Dear SPMers,
I am working on validation of SPECT/MRI
Registration algorithms and I try to simulate realistic HMPAO SPECT data from an
MRI data set (using Monte Carlo simulation).
Concerning anatomy, I think I will use the Zubal
phantom which provides a fine MRI T1 data set and the same data set anatomically
labeled.
Nevertheless, to simulate realistic SPECT data I
need a model of HMPAO perfusion (in uCi/cm3 for the simulation software), and I
would like to use SPM Spect Template to build this model.
Although only the relative distribution of
HMPAO-Tc activity is important for the model, to initialize "realistic"
absolute values I have to convert grey levels of the SPM Spect template in
uCi/cm3. That's why I would like to know what was the dose of Tc-HMPAO injected
in the 22 subjects used to construct this template. Can anybody help me
??
Moreover, I have to fill labeled
segments identified in MRI phantom (nucleus, temporal lobes, ...) with
activity values found in the SPM SPECT template. So I have to use spatial
normalisation of Zubal MRI data set on SPM T1 template. Is it then possible for
each voxel of one labeled structure (e.g. Thalamus, ...) to obtain its
coordinates in the T1 or SPECT SPM template. Actually I have to use information
saved in *sn3d.mat file without using Write Normalized option (because grey
levels of the volume to normalize identify the different anatomic
labels).
I hope someone could help me, thanks in
advance
Best regards,
Christophe GROVA
PS : all remarks or suggestions are
wellcome.
*************************************************
Christophe
GROVA
Laboratoire SIM, faculte de Medecine,
Université de Rennes
1
