[Mristudio-users] Mean FA

[susumu mori]

Hi Linda,

I'm not sure what you exactly mean by "mean tract position". To calculate a
mean, you need a population. The "population" could be across 10 different
individuals, 10 repeated scans, 10 voxels within the same individual, etc.

I assume that you meant; "after tractography, it is common to obtain 10s of
streamlines that belong to the same pathway and can we calculate a "mean
streamline"  of the 10s of streamlines?"

If so, DtiStudio doesn't have such a function explicitly, but has the
following functions;

1) After tractography (and thus obtained 10s of streamlines), this
information is converted to 1/0 binary masking, in which voxels that contain
the streamline are assigned "1". Then it reports mean FA of the masked
voxels.
2) In the above method, a voxel that contains 1 streamline and 10
streamlines are both equally weighted (they are both assigned "1"). I
believe that there is a function to perform "weighted FA calculation".
Please ask Hangyi about the detail of this function. This may provide a
similar result to calculating a mean tract and measuring FA.

To divide the tractography results, DtiStudio can report FA of a tract at
each slice level. For example, if you have 50 axial slices, you can get a
report of FA at 50 axial levels.  Then the problem is, you have to identify
corresponding axial levels when you compare the reported slice-specific
numbers across individuals. For example, the FA of the CST at axial slice
#24 may correspond to slice #26 of another subject. When you get FA of the
entire CST from the pons to the motor cortex, you don't have this problem.
As soon as dividing the CST into "regions", you need spatial normalization
across individuals to make sure you are comparing the same "regions" across
individuals.

There are several ways to do this normalization. Probably the easiest way is
to normalize the orientation and the size of the brains across the
population using a linear transformation such that each axial level is more
or less consistent and thus the slice-by-slice reporting can be population
averaged. Or you can use our full brain segmentation using the
DiffeoMap/RoiEditor pipeline such that the CST pathway can be divided into
anatomical segments such as "pons", "midbrain", "internal capusule", "corona
radiata".

Hope this helps.

On Fri, Jul 15, 2011 at 6:14 PM, Linda Lanyon <llanyon at eyecarecentre.org>wrote:

>  Is there a way in DTI Studio (or MRI Studio) to find a mean tract
> position for a bundle of fibers along the same pathway, and then to
> calculate mean FA along that tract (preferably divided into segments along
> the tract)?
>
> The only method I can find in DTI Studio is to reduce the tracts down to a
> selected few and then perform statistics, which gives mean FA. - just
> wondering if there is a way to get the mean tract as I've seen this
> published with other software.
>
> Thanks!
> Linda
>
>
> ------------------------------------------------------------------------------------------------
> Linda Lanyon PhD BSc(Hons) CEng CITP MBCS
> Michael Smith Foundation for Health Research Post Doctoral Research Fellow
> University of British Columbia, Human Vision & Eye Movement Laboratory
> VGH Eye Care Centre, 3rd Floor Research Labs
> 2550 Willow St, Vancouver
> Phone: +1-604-675-8865
> Cell: +1-778-378-5545
> Fax: +1 604-875-4663
> email: llanyon at eyecarecentre.org
> web:
> http://www.neuroophthalmology.ca/UBCNeuroOp/JBarton/labstaff/Linda.htm
>
> Director, Society for Canadian Women in Science and Technology:
> www.scwist.ca
> *SCWIST is a non-profit association that promotes, encourages and empowers
> women
> and girls in science, engineering and technology.
> *
> ------------------------------------------------------------------------------------------------
>
>  "If we knew what it was we were doing, it would not be called research,
> would it?"
> Albert Einstein
>
> _______________________________________________
> mristudio-users mailing list
> mristudio-users at mristudio.org
> http://lists.mristudio.org/mailman/listinfo/
> Unsubscribe, send a blank email to:
> mristudio-users-unsubscribe at mristudio.org
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.mristudio.org/pipermail/mristudio-users/attachments/20110716/78e0f247/attachment.html