[Mrtrix-discussion] patch for DICOM mosaics

Wed Oct 29 15:32:58 PDT 2008

Hi Samuel,

That's very odd. I just double checked everything, and it works for
me. I went back to version 0.2.5 as released on the website, applied
the previous DICOM patch, and had the same problem as you. I applied
the new patch, and the problem was resolved. Are you sure everything
was installed properly? You might want to type "./build clean", and
then "./build" to make sure everything is up to date. Then also make
sure you delete the previous installation, and copy your new
executables & library over to your installation folder - there's a
chance you might have been using a previous version. As a check, when
you give the "-version" option to any mrtrix command, it will print
out the date it was compiled. If it prints an older date than today,
you are most likely using a previous version.

Hope this helps. If it doesn't, there's something very odd going on...
Cheers,

Donald.

On Thu, Oct 30, 2008 at 6:49 AM, "Samuel Gröschel" <S.Groeschel at gmx.org> wrote:
> Hi Donald,
>
> Unfortunately, this patch did not resolve the problem for me. It does not change the setting for the diffusion data after 'mrconvert'ing them from DICOM files, in fact it does not seem to change anything (I checked that the old files are correctly replaced by the new ones). Some problem with my system or installation? How can I check this?
>
> Thanks, Samuel
>
>
>> Hi Samuel,
>>
>> Guess that last patch still had a few issues...
>>
>> This (hopefully) final patch should fix your remaining problem with
>> the origin of the images being incorrectly set for "inverted" Siemens
>> DICOM mosaics. Install as before, will override any previous patch.
>>
>> Let me know how you go.
>> Cheers,
>>
>> Donald.
>>
>>
>> On Wed, Sep 24, 2008 at 4:45 AM, "Samuel Gröschel" <S.Groeschel at gmx.org>
>> wrote:
>> > Thanks, Donald, it works beautifully!! Samuel
>> >
>> > -------- Original-Nachricht --------
>> >> Datum: Tue, 23 Sep 2008 10:20:07 +1000
>> >> Von: "Donald Tournier" <d.tournier at brain.org.au>
>> >> An: "Samuel Gröschel" <S.Groeschel at gmx.org>
>> >> CC: mrtrix-discussion at www.nitrc.org
>> >> Betreff: Re: [Mrtrix-discussion] patch for DICOM mosaics
>> >
>> >> Hi Samuel,
>> >>
>> >> I think we've finally cracked your upside-down DICOM issue. The
>> >> remaining problem was due to the use of the comma as the decimal
>> >> separator in German (i.e. 3,14 rather than 3.14). When the system is
>> >> set to German (or French or lots of other languages), there was a very
>> >> specific portion of the MRtrix code that got a bit confused when
>> >> trying to read in a number written using a decimal point (i.e. the US
>> >> convention). I've got that fixed up in this latest patch. As far as I
>> >> can tell, the rest of the code is not prone to this problem. Obviously
>> >> do let me know if any other strange things happen!
>> >>
>> >> Install this patch as per the previous instructions. As before, don't
>> >> worry about whether you've installed the previous patches or not, this
>> >> new version should overwrite the previous changes anyway.
>> >>
>> >> Let me know if this works...
>> >> Cheers!
>> >>
>> >> Donald.
>> >>
>> >>
>> >> On Mon, Sep 8, 2008 at 9:37 PM, "Samuel Gröschel"
>> <S.Groeschel at gmx.org>
>> >> wrote:
>> >> > Hi Donald,
>> >> >
>> >> > Sorry to bring this up again but with the new patch it doesn't work
>> >> either. This time, no error message occurs but the image is turned
>> upside down
>> >> as it used to be. Initially I stored the DICOM files (T1, EPI, T2) from
>> one
>> >> subject all in one folder. After "seperating" the EPI-files in another
>> >> folder, the DICOM-reading process indicates "mrview: DICOM image
>> contains
>> >> mosaic files - reformating... 100%" which sounds perfect. Still, when
>> the
>> >> mrview-window pops up the image remains upside down.
>> >> >
>> >> > I find it strange that this does not happen when YOU try to read the
>> >> identical DICOM files. So there must be a problem with my setting. Can
>> you
>> >> think of anything that might be different?
>> >> >
>> >> > Thanks and regards,
>> >> >
>> >> > Samuel
>> >> >
>> >> > -------- Original-Nachricht --------
>> >> >> Datum: Mon, 8 Sep 2008 09:17:46 +1000
>> >> >> Von: "Donald Tournier" <d.tournier at brain.org.au>
>> >> >> An: S.Groeschel at gmx.org
>> >> >> CC: mrtrix-discussion at www.nitrc.org
>> >> >> Betreff: Re: [Mrtrix-discussion] patch for DICOM mosaics
>> >> >
>> >> >> Hi Samuel,
>> >> >>
>> >> >> No, you didn't do anything wrong. That message could easily have
>> been
>> >> >> caused by the patch. It sounds like there might be some odd cases
>> that
>> >> >> I hadn't considered when reading these DICOM mosaics. Here's a new
>> >> >> patch that *should* work no matter what (fingers crossed). Don't
>> worry
>> >> >> if you've already installed the previous patch, this one will
>> >> >> overwrite any changes made by the previous one. Obviously, you can
>> >> >> also apply this patch on an unmodified MRtrix 0.2.5 installation.
>> >> >>
>> >> >> Let me know how it goes.
>> >> >>
>> >> >> Donald.
>> >> >>
>> >> >>
>> >> >> On Fri, Sep 5, 2008 at 7:46 PM,  <S.Groeschel at gmx.org> wrote:
>> >> >> > Hi Donald,
>> >> >> >
>> >> >> > Thanks very much for the patch which should sort out my
>> >> slice-ordering
>> >> >> > issue. After applying it, however, the following error message
>> occurs
>> >> >> > when trying to 'mrview' the ("incorrectly-ordered") dicom files.
>> >> >> >
>> >> >> > mrview: reading DICOM series "ep2d_diff_tra_22x20x20_69#"... 100%
>> >> >> > mrview: lib/file/dicom/image.cpp:141: bool
>> >> >> > MR::File::Dicom::Image::operator<(const MR::File::Dicom::Image&)
>> >> const:
>> >> >> > Assertion `!gsl_isnan(distance)' failed.
>> >> >> > Aborted
>> >> >> >
>> >> >> > Did I do anything wrong during installation? Can anyone help?
>> >> >> >
>> >> >> > Regards, Samuel
>> >> >> >
>> >> >> >
>> >> >> >> Hi everyone,
>> >> >> >>
>> >> >> >> Here is a patch to fix occasional slice-ordering issues when
>> loading
>> >> >> >> DICOM mosaic images. The images would sometimes be loaded
>> >> upside-down,
>> >> >> >> due to my complete ignorance of Siemens' obscure
>> 'SliceNormalVector'
>> >> >> >> CSA entry in the image headers (thanks to the SPM5 code for
>> pointing
>> >> >> >> me in right direction...). Note that this problem really
>> shouldn't
>> >> >> >> affect anything else: if you haven't had this problem yet, there
>> is
>> >> >> >> nothing to worry about. You might want to wait until the next
>> >> release
>> >> >> >> of MRtrix which will contain this fix anyway.
>> >> >> >>
>> >> >> >> To apply the patch, simply unpack the files into the MRtrix code
>> >> >> >> folder - this should overwrite the 3 modified files. You should
>> then
>> >> >> >> issue the ./build command and install as normal. In other words,
>> >> >> >> assuming you have already downloaded the patch file into the
>> >> >> >> mrtrix-0.2.5/ folder:
>> >> >> >>
>> >> >> >> > cd mrtrix-0.2.5/
>> >> >> >> > tar xfjv mrtrix-0.2.5_dicom_mosaic_patch_2008-09-05.tar.bz2
>> >> >> >> > ./build
>> >> >> >>
>> >> >> >> and copy over into your final installation folder as you would
>> >> >> normally.
>> >> >> >>
>> >> >> >> Regards,
>> >> >> >>
>> >> >> >> Donald.
>> >> >> >>
>> >> >> >>
>> >> >> >> --
>> >> >> >> 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
>> >> >
>> >>
>> >>
>> >>
>> >> --
>> >> Jacques-Donald Tournier (PhD)
>> >> Brain Research Institute, Melbourne, Australia
>> >> Tel: +61 (0)3 9496 4078
>> >
>>
>>
>>
>> --
>> Jacques-Donald Tournier (PhD)
>> Brain Research Institute, Melbourne, Australia
>> Tel: +61 (0)3 9496 4078
>

-- 
Jacques-Donald Tournier (PhD)
Brain Research Institute, Melbourne, Australia
Tel: +61 (0)3 9496 4078