Dear Joao,
you can take realignment parameters into account by regressing your
data on them and taking the residuals prior to computing the CBF
values. I.e., you use the CBF computation formula on the residuals of
this regression.
I would not expect the influence of these parameters to be
substantial, due to the fact that the perfusion signal lives in the
subtraction between successive volumes.
Best wishes,
Roberto Viviani
Dept. of Psychiatry and Psychotherapy III
University of Ulm, Germany
Quoting Joao Pereira <[log in to unmask]>:
> Dear all,
>
> I am using ASL for a functional study, and I would like to ask you for your
> help on the inclusion of the movement parameters on the statistical (1st
> level) model.
>
> I start off with 144 volumes, realign them etc, and then I calculate the
> perfusion signal by subtraction, ending up with 72 volumes. The rp*.txt
> file generated in the initial realignment cannot be used directly in the
> model. I have tried to use only label or control parameters, as well as the
> average of these, but results deteriorate considerably (vs the model
> without these parameters - and not only a matter of statistical power as
> any threshold shows very different results). Is there any standard way of
> getting around this issue?
>
> Thank you very much indeed.
>
> Best,
>
> Joao
>
|