Dear Konstantinos,
What you are saying is pretty weird to me, I always successfully managed to inverse normalize atlases using normalise-write in SPM12.
Can you specify better the settings you are using?
One idea I have is that if you do an inverse normalization like you say, SPM is going to use as an output the bounding box and pixel size of the T1 image and not of the PET. Which would cause the error you're seeing.
Could this be the case? You can easily check with the display function if the pixel size and matrix dimension are the same as the T1
Best regards
Luca
-----Original Message-----
From: SPM (Statistical Parametric Mapping) <[log in to unmask]> On Behalf Of Konstantinos C
Sent: 06 June 2020 12:42
To: [log in to unmask]
Subject: [SPM] Origin - normalisation - SPM12
Hello,
I have been working for many years with SPM8 and I have been very happy with it. Recently and after a delay of many many years I attempted to update our pipeline into using SPM12 instead. I am working with PET data and the basic idea is that I perform the T1 segmentation with the standard settings (T1 are not deriving from a certain protocol and the data are heterogeneous) and then I use the resulting inverse transformation matrix in the "normalise write" to warp an MNI atlas to the T1 native space (bounding box and voxel size is set to NaNs). This procedure has worked adequately in the SPM8 and resulted to perfectly warped atlases to native T1 space that I later used for the regional kinetic modeling of the co-registered PET data. However, in SPM12 the warped, resliced atlas is not having the space (same bounding box but not voxel size) and origins of the T1 image. Therefore I cannot use the atlas for masking with the individual grey matter mask ("Warning: The images do not all have same orientation and/or voxel sizes") and perform the regional sampling. Is this a complication of the new SPM12 normalization, and in that case how do I bypass that issue? I want to avoid going back to SPM8 but if there is no other alternative I will do it.
Grateful for any help
Cheers,
Konstantinos
Rispetta l’ambiente: non stampare questa mail se non è necessario.
Respect the environment: print this email only if necessary.
|