Hi - just to clarify:
probabilistic tractography does not use an FA threshold like you
would see in streamlining tractography. Instead, when a tract reaches
an isotropic regions, the samples will tend to disperse, so voxels
after this region will have lower probabilities. In some ways, this
is akin to an adaptive threshold, based on the local uncertainty
(rather than the anisotropy).
The --usef flag that steve is referring to does something very
different. It weights the probability of following a direction by the
current estimate of the volume fraction of that direction. This has
the effect of constraining the tract very much to the white matter,
and upweighs dominant directions in a 2-fibre voxel. It does not just
stop the pathway at an arbitrary threshold. You should find that the
results change dramatically when you use the --usef flag, but you may
also find that you need many more jumps per sample, as many of the
jumps will be wasted (because they are part of the "isotropic"
portion of the voxel and hence do not move).
Cheers
T
On 4 Sep 2007, at 08:33, Steve Smith wrote:
> Hi - this is a binary flag (set in the advanced options in the GUI)
> - it's either on or off.
> Cheers, Steve.
>
>
> On 4 Sep 2007, at 05:59, Dianne Patterson wrote:
>
>> I tried to look this up on the archives but they were hanging and
>> misbehaving.
>>
>> Can someone tell me where the default anisotropy value for
>> probtrackx is set?
>>
>> I gather it is a static value between 0 and 1...
>> I'm imagining it is set in a script somewhere??
>>
>> Thankyou,
>>
>> Dianne
>>
>> --
>> Dianne Patterson, Ph.D.
>> [log in to unmask]
>> ERP Lab
>> University of Arizona
>> 621-3256 (Office)
>
>
> ----------------------------------------------------------------------
> -----
> 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
> ----------------------------------------------------------------------
> -----
|