Dear Alexander,
This is one of the best suggestions I've got so far.
I will try this and let you know.
Thank you,
Buyean
-----Original Message-----
From: Hammers, Alexander [mailto:[log in to unmask]]
Sent: Tuesday, June 06, 2006 1:34 AM
To: Buyean Lee; [log in to unmask]
Subject: RE: [SPM] Partial volume effects and Coregistration
Dear Buyean,
What happens to the _other_ caudate? If both are "off" in the same lateral
direction, it's likely to be a coregistration problem (even though that's
rare in SPM2), if both seem pushed away from the ventricle, your partial
volume effect explanation has some backing. Even then, it could still be a
voxel size problem in either PET or MRI.
As normalised mutual information coregistration is so reliable, it's worth
checking for other reasons, the usual suspect being movement:
Are you talking about dynamic ligand PET, or water PET? In either case, you
can derive time-activity curves from the dynamic image, either from your
ROIs drawn onto MRI or, more simply, directly onto the ADD image. Go for a
region that has high contrast (caudate is likely to be good, but obviously
this depends on the radioligand) and reasonable counts, outline the region
to the edge, derive the TAC, and look at it. If there has been movement this
will show up as "kinks" in the curve (beware noise - it requires a bit of
experience to look at these).
HTH, A
-----Original Message-----
From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]]On
Behalf Of Buyean Lee
Sent: 06 June 2006 02:25
To: [log in to unmask]
Subject: [SPM] Partial volume effects and Coregistration
Dear all,
I would liket ask help regarding coregistration of a T1 MRI to the dynamic
PET images.
I have been using SPM2 to transfer the volume of interest (VOI) drawn on T1
MRI image to 34 frames of dynamic PET images using the following steps.
1. Regalign 33 frames to 17th frame using 'Realign' and genereate the mean
PET image.
2. Coregister T1 MRI to the mean PET image and at the same time coregister
the VOI mask images to the mean PET image as other images.
The coregistration seems to be OK, but when I closely take a look at the
accuracy of coregistration, it is not as accurate as I would like to see,
especially in the striatum around the ventricle.
For example, the center of a circle drawn on the caudate of the coregistered
MRI DOES NOT match the hottest spot of the caudate of the coregistered mean
PET image. In generall, the activity (image) of the caudate, in the PET
image, seems to be a little bit off from the ventricle.
I suspect this might be true becasue of partial volume effects.
Due to low activity of ventricle, the activity near the ventricle will be
relatively lower than other parts of the caudate.
I wonder if anybody has a similar problem like this.
Any suggestions?
Thank you,
Buyean Lee.
----------------------------------------------------------
IMPORTANT WARNING: This email (and any attachments) is only intended for the use of the person or entity to which it is addressed, and may contain information that is privileged and confidential. You, the recipient, are obligated to maintain it in a safe, secure and confidential manner. Unauthorized redisclosure or failure to maintain confidentiality may subject you to federal and state penalties. If you are not the intended recipient, please immediately notify us by return email, and delete this message from your computer.
----------------------------------------------------------
|