[MINC-users] registering the 'old' way

[Sylvain MILOT]

Hello Marc,

You could look over Louis Collins phd thesis as he mentions various landmark
points to choose from (section 3) although you won't need as much to get
good results. Typically one needs around 12 to 15 point pairs. A short recipe
would be to identify the AC-PC line and then select point pairs which determine
brain extent in x,y,z.

See here for an example: (in register select transform type = full affine 12 parameter)

register /usr/local/mni/models/norm_avg_305_mri.mnc \
         /data/aces/aces1/sylvain/clinical/Bokos/mri/bokos_panorea_18703000_007007_mri.mnc \
         /data/aces/aces1/sylvain/clinical/Bokos/bokos_panorea_mrital.tag

I then create a .xfm file with

tagtoxfm -lsq12 bokos_panorea_mrital.{tag,xfm}

and feed this back to mritotal with

mritotal -transformation bokos_panorea_mrital.xfm ...

and if that doesn't do the trick which sometimes occur when I deal with GD enhanced mri volumes,
I keep the transformed (to xfm) tag file. Typically when I know I'm going to feed it back to
mritotal, I select roughly 8 point pairs covering the entire cerebrum and if mritotal still fails,
I add point pairs to bring the total to 12-15 point pairs.

HTH,

Sylvain



On Mon, 17 Oct 2005, Marc Schoenwiesner wrote:

> Hi,
>
> which anatomical locations should one tag when registering by hand with MNI
> register?
>
> Cheers,
> Marc
> _______________________________________________
> MINC-users at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
>

---
Sylvain Milot (sylvain at bic.mni.mcgill.ca)
              (trinity at bic.mni.mcgill.ca)
Brain Imaging Centre
Montreal Neurological Institute
Webster 2B, Room 208
Montreal, Qc., Canada, H3A 2B4
Phone  : (514) 398-4965, Fax: 398-8948
Mobile : (514) 712-1768
Office : 527 Av Des Pins O., Room 204
         Montreal, Qc., H2W 1S4