[Mrtrix-discussion] output file of "find_SH_peaks"

[Donald Tournier]

Hi Kerstin,

Not sure what you mean by 'higher resolution'. Do you mean spatial
resolution? If so all you'd need to do is to interpolate the CSD image to a
higher resolution prior to feeding it into find_SH_peaks. I think
mrtransform might do the trick if you give it a higher resolution anatomical
image (for example) as the template image.

If you meant angular resolution, then I should point out that the
find_SH_peaks command does a full Newton-Raphson type gradient ascent, and
finds the exact peak direction. The direction set that you need to provide
to the command simply specifies a set of initial directions from which to
start the search. So basically find_SH_peaks already provides 'full' angular
resolution.

Cheers,

Donald.


On 16 February 2011 12:58, Kerstin Pannek <uqkpann1 at uq.edu.au> wrote:

> Hi
>
> is it possible to get the peaks at a higher resolution than the input csd?
>
> Cheers
> Kerstin
>
>
>
> On 15/02/2011, at 9:42 AM, Donald Tournier wrote:
>
> Hi Jeong,
>
> I wonder the order of fiber directions saved in the output file of
>> "find_SH_peaks".
>>
>> This is what I did:
>>
>> gendir 100 dir100.txt
>>
>> find_SH_peak CSD8.mif dir100.txt SH_peaks_CSD8.mif
>>
>>
>> The output file consists of 9 different 3-D volumes.
>>
>> Which ones are for X, Y, Z to indicate the direction of the 1st fiber ?
>>
>> I am guessing the first three volumes indicates the X, Y, Z of the first
>> vector, next three indicates those of the second vector, and so on.
>>
>> Is it correct?
>>
>
> Yes, this is correct. Also, the norm of each of these X,Y,Z vectors
> provides the amplitude of the FOD along the corresponding peak direction.
>
>
>> Also, I would like to know whether 100 directions are good enough to
>> estimate the directions of three fibers.
>>
>
> Yes, that's more than enough. In my experience, 60 is already enough for
> lmax 8.
>
> Cheers,
>
> Donald.
>
>
>
>
>> _______________________________________________
>> Mrtrix-discussion mailing list
>> Mrtrix-discussion at www.nitrc.org
>> http://www.nitrc.org/mailman/listinfo/mrtrix-discussion
>>
>>
>
>
> --
> Jacques-Donald Tournier (PhD)
> Brain Research Institute, Melbourne, Australia
> Tel: +61 (0)3 9496 4078
>  _______________________________________________
> Mrtrix-discussion mailing list
> Mrtrix-discussion at www.nitrc.org
> http://www.nitrc.org/mailman/listinfo/mrtrix-discussion
>
>
>


-- 
Jacques-Donald Tournier (PhD)
Brain Research Institute, Melbourne, Australia
Tel: +61 (0)3 9496 4078
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.nitrc.org/pipermail/mrtrix-discussion/attachments/20110216/4dddcd21/attachment.html