Hi,
If you use the tool "avscale" then you can get the transformation parameters that correspond to this matrix and then use these. Alternatively, if all you want is an rms displacement, then use "rmsdiff" with this matrix (compare it to the identity matrix in $FSLDIR/etc/flirtsch/ident.mat) and this will give you a simple mm result.
I hope this helps.
All the best,
Mark
> On 4 Oct 2017, at 17:26, Rodrigo Perea <[log in to unmask]> wrote:
>
> Hi FSL experts,
> I am trying to understand a strategy to calculate head motion after applying eddy in 4 sets of DWIs. Before, I combined all these images and applied eddy but now I am decided to do them separately (eddy on each set and the flirt each dataset to the fist collected set).
>
> So after running eddy (on each set of DWIs) I get a "my_eddy_output.eddy_restricted_movement_rms" output file that denotes the movement relative to each image and relative to the first (as explained here: https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/eddy/UsersGuide ). With this information I can calculate the average head motion for each set of DWI. Then I using FLIRT to align all the other sets of DWIs to the first set. When I do this, I output the FSL matrix transformation DWIset1_to_setXX.mat.
>
> Essentially, my question is how can I use the FSL transformation matrix (e.g. DWIset1_to_setXX.mat ) to calculate the head motion between the first set of DWIs and the others?
>
> Any suggestions, comments will be greatly appreciate it.
>
> Thanks in advance,
>
> Rodrigo
|