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 Diffusion Toolkit Usage number of directions with Q-BALL

number of directions with Q-BALL
Page:   1  
UserPost

3:15 pm
November 17, 2011


camocazi

Member

posts 6
Print this Post
1

Hi diffusion MR experts,

 

I  was hoping to ask a general methods question – I understand that having > 60 directions and > bval=3000 is optimal for running Q-BALL. What will happen if I try to run it on 30 directions and bval=1000? Will the results at least be reasonable, and can it resolve at least some of the crossing-fiber issues?

 

We have a great deal of data that was acquired with 30directions, and we are trying to get our results to be at least somewhat comparable to a HARDI sequence. I understand that spherical decconvolution is another method that may work on less than 60 directions, but I am still wondering what will happen if we try QBI.

 

Thank you very much for any knowledge or advice,

 

David

12:14 pm
November 18, 2011


Ruopeng

Admin

posts 406
Print this Post
2

Hi David,

I don't have a firm answer. You just need to go ahead and try it. And investigate the result by yourself. It is possible that qball recon would simply fail on 30 direction data and generate totally non-sense result, as I have experienced before. In other cases when it didn't fail, the result looks a whole lot like DTI at first sight. But after taking a closer look, there are actually some fiber crossing resolved. I would just suggest you just do a DTI and Qball construction on the same data and compare them yourself.

Ruopeng
Page:   1