[Mristudio-users] DtiStudio and multiple acquisitions

[susumu]

DtiStudio and multiple acquisitionsHi Jeff,

Let me try to answer your question in a generic way. As you know, we need at least 1 b0 and 6 DWIs for DTI. If you have these 7 images, obtaining the tensor elements requires to "solve" the tensor equation.

If you have more than 6 DWIs, then you have to do "fitting" to obtain the elements. This is very similar to the simple linear fitting. If you have two points, you solve the equation to get the slope. If you have 3 points, you do fitting for the best fit.

When you have more than 6 DWIs, you have two choices; more orientations or more repetitions. In the analogy of the linear fitting, if you have 4 points, you can spread the 4 points along the x axis, or you can sample twice at two locations along the x axis. Statistically, there is not much difference between these two approaches. 

I guess my point is, the two datapoints at a fixed x value (or two DWIs with the same b-orientation) doesn't have to be averaged or need any special calculation. These are just dataponts to be fitted. For the tensor fitting, we can use the same math regardless of multiple points spread over different gradient orientations or repeated with the same orientation set.

Hope this answers your question.

Susumu


  ----- Original Message ----- 
  From: Riley, Jeffrey 
  To: mristudio-users at mristudio.org 
  Sent: Monday, April 21, 2008 6:27 PM
  Subject: [Mristudio-users] DtiStudio and multiple acquisitions




  Hi -

  I have a question regarding multiple dti acquisitions in DtiStudio; namely, how does the program incorporate both acquisitions into the tensor calculations (i.e. are the two runs averaged, and then the tensor calculation is done?) and at what part of the processing pipeline does this occur?

  Thanks for your help,

  Jeff 



------------------------------------------------------------------------------


  _______________________________________________
  Mristudio-users mailing list
  Mristudio-users at mristudio.org
  http://lists.mristudio.org/mailman/listinfo/mristudio-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.mristudio.org/pipermail/mristudio-users/attachments/20080421/b1b6dcc8/attachment.html