Dear Luis,
Keep in mind that I'm not very experienced in SPECT. However, if I understand correctly, when your perform subtraction, you should have an image that's positive where there's hiper-perfusion and that's negative where there's hypoperfusion. Then you feed all these images in a one sample t-test in SPM.
If this is the case, when you are looking at the results in SPM if you select a t-contrast and input "1" (or +1) as contrast SPM is going to show you where your input images are on average significantly positive (hyperperfusion in my assumptions). If you input -1 as contrast, instead SPM is going to show you where your input images are on average significantly negative (hypoperfusion in my assumptions).
(if you use an f contrast instead it's going to show you everywhere the images are significantly non-zero, in both directions).
Hope this can help
Luca
________________________________________
Da: SPM (Statistical Parametric Mapping) <[log in to unmask]> per conto di Luis Fernado Silva Castro de Araújo <[log in to unmask]>
Inviato: mercoledì 15 febbraio 2017 01.10.43
A: [log in to unmask]
Oggetto: [SPM] SPECT subtraction analysis and SPM
Hi specialists,
This is my first post in here, thus thanks the authors of SPM for the excellent tool. I am also new to neuroimaging, please be patient.
I want to analyse inter-intra epileptic SPECTS of patients that have been diagnosed with non-epileptic seisures. The sample size is seven.
During my investigations of the methods, I noticed that most of publications use the subtraction of these images, instead of comparing them using a paired t-test in SPM. The next step was then to find a way of performing these subtractions.
I noticed that each researcher uses his own formula to acquire the subtraction and, in the interest of reproducibility, decided to use an older, but open tool. I used ISAS Bioimage Suite to acquire the subtractions, which were later realigned, centered to the anterior comissure, and coregistered to the T1 MNI template (provided by ISAS) in SPM. Some significant clusters were found in a one sample t-test.
Now to the question:
Can I determine the direction of these clusters? Are these hyper or hypoperfusion? Is there some way of assess the direction of these using SPM?
Does the use of subtraction precludes determining the direction of these clusters?
Thank you,
Luis
Rispetta l’ambiente: non stampare questa mail se non è necessario.
Respect the environment: if it's not necessary, don't print this mail.
