Ah, I figured out why I was getting all 0s. The syntax in the example on the website under "Using non-FA Images in TBSS" says:
tbss_1_preproc -f 1000000 *.nii.gz
This results in a "multiplicative factor of -f" and images with all 0 values. It works without the -f. Earlier it worked by accident for me when I typed
tbss_1_preproc 100000000 *nii.gz
Now I got 5000000 to work.
Thanks,
Dana
-----Original Message-----
From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Dana Perantie
Sent: Wednesday, May 14, 2008 10:58 AM
To: [log in to unmask]
Subject: Re: [FSL] scaling for MD with tbss_non_FA
Thanks for your response, Steve. I just tried using 5,000,000 but the resulting images have only 0 values. Does anyone have a suggestion?
In a preliminary attempt I used a scale of 100,000,000 and the resulting values were in a range of about -30000 to +30000. There were no negative values in the original MD images (maybe some upper limit was exceeded and the values wrapped around to be negative?).
Thank you,
Dana
-----Original Message-----
From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Steve Smith
Sent: Wednesday, May 14, 2008 6:04 AM
To: [log in to unmask]
Subject: Re: [FSL] scaling for MD with tbss_non_FA
Hi,
Yes, a scaling of about 5,000,000 is about right - as long as you
apply the same scaling to all MD images (which is what the script will
do) then that's fine - and what you want to interpret any final
results you can just divide by 5,000,000 to get back to the original
MD units that your data had.
Cheers, Steve.
On 13 May 2008, at 21:04, Dana Perantie wrote:
> Dear FSL list,
> I have a region of interest from TBSS analyses of FA images and
> would like
> to find the mean diffusivity (MD) values for the same region. To
> get the
> MD in the same space, I am running tbss_1_preproc (after re-naming
> them
> the same as the FA), and then tbss_non_FA. tbss_1_preproc requires a
> scaling argument -- my understanding is that the goal is to get the
> range
> on the order of 0 to 10,000. Here are a few examples ranges of our MD
> images:
> .00046336 to .0021267
> .00063775 to .0022076
> .00058721 to .0018281
> So, should I be using something like 5,000,000 for scaling? Is it a
> little arbitrary -- what's the effect of 5,000,000 versus 4,800,000 in
> this situation? What will be the meaning of the values after this re-
> scaling? Will the values relate in any way to the usual units for
> diffusivity (something like x10^3 mm2/s)?
>
> Thank you,
> Dana
>
---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------
|