Print

Print
Hi David,

Have you checked that your equivalent image to MNI152_T1_2mm_strucseg is more than just a mask, and contains different label values?  The default view in FSLView is misleading - set the display maximum to be 5 in order to see all of the labels.
If you have this set fine then you should be able to get the regional outputs to work OK.

All the best,
Mark



On 29 Aug 2012, at 17:04, David Hanbury <[log in to unmask]> wrote:

Hi Mark,
 
This works great, except I am getting a 0.0 value for vCSF.  I don’t think it’s the image because I got vCSF values with the other atlas, leading me to believe it is the rhesus atlas I used, which, by the way, does delineate CSF.  Any thoughts on this?
 
Thanks
 
David
 
David B. Hanbury, Ph.D.
Research Fellow - Pathology/Comparative Medicine
Wake Forest University School of Medicine
Medical Center Boulevard, Winston-Salem, NC 27157-1040
336-716-1591
 
From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Mark Jenkinson
Sent: Thursday, August 23, 2012 11:55 AM
To: [log in to unmask]
Subject: Re: [FSL] Sienax with monkey brains?
 
Hi,
 
The easiest way is to make a copy of $FSLDIR/bin/sienax in your own directory somewhere and edit this.
You need to change all instances of ${FSLDIR}/data/standard/MNI152_T1_2mm and replace them with your own template instead.
However, note that you need more than just the template image itself, you all need other processed versions of it such as:
 
MNI152_T1_2mm_brain
MNI152_T1_2mm_brain_mask_dil
MNI152_T1_2mm_skull
MNI152_T1_2mm_strucseg
MNI152_T1_2mm_strucseg_periph
 
Once you have all of these then you can proceed.
Note that depending on the options you pass to sienax, you may not need all of these files.
Have a look in the script to see (and if that's not easy for you then just make a version of each of the above images, and use the existing ones here to show you what is needed).
 
All the best,
            Mark
 
 
 
On 23 Aug 2012, at 15:22, David Hanbury <[log in to unmask]> wrote:


Hi all,
 
I am wanting to use SIENAX on some monkey MRIs that I have.  BET2 and FAST did not work well alone, even after tweaking the parameters, but I was getting decent results when running them through Sienax (as BET2 and FLIRT work together to eliminate more skull, thus making FAST more efficient).  The problem, however, is that because FLIRT utilizes the MNI standard template, the skull-strip does not exactly fit with monkey brains.  Moreover, I was unable to figure out (if it is even possible) how to use a rhesus template instead of the MNI.  Is it even possible?  Can anyone help with this?
 
Thanks,
 
David
 
David B. Hanbury, Ph.D.
Research Fellow - Pathology/Comparative Medicine
Wake Forest University School of Medicine
Medical Center Boulevard, Winston-Salem, NC 27157-1040
336-716-1591