[MINC-users] DTI DICOM to Minc Conversion

[Andrew Janke]

Hi Andrew,

While I probably should fix dcm2mnc to handle this case I've always
been of the opinion that converters should be written in scripting
languages to allow quick hackery, so my usual fix for such things is
to write something dodgy in perl.

I once wrote something called dcmsort that lives in here:

   https://github.com/andrewjanke/mnc2dcm

That will give you a quick idea of all the DICOM files in a directory.
It uses a version of DICOM.pm from John Harlap (and others), I should
get around to changing this to something like dcmtk but have never
found a need just yet. Yes, there is a version of my mythical mnc2dcm
converter there that will sort of work but I wouldn't recommend anyone
uses it for anything real...

The idea is that you run this on your input series of DICOM files,
that are then renamed to something sensical and put into folders in a
reasonably organised structure. At which point you pass the series of
folders into dcm2mnc and then play mincconcat after if that is what
you want. I tend to store Diffusion data in separate files (one for
each direction, B0 image) as this allows me to do quick math across
them if needs be.


a

On 4 July 2014 00:13, Andrew Wood <andrew at biospective.com> wrote:
> Hi all,
>
> I'm wondering if anyone out there has worked with ADNI DTI data. When I run
> the DICOM through dcm2mnc, the time dimension gets lost.. I've had success
> with other DTI datasets in the past (e.g. PPMI), where the different
> directions are converted along the time dimension, so there is something
> about this ADNI data is confusing the converter.
>
> I've tried several different versions of dcm2mnc, and having no luck with
> any of them.
>
> I'd appreciate any advise on how to best get this conversion working.
>
> Thanks,
> Andrew
> _______________________________________________
> MINC-users at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users