summed across subjects to produce group probability maps
>
> 4.these group probabilistic maps were threshold (present in a minimum of one-third of subjects) and then both group probabilistic maps were combined together as a mask
>
> 5. voxel-by-voxel-based comparison of FA values were performed between groups. The mask produce from step 4 was used as explicit mask.
>
> The validity of the step 4 is my major point want to verify .
>
> Thank you for your help!
>
> YX Li
>
--
Saad Jbabdi
University of Oxford, FMRIB Centre
JR Hospital, Headington, OX3 9DU, UK
(+44)1865-222466 (fax 717)
www.fmrib.ox.ac.uk/~saad
|