 |
 |
Ok, thank you! For some reason I understood that rc images were subject to an affine deformation, not to a rigidid one! If that’s the case, everything makes sense now! Thank you! Luca Da: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] Per conto di John Ashburner Inviato: mercoledì 22 marzo 2017 16:16 A: [log in to unmask] Oggetto: Re: [SPM] AW: [SPM] How should rc* images look like? That's correct. The "imported" images are rigidly transformed to be as close to the MNI space as a rigid transform will allow. This is determined by a Procrustes analysis ( https://en.wikipedia.org/wiki/Procrustes_analysis ) of the nonlinear deformations, weighted by the grey matter. Brains in MNI space are a bigger than any typical brain, so the imported images will all appear smaller by comparison. Best regards, -John On 22 March 2017 at 14:18, michel grothe <[log in to unmask]<mailto:[log in to unmask]>> wrote: Hi Luca, to my understanding the imported images are only rigidly transformed, so the size of the brain should not change compared to the original native space scan (and would be expected to be smaller compared to the MNI space TPMs). Best, Michel ________________________________ Von: SPM (Statistical Parametric Mapping) <[log in to unmask]<mailto:[log in to unmask]>> im Auftrag von PRESOTTO LUCA <[log in to unmask]<mailto:[log in to unmask]>> Gesendet: Mittwoch, 22. März 2017 14:51 An: [log in to unmask]<mailto:[log in to unmask]> Betreff: [SPM] R: [SPM] How should rc* images look like? Thank you for your answer! I was referring to the x,y,z dimension. I mean that the brain is smaller. I attach a picture of 5 rc images compared with a contour drawn on the gray matter TPM. The segmentation for each subject looks perfectly fine. The starting MRI scans are all 3D T1 volumetric scans (1 mm isotropic voxel) acquired on different scanners with different field strengths (these are scans downloaded from the ADNI database). Best Regards, Luca Da: John Ashburner [mailto:[log in to unmask]<mailto:[log in to unmask]>] Inviato: mercoledì 22 marzo 2017 14:26 A: PRESOTTO LUCA <[log in to unmask]<mailto:[log in to unmask]>> Cc: [log in to unmask]<mailto:[log in to unmask]> Oggetto: Re: [SPM] How should rc* images look like? When you say "all my images are significantly smaller than the TPM", are you referring to the size of the files or the actual dimensions of the images? The "imported" rc files should be smaller in size, but their x, y and z dimensions should be the same as those of the TPM.nii. The reason for this is that the TPM encodes more than just a single tissue class. Best regards, -John On 22 March 2017 at 12:42, PRESOTTO LUCA <[log in to unmask]<mailto:[log in to unmask]>> wrote: Dear experts, I'm trying to use dartel for my first time. After reading the manual and a number of tutorials from google I've used segment on a bunch of t1 images and I've set the algorithm to save also "dartel imported" images. It is my understanding from what I've read that rc images are the same maps from the segmentation, with only the affine transformation, as determined before the beginning of the segmentation, applied. Is my understanding correct? If so I'd expect them all to be around the same size of the tpm, just not of the correct shape. Instead, all my images are significantly smaller than the TPM. Am I missing something? Thank you! Luca Rispetta l’ambiente: non stampare questa mail se non è necessario. Respect the environment: if it's not necessary, don't print this mail.