[Mrtrix-discussion] Exporting diffusion data from Bruker systems

[Donald Tournier]

Hi Serena,

MRtrix can handle NIfTI and Analyse natively, so if you can convert your
data to these formats, this might be simplest approach. It may or may not
give you the gradient directions unfortunately, but I'm sure there are ways
to obtain these by other means.

Otherwise, you can try to import the data into MRtrix's own .mih format,
which is documented on this page. I don't have much experience with Bruker
data, but if I recall correctly, they're stored as a single raw data file
with a separate text header. This actually makes it quite easy to deal
with. The .mih format is just a plain text header with 'key: value'
entries. All you'd need to do is create such a file in a text editor, with
'mrtrix image' as the first line, and the last line with 'file:
NAME_OF_RAW_DATA_FILE 0' (edit as appropriate). Other lines will be needed
to specify the dimensions, voxel size, layout, datatype, and optionally the
transform and DW directions. If in doubt, you can always convert any
vaguely similar data that you have to the .mih format, and simply edit the
resulting .mih file to suit your data (you can then discard the raw data
.dat file generated by the conversion).

Of the required entries, I expect the layout might be the trickiest to
figure out. You can find an explanation of this on this page. Essentially,
it provides enough information to figure out the correspondence between a
voxel's coordinates and the position of its value in the file (i.e. its
file offset). I recommend you simply experiment with it until you can
display a sensible looking image in MRView...

If you can, I'd also recommend you try to figure out the transform, unless
you know the slices were not oblique. You may also find information related
to the DW gradient orientations in the Bruker text file header, which you
can also add in using the DW_scheme entries. In both of these cases, you
might find that things don't necessarily match up correctly (the transform
might not be quite right, the DW directions might be inverted, etc), in
particular for data acquired in a non-transverse orientation and/or using
oblique slices. This can be tricky to handle, since there are quite a few
conversions between coordinate systems going on behind the scenes...

If you're having trouble with this, feel free to send me the Bruker header
file, and I'll try to help out.

Hope this helps,

Cheers,

Donald.


On 15 May 2013 23:39, "Counsell, Serena" <serena.counsell at kcl.ac.uk> wrote:

>  Dear All
>
> I am having trouble exporting diffusion data acquired on a Bruker system
> in a format that can be used by MRtrix. Does anyone have any experience or
> tips?
>
> Many thanks,
> serena
>
>
>
> _______________________________________________
> 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/20130516/94f28944/attachment.html