Hi - are you sure that your gradient orientations are in the coronal  
frame of reference.
Often, you will find that gradient orientations are in the scanner  
frame of reference.

If this is true, then you might find everything works fine when you  
have reoriented your
images into axial sections.

If the gradient orientations are really in the coronal frame of  
reference, then you will have to do exactly the same manipulations to  
them as you do to your images to get them into axial.

i.e.
The following might work to make coronal sections into axial.
avwswapdim data x z y data

This will swap z and y - you might need to introduce some negative  
signs to make things the right way up.

Anyway, assuming this works, you will then have to swap the y and z  
elements in your bvecs file (again with the appropriate negative signs).

Be very careful to check dti_V1 in lines mode in fslview to make sure  
the vectors line up in all planes of section.

Hope this helps

T


On 4 Jan 2007, at 19:40, bs Jeong wrote:

> Dear FSLer
>
> I have 52 directional DTI data which were acquired with coronal  
> direction
> (posterior to anterior) and also 52 vector values of images  
> coronally acquired.
> If I have DTI data which were acquired with axial direction,  
> FSLview shows
> axially sectioned image at left inferior quater of fslview window.
> But FSLview shows coronally sectioned image at left inferior quater of
> fslview window because my DTI images were acquired with coronal  
> direction.
>
> My question is when I have to convert 4d DTI data coronally  
> acquired to the
> same orientation with avg152T1 (LAS system).
> One possible method would be that I reorient 4d DTI images BEFORE  
> putting
> them into DTIfit. In case, do I have to change 52 vector values of  
> images
> coronally acquired and is it possible?
>
> Another method would be that I reorient the products (FA images, L1..,
> ADC...) of 4d DTI images AFTER putting them into DTIfit. In case,  
> can I use
> 52 vector values of images coronally to get FA, ADC, L1... from 4d  
> DTI image
> acquired coronally. Also I should have to B0 and brain mask with  
> direction
> before reorienting.
>
> Thanks in advance