John,
Will this new procedure in SPM5b allow for proper segmentation when the
FLAIR scans are rife with ischemic white matter pathology...which is
typically mischaracterized as bone in the current SPM2 segmentation routine.
Or, am I still stuck with tedious masking?
Thanks,
Jeff
--
Jeffrey N. Browndyke, Ph.D.
Assistant Professor - Dept. of Psychiatry
Duke University Medical Center
Bryan Alzheimer's Disease Research Center
2200 West Main St., Suite A230
Durham, NC. 27705
(919) 416-5383 voice
(810) 821-0424 fax
-----Original Message-----
From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]]On
Behalf Of John Ashburner
Sent: Thursday, May 12, 2005 8:22 AM
To: [log in to unmask]
Subject: Re: [SPM] FLAIR-template
> > I would like to normalize my FLAIR data, but cannot find a
> > FLAIR-template. Does anyone have a FLAIR-template? Or is there another
> > way of normalizing this data, e.g. by using a T1-template?
>
> I've create once a Flair template (which, unfortunately, I can not make
> available at the moment), by coregistering the Flair images with the T1
> image of the same subject (do not necessarily need to be a 3D, an axial T1
> images should work as well, I guess). Finally, you can normalize the T1
and
> apply the normalization parameter to the Flair image.
>
> If you have only the Flair images, you can probably use the T2 template
and
> use only an affine transformation. You should than average all the
> normalized Flair images, in order to create a new template, which you use
> as the templatee for the second round of normalization, by normalizing the
> original Flair images onto the new template.
You may also want to try the Segment button in SPM5b, which is now available
(this is a very subtle release announcement for the beta release of SPM5 for
those of you who are taking notice). This will allow to to spatially
normalise a much wider range of images with different contrasts. Basically,
the segment button will now produce sn.mat files, which can be applied using
the normalise button. Note that the images will be written as NIFTI-1
format, which means that once you switch, there will be no going back to
SPM2. The idea is that a few enthusiastic people will try some beta testing
of this early (and possibly buggy) version of SPM5.
Best regards,
-John
|