Narly,

You should install the 64bit Centos freesurfer build only if you have a 64bit machine, which you can detect by typing:

uname -m

which would print:

x86_64

Freesurfer includes a Tcl/Tck/Tix/BLT library combo that is known to work with Freesurfer, so it won't use the Tcl/Tk installed on your system.

Nick

On Tue, 2007-06-12 at 09:02 +0100, Narly A Golestani wrote:

Dear Nick,

Thanks for your input re- installing FreeSurfer on Mandrake 10.2. Which of the two Centos distributions should I install? Or should I try one and then the other if the tests fail with the first?

Also, will the FreeSurfer tests that you have pointed me to detect problems with the fact that FreeSurfer looks for Tk 8.3 (I am deducing this from the mailing list posts), rather than Tcl/Tk 8.4 that Mandrake 10 (and presumably later versions) come with?

Thanks in advance, Narly.

On Sat, 9 Jun 2007, Nick Schmansky wrote:

...

Narly,

The Centos distribution would probably be the best choice. Once installed, be sure to try tkmedit and tksurfer, as shown here:

http://surfer.nmr.mgh.harvard.edu/fswiki/TestingFreeSurfer

Nick

On Thu, 2007-06-07 at 15:59 +0100, Narly A Golestani wrote:

...

Hello, I would like to install FreeSurfer on Mandrake 10.2 (Mandriva), and would like to know which version I should download. The available versions appear to be for: CentOS 4, CentOS 4 x86_64, or Red Hat 9. Also, posts to the archive seem to suggest that Mandrake may not be an optimal platform for Freesurfer, and that it may be necessary to compile tk8.3. Unfortunately I am not a unix expert and do not know exactly how to proceed. I would appreciate any guidance / suggestions. Thanks, Narly.

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

==================================== Narly Golestani Institute of Cognitive Neuroscience 17 Queen Square London WC1N 3AR United Kingdom Tel: +44 (0)20 7679 7529 ====================================

Hi Narly,

1) we've run a couple of mdeft and they seem fine.

2) It really depends on the CNR. The tools are much more robust these days, and a single acquisition is usually fine. It also depends on what you want to do with the data. If it's just for analysis of functional data then the requirements are not as stringent as if you want to do a thickness study.

cheers, Bruce

On Mon, 9 Jul 2007, Narly A Golestani wrote:

Dear all, I am a new FreeSurfer user, and have a quick question before beginning to process T1 data: on the beginners guide (entitled 'Cortical Reconstruction with FreeSurfer'), it says that the input ought to be from MPRAGE or from SPGR sequences. 1) Are these the only sequences that FreeSurfer can process? I have data obtained using a MDEFT sequence, with fat-insensitive excitation pulses and a FLASH-EPI hybrid readout. 2) Also, I only have one structural run per brain - but the tutorials / set-up pages seem to suggest that it is necessary to have 2 structural runs per brain. I would appreciate more information on the type of T1 input sequences that FreeSurfer can process. thanks, Narly. _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

the easiest thing is just visual inspection. It depends on the coil and the field strength. We routinely do a single 6.5 minute acquisition with a 12 channel coil at 3T and it seems fine. You could do test/retest studies of course, but that will tell you reliability, not sensitivity.

Bruce On Mon, 9 Jul 2007, Narly A Golestani wrote:

Hi Bruce, Thanks for your answers. It actually is for a thickness study. I hope that one run will be adequate. Are there validity checks for the analysis output? thanks again, Narly.

On Mon, 9 Jul 2007, Bruce Fischl wrote:

...

Hi Narly,

1. we've run a couple of mdeft and they seem fine.

2. It really depends on the CNR. The tools are much more robust these days,

and a single acquisition is usually fine. It also depends on what you want to do with the data. If it's just for analysis of functional data then the requirements are not as stringent as if you want to do a thickness study.

cheers, Bruce

On Mon, 9 Jul 2007, Narly A Golestani wrote:

...

Dear all, I am a new FreeSurfer user, and have a quick question before beginning to process T1 data: on the beginners guide (entitled 'Cortical Reconstruction with FreeSurfer'), it says that the input ought to be from MPRAGE or from SPGR sequences. 1) Are these the only sequences that FreeSurfer can process? I have data obtained using a MDEFT sequence, with fat-insensitive excitation pulses and a FLASH-EPI hybrid readout. 2) Also, I only have one structural run per brain - but the tutorials / set-up pages seem to suggest that it is necessary to have 2 structural runs per brain. I would appreciate more information on the type of T1 input sequences that FreeSurfer can process. thanks, Narly. _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

==================================== Narly Golestani Institute of Cognitive Neuroscience 17 Queen Square London WC1N 3AR United Kingdom Tel: +44 (0)20 7679 7529 ====================================