processing-scripts > Mapping from Functional to Anatomical
Showing 1-3 of 3 posts
Display:
Results per page:
Dec 13, 2011  10:12 PM | Chad Z.
Mapping from Functional to Anatomical
Hello,

I am trying to map the post-processed functional data to the original T1 anatomical data because I have used other software for processing the T1 data and would like to apply the voxel labelling from this to the functional data.  Could you please recommend what you think would be the easiest way to go about doing this?

I see that the script seems to do this for only the 8th time slice in 2_funcpreproc.sh and 3_registration.sh and that perhaps one viable solution is to simply loop over this step for each slice and recombine them to for the complete time series.  Does this sound like a good solution?

Regards,
- Chad
Dec 13, 2011  10:12 PM | Chad Z.
RE: Mapping from Functional to Anatomical
Nevermind, I figured it out I think.
Dec 15, 2011  03:12 AM | Maarten Mennes
RE: Mapping from Functional to Anatomical
Hi Chad,

In the scripts the registration works by calculation the transformation matrix of the functional image to the T1 using the 8th volume of the functional timeseries. Later on this transformation matrix is then applied to the whole functional timeseries to bring it into T1 space (Afterwards the T1 to standard non-linear warp is applied to bring the functional image into standard space).

We use the registration to the 8th image because its faster than to the whole volume and we obtain a transformation matrix that we can then use to register other similar images from that subject or later on when we register to standard space.

In your case I would be pretty straightforward to use your preprocessed T1. Just use that as the reference image to calculate the transformation from functional to T1 space. You can choose to also use the 8th image or just apply the transformation to the whole timeseries.

e.g., flirt -ref yourT1 -in func_timeseries -out func2highres -omat func2highres.mat -cost corratio -dof 6 -interp trilinear

I hope this matches you figuring it out!
Cheers,
Maarten