Maria - when you look at the DWI volumes one after another is it that the edge of the brain is misalligned or do the images look different?
If it's the former then you're right and the motion correction isn't doing a great job, but if you're looking at the actual brain (esp white matter) then each of those volumes should look very different because they have different diffusion gradients!
It might be useful to post a few screen shots of consecutive volumes that you think are misaligned to see how much of a problem you have. (Optionally overlaid on the original scan to see what efforts eddy has applied already).
Kx
Sent from my iPhone, please excuse any typos or excessive brevity
> On 27 Jan 2016, at 21:02, SUBSCRIBE FSL Anonymous <[log in to unmask]> wrote:
>
> Thank you for your reply. I did initially run eddy being aware that it can correct motion artifacts. Then checked the motion visually with the movie loop. The DWIs seem to still shift. Afterwards, I tried running mcflirt on the b0s and separately on the rest of DWIs. However, the DWIs, do not seem to be coregistered with much improvement.
>
> The only other solution that I can think of is registering each DWI volume to my reference DWI with flirt, having a weight volume that basically tells to ignore the regions that are varying a lot from one volume to another.
>
> Best,
> Maria
|