Hi,
The flirt command looks correct, but I agree that your result does not look good.
What happens if you use the example_func as the input? Does it still look bad?
If you still cannot get a good result then upload the relevant images and transformation files (mat files and any warp files) to:
https://oxfile.ox.ac.uk/oxfile/work/extBox?id=68312615463381F4C
All the best,
Mark
On 4 Feb 2014, at 19:36, Agaa B <[log in to unmask]> wrote:
> Hi!
> I used the BBR registration of functional to subject's MPRAGE and 12 DOF linear to MNI and the result. As viewed in FSL output looked correct.
> For another type of analysis, I need to apply the transformations created in the above step to create functional images in the MNI space for each subject.
> I did that by applying the following command:
> /usr/local/fsl/bin/flirt -in smooth_func.nii.gz -ref /usr/local/fsl/data/standard/MNI152_T1_2mm_brain -out ${sub}smooth_func_2mmMNI152.nii.gz -init ${mypwd}${sub}/image/${datatype}/melodic_unwarp.ica/reg/example_func2standard.mat –applyxfm
>
> The attached image shows that the resulting functional image is much inflated compared to the MNI standard brain (in red).
>
> The above transformation was done on smoothed data. I repressed this on unsmoothed data, but the result did not change.
>
> I need the functional brain to fit well to the standard for various reasons, such as using atlases or overlaying the image with tissue priors.
>
> If you have any clue what may be wrong with what I am doing, this would be a great help!
>
> Best,
> Aga
> <Screen Shot 2014-02-04 at 1.33.29 PM.png>
|