Thank you Helmut for your help
I've put some images on imgur: https://imgur.com/a/qokH5 . FDG PET images are peculiar -in a good way- because like 90% of the uptake is in the gray matter, so the anatomical details are pretty good for a PET, and there's nothing outside confounding. I've attached also images from the MRI segmentation superimposed with the native space of the 2 subjects, and you can see that everything looks definitely fine there. When you switch to the normalized PET images, instead, you clearly see many differences in the cerebellum, at the top, and even some at the sides of the coronal plane (but maybe they're minor).
BTW, I've generated ad-hoc images smoothed only 5 mm to highlight the spatial differences better.
Thanks again,
Luca
-----Messaggio originale-----
Da: MRI More [mailto:[log in to unmask]]
Inviato: mercoledì 27 settembre 2017 14:30
A: [log in to unmask]; PRESOTTO LUCA <[log in to unmask]>
Oggetto: Re: PET normalized via DARTEL don't look perfet
Dear Luca,
The swmc1 files are obviously skull-stripped, so they should indeed look similar to each other. It does not matter whether the field of view for the T1 image of a certain subject was shifted somewhat more dorsally or ventrally as long as the brain was covered. This would not necessarily be the case for e.g. fMRI data with a limited number of slices and possibly also holds for the PET images. Depending on original FOV you would have non-zero voxels in more ventral slices in the normalised data of some subjects but not in others simply because these regions were covered in only some of them.
Hopefully this explains your findings. Otherwise it might be useful if you could upload some screenshots somewhere to get a visual impression of your issue.
Best regards
Helmut
[BannerOSR_731X129-01]<http://www.5xmille.org/>
IL TUO 5XMILLE AL SAN RAFFAELE DI MILANO
Devolvi il tuo 5 per mille all’Ospedale San Raffaele: perché al centro della Ricerca ci sei TU. CODICE FISCALE: 07636600962, nel riquadro RICERCA SANITARIA. Non c’è cura, senza ricerca. Non c’è ricerca, senza il tuo 5xmille. Scopri come su http://www.5xmille.org
Rispetta l’ambiente: non stampare questa mail se non è necessario.
Respect the environment: if it's not necessary, don't print this mail.
|