 |
 |
Hi Priti You may find it useful to read the discussions on these topics in the following papers: DK Jones, Imaging Med. (2010) 2(3), 341–355 Jbabdi and Johansen-Berg, Brain Connectivity Volume 1, Number 3, 2011 The best ways to ensure that you are not just looking at noise are: 1. qualitative: look at the trajectories and ensure they follow anatomically plausible routes 2. quantitative: look at reproducibility across subjects Cheers, Saad On 23 Nov 2011, at 22:47, Priti Srinivasan wrote: > Hello FSL and ProbtrackX Experts, > > I'm using the probtrackx tool to get tracts going from PAG to ACC (based on Hadjipavlov 2005 paper in PAIN). I divided my ACC into 10 different masks and ran probtrackx from PAG to each of the smaller masks of ACC (ACC masks were obtained from FreeSurfer Segmentation and PAG masks were hand drawn on each subject). I used 15000 samples to sample the PDF as the tracts are smaller and non-dominant. When I look at the results, my waytotal values range from 0 to 10 (15, 20 etc) . My understanding of the waytotal is the number of stream lines that successfully went from the seed to the target. I'm trying to interpret these results and I'm not able to say for sure if these results are true or just false positives. Especially when you have about 15000 samples sent out from each voxel in the mask (the mask would have about 40-50 voxels), the total no. of streamlines sent out is close to about 600000 streamlines and if only 10-15 survive out of it, then I'm not sure if these connections actually exist. Has anyone come across this situation before? Could anyone throw some light on this? > > Thank you so much for all your help, > Priti > -- Saad Jbabdi University of Oxford, FMRIB Centre JR Hospital, Headington, OX3 9DU, UK (+44)1865-222466 (fax 717) www.fmrib.ox.ac.uk/~saad