thanks

 

Stefano Marenco, MD

Senior Staff Fellow

NIMH/CBDB

10 Center Drive, Room 4S235

Bethesda, MD 20892

tel (301) 435-8964

fax (301) 451-4269

 

-----Original Message-----
From: Mark Jenkinson [mailto:[log in to unmask]]
Sent: Friday, January 09, 2004 7:05 PM
To: [log in to unmask]
Subject: Re: [FSL] flirt problems

 

Hi,

 

Are you running this on a linux machine?

There appears to be a problem with FLIRT compiled on Redhat 8 and 9. If

you are getting segmentation faults when running flirt, then download

the non-optimised version from

www.fmrib.ox.ac.uk/fsldownloads/flirt_noopt

<http://www.fmrib.ox.ac.uk/fsldownloads/flirt_noopt> which, although it

runs slower, should not crash. In the unlikely event that it still

crashes, please post the details on the FSL email list.

 

To install this version of flirt do the following:

1 - download the file from the website (as flirt_noopt)

2 - mv $FSLDIR/bin/flirt $FSLDIR/bin/flirtopt

3 - mv flirt_noopt $FSLDIR/bin/flirt

4 - chmod a+x $FSLDIR/bin/flirt

 

You may need an administrator to do the last 3 steps, depending

on whether you have permission to write to $FSLDIR or not.

 

All the best,

Mark



On Friday, January 9, 2004, at 11:39 pm, Marenco, Stefano (NIH/NIMH) wrote:

 

Hi, I am trying to register hi resolution T1 weighted images (1x1x1.5mm) 16 bit/pixel to T2 images with lower res (2x2x2), which are floating point. I use a mutual information cost function. I get a segmentation violation on some scans and not others. When I run flirt from within medx this does not happen. I think I have the latest version of the software, but which files should I check to make sure?

 

 

 

Stefano Marenco, MD

 

Senior Staff Fellow

 

NIMH/CBDB

 

10 Center Drive, Room 4S235

 

Bethesda, MD 20892

 

tel (301) 435-8964

 

fax (301) 451-4269