ANTS and SPM compatibility

[zeynabre]

Hello,
I have an mr image of the iterative averaging with ANTS. When I try to import it in SPM I get this error:
"There is not enough overlap in the images
to obtain a solution.

    Please check that your header information is OK.
    The Check Reg utility will show you the initial
    alignment between the images, which must be
    within about 4cm and about 15 degrees in order
    for SPM to find the optimal solution."

Do you have any solution how can I have a ANTS average image which is compatible with SPM?

Thanks

[ntustison]

There are known header issues affecting compatibility between different software packages. See here. We don't really have anything to harmonize these issues but you might look at something like nipype which does deal with such matters.

[satra]

@zeynabre - if you affine transform the average image to MNI space that should work fine with SPM

[zeynabre]

You mean for example using FLIRT?

[satra]

FLIRT or ANTs - doesn't matter

[zeynabre]

I do not have raw data. I only have average images from ANTs. And if I run FLIRT for this image it works in SPM.

[cookpa]

Hi @zeynabre, would you mind sharing the template image before and after you ran FLIRT? I'd like to look at the headers so I can update the ANTs Wiki for other SPM users.

[zeynabre]

Sure. I have attached both images here. I used "MNI152_T1_1mm.nii.gz" as reference image.

ANTS.nii.gz
FLIRT1.nii.gz

[cookpa]

Thanks.

I see in the FLIRT1.nii.gz the qform and sform codes are set to 4. This is different to what antsRegistration would produce, but it may not matter unless SPM specifically looks for that code.

[John1231983]

Dear all, related to flirt. If I want to algin an image to another image, Should I resize (zero padding) to these image to the same size before applying alignment in flirt? My first image size is 120x128x256 and the second is 122x 125x 240. Thanks all