Hi, just guessing so this may be totally useless...
If you have coordinates of the fiducials from both scanning sessions, you
should be able to coregister them and get a registration matrix (not sure if
you can do this in FSL, just putting in my 2 cents worth, perhaps with
matlab or some geometry software - I have a matlab coregistration function
for 3 points on two objects, but not for six points on two objects). If you
can get that registration matrix, maybe you can convert it to an FSL .mat
file somehow and then apply it with the FSL tools. Anyhow, maybe there is a
way to do it with FSL. You might also consider the tkregister program in
FreeSurfer, which might work with fiducials (usually for coregistration of
electrodes to a head) and it allows you to manually adjust the registration.
Regards, Darren
----- Original Message -----
From: "gamal akabani" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Tuesday, April 29, 2003 6:59 AM
Subject: [FSL] Need advice on registration of SPECT and MRI using fsl
(FLIRT)
> Hi all:
>
> I need badly some advice on a current problem I want to tackle using fsl.
> I have an MRI data set (reference) and a SPECT data set (input). The SPECT
> data set does not provide any information regarding brain morphology.
>
> To register the MRI and SPECT images I used fiduciary markers on the
> patient's head (about 4 to 6 markers) to be able to register the data
sets.
> Thus, this is a 6 or maybe 9 traditional parameter model.
>
> My question is how can I implement a registration (using FLIRT) using the
> MRI and SPECT data that show only the markers. I have tried and the
results
> are not satisfactory.
>
> Any ideas?
>
> Thanks in advance
>
> gamal
>
|