Hi Andreas,

It is extremely different to get good registrations manually (as you have to work in 3D and you only see 2D planes), and you'll most certainly not be able to achieve the accuracy of an automated method (especially not of robust_register/robust_template, which are pretty damn accurate, if you ask me :-). So I strongly advise against that (unless you want to show that manual registration messes things up).

Now to answer your question: You could theoretically use mri_robust_template to construct the template from your three inputs. Then stop the stream, mri_concatenate_lta your transform from image_i to image_0 with the transform that roubst_template gives you from image_0 to template. Then reconstruct the template using mri_robust_template with --ixform (passing these transforms) and --noit to not iterate, this will recreate the median template image.
Also replace the lta files in the base/mri/transforms with the new ones that you got from the concatenation, so that the base and later the long streams know.
Resume the recon-all stream.

This, obviously, has never been tested and requires some looking into recon-all to make sure you do things right. E.g. you need to use the norm.mgz of the inputs and make sure you overwrite the norm_template or similar (the one that get's created in the base step).

Best, Martin


On 04/09/2014 05:35 AM, Andreas Werner wrote:
Dear Experts,
 
i am analyzing a longitudinal study (three timepoints for each participant).
I manually register the second and third timepoint to the first timepoint using freeview.
Question: Is it possible to use these manually created transformation matrices to create a common template that is equidistant from all three timepoints to balance interpolation ? If so, which commands do I need to use ?
(As I understand " mri_robust_template " does this, but I would like to use manual rather than automatic registration.)
Many thanks in advance!
Andreas


_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

-- 
Martin Reuter, Ph.D.

Instructor in Neurology
  Harvard Medical School
Assistant in Neuroscience
  Dept. of Radiology, Massachusetts General Hospital
  Dept. of Neurology, Massachusetts General Hospital
Research Affiliate
  Computer Science and Artificial Intelligence Lab,
  Dept. of Electrical Engineering and Computer Science,
  Massachusetts Institute of Technology

A.A.Martinos Center for Biomedical Imaging
149 Thirteenth Street, Suite 2301
Charlestown, MA 02129

Phone: +1-617-724-5652
Email: 
   mreuter@nmr.mgh.harvard.edu
   reuter@mit.edu
Web  : http://reuter.mit.edu