You must be logged in to post Login Register


Lost Your Password?
Search Forums:


 






Wildcard Usage:
*    matches any number of characters
%    matches exactly one character

Forum TrackVis Usage Tracks through Corpus Callosum look wrong

Tracks through Corpus Callosum look wrong
Page:   1  
UserPost

7:03 pm
October 6, 2011


johnvm

Member

posts 3
Print this Post
1

I have data from a Siemens Tim Trio that was acquired using an 80 direction gradient table. The FA map looks fine but the tracks through the CC curve down instead of up. I've tried all combinations of the Invert and Swap options to no avail. Any help would be appreciated. Love the software!

Droopy Tracks

 

11:14 am
October 11, 2011


Ruopeng

Admin

posts 406
Print this Post
2

It's a little hard to tell from this image. Would it be possible to show a trackgroup with only a Coronal slice filter (no ROI sphere)?

Ruopeng

12:55 pm
October 11, 2011


johnvm

Member

posts 3
Print this Post
3

Ruopeng said:

It's a little hard to tell from this image. Would it be possible to show a trackgroup with only a Coronal slice filter (no ROI sphere)?

Ruopeng

Thanks for looking into this. Attahed is another picture without the sphere and just the coronal section of the FA map slightly behind the tracks. As you can see the tracks initially start curving upward but then make a radical downward turn instead of continuing up. I've also tried decrease the maximum angle but it still happens. In DTIStudio, I would need to use the option to "Flip the X-Component of the Eigen Vector". I've tried all the combinations of Inversion and Swapping in your software but none of them seem to be able to fix this problem.

Thanks again!

-John 

 

2:39 pm
October 11, 2011


Ruopeng

Admin

posts 406
Print this Post
4

Post edited 6:45 pm – October 11, 2011 by Ruopeng

Sorry, I meant to say use a Sagittal slice filter. What you can do is add a new trackgroup, and check x-filter (also uncheck y-filter if there is one applied) in Slice filters settings. That way you can see the mass corpus Callosum tracks instead of only this little bundle. It's easier to get a feel what's wrong when look at the "big view". (see picture below)

Anyway, even in a good data, there will be scenerio that CC bundle curves down instead of up. Because that's the nature of DTI. There's only one princple eigenvector in each voxel. So in fiber crossing area, you will see tracks deviated by other major bundle (such as cortico spinal bundle in this case).

Ruopeng

4:17 pm
October 11, 2011


johnvm

Member

posts 3
Print this Post
5

OK. Here's the latest pic. They do curve up in the anterior portion. I'll check how this subject's data looks in DTI Studio. Otherwise, I guess it's working.

 

Thanks!

-John

 

11:24 am
October 12, 2011


Ruopeng

Admin

posts 406
Print this Post
6

Yes, it appears to be right.

Ruopeng
Page:   1