[Mristudio-users] about Tractography

[Jeong Pyo Son]

Hello DTI experts!

I'm Jeongpyo Son.

I have some questions regarding* how to acquire tractography data using
homemade calculated images*.
For tensor calculations, DTI studio cannot calculate my DTI data (rat DTI
data acquired by using Bruker 7T scanner) correctly.
Hangyi gave me some advice about calculating tensor using DTI studio but
unfortunately it didn't help. (Regardless of the results, I sincerely
appreciate your help Hangyi ^^)
So, I made MATLAB M-files and acquired the FA, tensor (Dxx, Dyy, Dzz,
.....) and directional diffusivity maps and I saved all the calculated data
in the nii format.

Now, here are some questions that I have about fiber tracking methods:

1. As far as  I know, fiber tracking is available* by tensor images (Dxx,
Dyy, Dzz, .....).*
*How can I create fiber tracts using tensor images?*
My tensor images are 3D so the number of tensor images is 6(Dxx, Dyy, Dzz,
Dxy, Dxz, Dyz).
In the fiber tracking menu, only 1 image can be chosen in the "Tensor
Image" tab.
Should I convert my Tensor images to 4D? If so, then how should I
arrange the Tensor images?

2. Another method, *using principal vector and FA map.*
I made the FA map using my own M-files but I don't know how to create the
principal vector map.
The calculation method of the M-file are same as following.
----------------------------------------------------------------------------------------------------------------------------------------------------
          for z = 1:dz
            for y = 1:dy
              for x = 1:dx
                y_temp = [log(S1(x,y,z)); log(S2(x,y,z)); log(S3(x,y,z));
log(S4(x,y,z)); log(S5(x,y,z)); log(S6(x,y,z))];
                D = inv(b_matrix)*y_temp;
                Dxx(x,y,z) = D(1,1);
                Dyy(x,y,z) = D(2,1);
                Dzz(x,y,z) = D(3,1);
                Dxy(x,y,z) = D(4,1);
                Dxz(x,y,z) = D(5,1);
                Dyz(x,y,z) = D(6,1);
              end
           end
         end
-------------------------------------------------------------------------------------------------------------------------------------------------------
This M-file is made for calculating the data acquired using 6 diffusion
encoding gradients scheme, and S# means "signal of the original DWI/signals
of the b0 image".
So, y_temp means ln(S/S0).

After the Tensor calculations, the eigen values and eigen vectors are
acquired.
*The eigen vectors are formed as 3x3 matrix and by them, how can I acquire
principal eigen vectors?*
Please let me know the formula for calculating principal eigen vector.



Best Wishes.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.mristudio.org/pipermail/mristudio-users/attachments/20120716/9dd2858e/attachment.html