[Brains-users] Problems with tissue class models due to dark T1images

[Ronald Pierson]

Thanks for your patience with my reply.  

Other image dimensions shouldn't give you any problems, as long as the whole
brain is in there.  It uses the Talairach.bnd to set the coordinate system,
so that just needs to be picked as you normally do in workup.

I have been working on converting the script to tcl, which should be
run-able through the console.  One question - on our version of brains2,
under the Utility option, is there an Advanced Console option?  If so, that
would be great.  

Following is a step-by-step way to run this without the script, but you need
to console for one command "b2 clip image"


1 Load your T1 image.

FIND THE IMAGE INTENSITIES INSIDE THE BRAIN.
2 Load the Talairach.bnd
3 Load the Talairach box for the brain (load, load Talairach box, choose
brain_box).
4 Convert the box to a mask (convert, Talairach box to mask, OK).
5 Measure the image statistics in this mask (Measure, Measure Image Mask,
select the mask and the image, Meaure).
6 In the measurements box find the maximum intensity.  Add 20 to this to
make it a nice threshold.

REMOVE BRIGHT STUFF FROM THE IMAGE
7 Utility, Image Tools, Threshold image.  Set the threshold to 0, then okay.
This creates a mask of the entire image.
8 Open advanced console (Utility, Advance Console)
9 Enter this command in the console:

b2 clip image i1 m2 upper-bound= $threshold

$threshold in the above command needs to be replaced with your threshold you
found in 6 for this image.

CREATE NEW IMAGE
10 Add the image to itself (Utility, Image Tools, Image Math)  Select the
clipped image for both images, and ADD for the operation.  

This should get you a much better looking T1.  Save this new image and use
it for segmentation.


Let me know how it works.

Ron



-----Original Message-----
From: brains-users-bounces at psychiatry.uiowa.edu
[mailto:brains-users-bounces at psychiatry.uiowa.edu] On Behalf Of Kevin Berne
Sent: Tuesday, October 04, 2005 5:03 AM
To: brains-users at psychiatry.uiowa.edu
Subject: [Brains-users] Problems with tissue class models due to dark
T1images


Hello,
Thank you very much for sending me the thresholding script along with the
explanation of it. Since I myself sadly have no idea how to run the
mentioned script under Cygwin I would like to ask you to perhaps describe a
process through which I might be able to get the same results manually, if
that process exists. I know it has already been described what the script
does exactly, but I am unable to translate that into anything that I might
be able to perform manually. So, if there is any way to do it manually,
please guide me through it. If anyone knows how to do it and finds the time
to write it down, I would be very grateful. If anyone even knows how to
actually run the script under Cygwin, that would be even better...
By the way, we do not have PD images, so the T1 and T2 images are the only
things to work with.
Finally, I had some time ago raised the question whether resampling the
images with odd values (356x356x292 for example) to prevent a cut-off-effect
we have been experiencing actually affects the measuring process later on.
Since I still do not know the answer, I would still appreciate it if anyone
knew anything about that matter.
Thank you again,
Kevin Berne
Central Institute of Mental Health http://www.zi-mannheim.de
Department of Neuropsychology D-68159 Mannheim J5
and Clinical Psychology  
 
 

______________________________________________________________________
XXL-Speicher, PC-Virenschutz, Spartarife & mehr: Nur im WEB.DE Club!

Jetzt gratis testen! http://freemail.web.de/home/landingpad/?mc=021130

_______________________________________________
Brains-users mailing list
Brains-users at www.psychiatry.uiowa.edu
http://www.psychiatry.uiowa.edu/mailman/listinfo/brains-users