[Mrtrix-discussion] termination mask

[Robert Smith]

Chris

It's been a while since I've worked with the 0.2 tracking code, but from a
quick glance it seems that the behaviour is the following:
* If the mask is exited, the first point outside the mask is *not* included
in the streamline.
* If the relevant diffusion metric (FA / FOD amplitude) is below threshold,
the first point at which this criterion fails *will be* included in the
streamline.
* If the curvature threshold is exceeded, the point at the end of the step
that has exceeded this criterion is *not* included (this applies for
DT_STREAM and SD_STREAM only).
* The behaviour on entering an exclusion ROI is irrelevant as the
streamline is rejected regardless.

When we eventually get around to releasing the 0.3 code, this behaviour is
made much more explicit in the code design, and can therefore be modified
if you prefer specific behaviour for specific termination criteria.

Regarding termination of a streamline if it re-enters the seed region, I
think the answer depends on the behaviour you are seeking; do you want the
streamlines to be simply terminated if they re-enter the seed region, or
would you like them to be rejected as artefact? If the latter, a
modification of the filter_tracks command may be more digestible; you could
generate a vector specifying whether each point along the streamline is
within the seed mask, and reject any streamlines for which there are two
discrete streamline segments within the seed mask. Implementing the former
may be a little trickier, as you would need to deal with confounds such as
multiple seed regions and bi-directional tracking, but it can certainly be
done; let me know if you try but get stuck.

I believe there was a streamtrack modification distributed on the
discussion board some time ago that wrote each streamline seed to a text
file, though I'm having trouble finding it. Unfortunately it's not
currently possible to embed the data for each streamline seed within the
.tck file, so this needs to be handled externally. Of course, if you are
tracking unidirectionally, the seed point will always be the first point on
the streamline.

Best regards
Rob

--

Robert Smith
PhD Candidate

The Florey Institute of Neuroscience and Mental Health
Melbourne Brain Centre - Austin Campus
245 Burgundy Street
Heidelberg Vic 3084
Ph: +61 3 9035 7128
Fax: +61 3 9035 7301
www.florey.edu.au



On Tue, Nov 6, 2012 at 2:43 AM, Parker, Christopher <
christopher.parker.10 at ucl.ac.uk> wrote:

>  Hi Donald,
>
>  I have a few questions about streamtrack. When streamlines terminate
> (i.e. they enter an exclusion ROI, leave the mask ROI, encounter low FA or
> high curvature), are the coordinates at the current point (i.e. the
> coordinate that fulfils the termination criteria) recorded, or are the
> streamline coordinates terminated just before this point (the step before
> termination). I'm guessing for high curvature it is the latter but for the
> others it is the former.
>
>  I was also wondering if there is any way to specify a termination mask
> that is also a seed mask using streamtrack, so that streamlines that
> re-enter the mask are terminated, but are free to initially propagate
> within the mask. It is not possible with the command line tools as far as I
> can see. This would be really useful- maybe I can modify the streamtrack
> command to make this possible?
>
>  Another thing that would useful is to modify the streamtrack command so
> that it records the streamline starting coordinates.
>
>  Kind Regards,
> Chris
>
> _______________________________________________
> Mrtrix-discussion mailing list
> Mrtrix-discussion at www.nitrc.org
> http://www.nitrc.org/mailman/listinfo/mrtrix-discussion
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.nitrc.org/pipermail/mrtrix-discussion/attachments/20121106/f15bcdfe/attachment.html