[Mrtrix-discussion] ignoring slices?

[Donald Tournier]

Hi Kerstin,

I'm afraid there is no such option for csdeconv. In my experience, CSD is
relatively robust to one or two corrupted images, but I've certainly not
gone as far as validating that particular claim...

There are ways you could do the analysis, but they're not trivial.
Essentially you'd need to do the full csdeconv run, then for each corrupted
slice:
- generate a set of DWI with the relevant volumes taken out (mrconvert
--coord);
- get the DW encoding (mrinfo --grad);
- edit it to remove the corresponding gradient for that volume;
- do the csdeconv analysis with the reduced DWI set and its corresponding
encoding (csdeconv -grad).
Repeat as necessary for any other volumes/slices, and finally splice your
data back together using various combinations of mrconvert --coord to
extract the slices you want from the CSD outputs you trust for those slices,
and mrcat to concatenate them back together in the right order...

Not sure that all makes sense. It might just be easier to add the
functionality to csdeconv if you feel brave enough...

Cheers,

Donald.


On 1 April 2010 11:53, Kerstin Pannek <k.pannek1 at uq.edu.au> wrote:

> Hi all
>
> I have some datasets where a few slices have severe artifacts (1-2 slices
> out of the full dataset with 65 volumes). I was just wondering whether I
> should ignore these slices in csdeconv, and if so, how? dwi2tensor has an
> option ignoreslices, but csdeconv doesn't have this option.
>
> Cheers
> Kerstin
> _______________________________________________
> 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/20100407/1b0a7643/attachment.html