The lme toolbox
allows you to have different regression models across locations. If
there is a missed data value at some location (vertex/voxel) for a
given time-point for a given subject then you can exclude the
corresponding row of the full design matrix at that location using the
parameter Xrows of :
[stats,st] =
lme_mass_fit_vw(X,Zcols,Y,ni,maskvtx,fname,prs,e,Xrows)
This allows you to
use a different number of rows from your design matrix across
locations. Xrows is just a matrix of zeros and ones. Look at the
header of lme_mass_fit_vw for more info.
If you also want to
have a different number of columns from your
"maximal model" design matrix (this the design mtrix for the model that contains the
maximum number of regressors you are going to use at any location)
then you need to work directly with
[stats,st] =
lme_mass_fit(X,Xcols,Xrows,Zcols,Y,ni) (this script is call by
lme_mass_fit_vw)
Look at the header
of this script for more info. The parameter Xcols is a matrix of
zeros and ones that allows you to choose a different subset of
columns from your “maximal design matrix” at each location.
Hope this makes
sense.
Best
-Jorge
El Miércoles 30 de abril de 2014 18:47, Marnie Shaw <marnie.shaw@anu.edu.au> escribió:
Dear Freesurfers,
Is there a way to include a per voxel regressor in an LME model within the matlab LME toolbox? Just like you can do with mri_glmfit?
Cheers,
Marnie
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurferThe information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline
. If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.