[caret-users] Question about surefit and caret (intensity inhomogeneities in images)

[Donna Dierker]

Hi Radhika,

The only way to answer that question (will 3dUniformize work) is to try 
it on your data.  You should also try the other methods listed in the 
earlier thread:

http://www.mail-archive.com/caret-users@brainvis.wustl.edu/msg01429.html

Based on previous experience with surface coil anatomical data, I 
wouldn't bet your house that it will work well enough to get a decent 
segmentation out of Caret; however, previous experience also shows that 
the term "decent" is very subjective, depending not only on the person 
judging, but also his/her purposes for the surfaces.

The F99UA1 images shown in David's 2004 paper were definitely NOT 
generated from a surface coil anatomical.  The structural MRI that 
yielded those fine surfaces came from Nikos Logothetis, using 
acquisition methods that inspire envy in almost everyone who sees it.  
The M3, if I read the paper correctly, came from Katrin Denys, and 
having had a quick look at some intermediate anatomical volume data from 
that monkey, it also looks considerably better than the old surface coil 
scans I used to see here at Wash U.  It is unlikely that any surfaces 
you generate using your anatomical data will be as high quality, but 
they might still be useful for some purposes.

You won't know whether they will be useful until you try the 
nonuniformity methods; select the best resulting scan; and try caret.  
The worst that can happen is that you'll have more concrete evidence to 
justify improved acquisition methods.

Donna

On 08/06/2008 06:18 AM, Radhika Palkar wrote:
> Hello all!
>
>  I was reading the following paper "Van Essen, D. C., (2004). Surface-based
> approaches to spatial localization and registration in primate cerebral cortex.
> Neuroimage. 23:S97-S107" in regards to the use of caret and surefit. My question
> is regarding the images used to develop the cortical surfaces.
>  In my lab we are trying to create similar surfaces but have severe intensity
> inhomogeneity in the images due to the use of surface coils. I have not yet
> tried caret on my images and I was wondering if the intensity correction
> program it has (3dUniformize) will work on my them. My images actually look
> exactly like the example shown not to work in the tutorial! I have used
> freesurfer's program nu_correct but it does not get rid of the problem
> sufficiently.
>  I want to know if this paper used better images or a better software to correct
> the intensity problems?
>
> Sincerely,
>
> Radhika
> _______________________________________________
> caret-users mailing list
> caret-users at brainvis.wustl.edu
> http://brainvis.wustl.edu/mailman/listinfo/caret-users
>