Dear Davide,
There are a couple of problems with your approach. As a first thing, I don't understand why you perform the realignment if your scan isn't a dynamic one.
Then, before running the normalization, you should set the origin of the scan so that the image is aligned with the MNI space. Many reccomend doing it manually. Personally I do this in an automated way by first setting the origin to the center of mass of the image and then running a coregistration. (probably also setting the origin to the central voxel would be ok. But you still need a coregistration afterwards. Try checking where 64.5 64.5 51.9 falls).
Finally, there's a huge problem with normalizing florbetaben images. The "new normalize" method doesn't work really well on PET images, it's meant for MRI. Anyway, amyload imaging carries very little anatomical information, and, even more problematically, the expected tracer distribution changes markedly between positive and negative scans.
I'd truly reccomend either acquiring an MRI for each patient or using other information, if you have them (an FDG PET or whatever you got where you can recognize some anatomy)
Cheers,
Luca
________________________________________
Da: SPM (Statistical Parametric Mapping) <[log in to unmask]> per conto di Davide Sardina <[log in to unmask]>
Inviato: venerdì 23 settembre 2016 10.31.30
A: [log in to unmask]
Oggetto: [SPM] PET florbetaben normalisation SPM12 problem
Dear SPM Users,
I’m trying to perform a normalisation step in order to convert a pet florbetaben into MNI space.
I started from 47 slices in dicom format and imported them with DICOM import obtaining a unique nii file (dimension: 128 x 128 x 47, origin: 64.5 64.5 51.9). Then, I used “Realign (Est & Res)” function on this file and got ‘means’ and ‘realigned’ files.
Now I need to perform normalisation by using “Normalise (Est & Write)”. As suggested, I changed the smothness parameter to 5 and pass the nii ‘means’ file obtained from the realign step as “image to align” and the ‘realigned’ nii file as “images to write”. Left the other parameter as default.
I obtained a normalised result (dimension: 79 x 95 x 79, orgin: 40 57 36) that miss half of the brain (top: realign pet; bottom: normalised pet ) as you can see in figure: https://www.dropbox.com/s/oho995ry2pcwnjx/check-reg.png?dl=0
I think that the problem is related to bounding box option. Could someone suggest a way to solve this problem? Is the normalisation correct or is there some limitation with pet in SPM?
I’m looking forward to hear from you!
Davide Sardina
PhD student
University of Catania
Rispetta l’ambiente: non stampare questa mail se non è necessario.
Respect the environment: if it's not necessary, don't print this mail.
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
[http://www.5xmille.org/img/731x129.jpg]
|