Hey Saad, 

Thanks for the quick reply. I agree I havn't given you much information. 

But because the script that I am using is so long, I'm not sure where things may have gone wrong. I have been checking the data along the way- but I'll definitaly have to comb through it again. 

I put this post up to get some feedback on common errors that happen to others that may result in missing data. Do you have any further suggestions on common error prone sections of the analysis? Having a few starting points may assist me in combing through my data. 

Just to give you more info on my seeds/target. I am using a surface map  produced by freesurfer to define the different gray/white matter regions. The tracts are based on connections between these gray/white matter regions. Do you believe that my lack of directions would affect the efficiency of this type of analysis?

thanks again

On Sat, Apr 13, 2013 at 8:56 AM, Saad Jbabdi <[log in to unmask]> wrote:

Hi Erica

I am afraid it is difficult for us to diagnose your problem given the information that you gave us.

I am assuming that you are running tractography in a bunch of subjects using some sort of automated script. And for each pathway, the script uses different sets of inclusion/exclusion masks to isolate the pathway. Right?  And you are saying that tractography fails to recover some pathways for some subjects (i.e. the corresponding fdt_paths is zero).  In this case, you are going to have to examine those cases individually to figure out what has happened. 

For example, the bvecs may be wrong for a subset of your subjects. Or the inclusion/exclusion masks may not be appropriate for those subjects/pathways, etc. etc.

Hope this helps a bit

Saad

On 12 Apr 2013, at 21:17, Erica Tatham wrote:

Hi FSL experts.

I have run my tractography analysis on the majority of my participants and I noticed that I am missing DTI data (FA, RD, MD) for many of my participant's tracts or I am only getting DTI information for ~40% of the tracts (note: the tracts with DTI information are not consistent across all participants).

I have maintained the standard protocol for all analysis and was wondering if anyone had any insight into why I may be loosing data. I think the problem is that the tracts are not being located, and therefore DTI information can not be calculated for those tracts. Does anyone have any suggestions on the most standard way to change my analysis so that I obtain more information - or should I just work with that I have?

Below is some info on my analysis......

I have 12 directions in my data

The tracts I am looking at:
Superior Longitudinal Fasciculus
Inferior Longitudinal Fasciculus
Cingulum Bundle
Cortico-spinal tract
Inferior Frontal-occipital fasciculus
Uncinate Fasciculus

Thank you, thank you, thank you for any help :)
I really appreciate that there is a wonderful forum like this, and hopefully when I know enough about FSL I can help other people out.

Cheers
Erica

--

Saad Jbabdi

University of Oxford, FMRIB Centre

JR Hospital, Headington, OX3 9DU, UK

(+44)1865-222466  (fax 717)

www.ndcn.ox.ac.uk/team/researchers/saad-jbabdi

--

Erica Tatham, B.Sc

M.Sc. Candidate 

McMaster Integrative Neuroscience Discovery & Study

McMaster University

1280 Main St W., Hamilton On

Email: [log in to unmask]