you should use the "source weighting image" option in the old normalise routine. Build a mask, with 1 where you have good signal and 0 everywhere else. This *should* provide the best normalisation possible given your data.
Da: SPM (Statistical Parametric Mapping) <[log in to unmask]> per conto di Paolo Taurisano <[log in to unmask]>
Inviato: mercoledì 6 luglio 2016 15.08.16
A: [log in to unmask]
Oggetto: [SPM] partial acquisition normalization
we have acquired fMRI images with a 20slices/5mm thickness protocol which did not record data from rather large areas in the occipital and temporal lobe. For some of the acquisitions, we also have structural data. We need to preprocess functional data with the “old normalize” spm12 module, but we are aware that normalizing partial acquisitions with this algorithm could lead to a processing bias and to bad normalization output (as indicated in SPM12 manual, page 205). Indeed, in our experience SPM tries to "amend" the partial acquisition by inflating the volume like a baloon to cover the areas which were not acquired. In turn, this leads to increased spatial correlation between voxels, increased number of multiple comparisons for data not acquired, and to distorsions of the correlation structure obtained in the smoothing step, as is evident when we estimate smoothing post-hoc, e.g. with the REST toolbox.
Do you know any valid strategies to preprocess this type of data with the “old normalize” module in order to avoid these problems?
Many thanks in advance for your precious help.
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