[Insight-users] ImageSeriesReader

Tue, 22 Apr 2003 16:10:33 +0200 (CEST)

Hi Jolinda,

this is exactly what I was looking for. Now it works
fine with my 4D cardiac data set, producing one 3D
volume (MHD + RAW) for each phase of the heart cycle.
Big thanks, I think your program will save me a lot of
time :-)

Mathias

 --- Jolinda Smith <jolinda@darkwing.uoregon.edu>
schrieb: > Hi Matthias,
> 
> I added "trigger time" to the list of things that
> identify volumes in a
> series. Could you try out the new version? I don't
> have any 4D cardiac
> sequences to test it on (we have a head-only
> machine), so I'd appreciate
> you letting me know if it works.
> 
> Thanks,
> 
> Jolinda
> 
> ----- Original Message -----
> From: "Mathias Seitel" <mathiasseitel@yahoo.de>
> To: "Jolinda Smith" <jolinda@darkwing.uoregon.edu>
> Cc: <insight-users@public.kitware.com>
> Sent: Wednesday, April 16, 2003 8:04 AM
> Subject: Re: [Insight-users] ImageSeriesReader
> 
> 
> > Hi Jolinda,
> >
> > The software worked fine, and the new
> functionality
> > for specifying any required tags is really nice.
> > However I'm somewhat confused as to how the
> individual
> > slices of a DICOM series are assembled in the RAW
> > file.
> >
> > Specifically, I've got a 4D-MRI cardiac volume
> series,
> > depicting one cardiac cycle in 3D. It consists of
> 400
> > slices in total, 16 for each volume (thus 25
> cardiac
> > phases). The image numbers range from 1 to 400.
> The
> > volume of the first cardiac phase consist of
> images 1,
> > 26, 51, the second volume of images 2, 27, 52 and
> so
> > on.
> >
> > MRIConvert extracts just the first volume (1, 26,
> ...,
> > 376), thus the resulting RAW file is 2 MB in size
> (256
> > x 256 x 16 Bit). I'm wondering how the program
> knows
> > just which slices to use for the volume, skipping
> the
> > remaining 24 each time. Since no Slice Location
> tag is
> > provided in the DICOM files, all I can figure out
> > would be either use the z-coordinate of the Image
> > Posistion Patient tag, or the Trigger Time tag,
> both
> > of which could be used to correctly identify the
> > individual volumes. Or am I ignoring something?
> >
> > On the other hand, would it be possible to extract
> not
> > only the first volume, but all volumes of the 4D
> data
> > set? This would be a very convenient solution...
> >
> > Thanks,
> > Mathias
> >
> >
> >  --- Jolinda Smith <jolinda@darkwing.uoregon.edu>
> > schrieb: > Okay, it's finally done. You can now
> use
> > MRIConvert
> > > to create metaimage
> > > headers without raw files, and can save any
> known
> > > DICOM tag to the
> > > header. Check it out:
> > >
> http://lcni.uoregon.edu/~jolinda/MRIConvert.html.
> > > Let me know if you have any trouble -- the new
> > > version hasn't exactly
> > > been thoroughly tested. BTW, if there's a tag
> you
> > > need to save that
> > > isn't in the list, just add it to the
> "generic.txt"
> > > file. It's not a
> > > complete list of DICOM tags, just the most
> common
> > > ones for MR images.
> > >
> > > Jolinda
> > >
> > > ----- Original Message -----
> > > From: "Mathias Seitel" <mathiasseitel@yahoo.de>
> > > To: "Jolinda Smith"
> <jolinda@darkwing.uoregon.edu>
> > > Cc: "insight users"
> > > <insight-users@public.kitware.com>
> > > Sent: Friday, March 21, 2003 6:10 AM
> > > Subject: Re: [Insight-users] ImageSeriesReader
> > >
> > >
> > > > Hi Jolinda,
> > > >
> > > > For our purpose image modality and
> study/series
> > > UIDs
> > > > would probably be needed in addition. Rescale
> > > > slope/intercept and window width/center or LUT
> > > would
> > > > also be useful, but I'm not sure if this is
> > > required.
> > > >
> > > > Would it be possible to create a meta image
> header
> > > for
> > > > the DICOM files directly, without the
> intermediate
> > > RAW
> > > > file? I'm looking for a way to let the user
> just
> > > > select the set of DICOM files he wants to
> load.
> > > >
> > > > Thanks,
> > > > Mathias
> > > >
> > > >  --- Jolinda Smith
> <jolinda@darkwing.uoregon.edu>
> > > > schrieb: > Currently, I'm not saving all
> possible
> > > > metaimage
> > > > > tags in the header,
> > > > > just the ones I need for my own purposes.
> Let me
> > > > > know what information
> > > > > you need from the DICOM file, and if there
> are
> > > > > corresponding metaimage
> > > > > tags available, I'll add them to the
> converter.
> > > > >
> > > > > ----- Original Message -----
> > > > > From: "Mathias Seitel"
> <mathiasseitel@yahoo.de>
> > > > > To: <insight-users@public.kitware.com>
> > > > > Sent: Wednesday, March 19, 2003 4:47 AM
> > > > > Subject: Re: [Insight-users]
> ImageSeriesReader
> > > > >
> > > > >
> > > > > > Hi Jolinda,
> > > > > >
> > > > > > thanks alot for your program, it really is
> a
> > > nice
> > > > > > tool. I tried it with some of my samples
> and
> > > it
> > > > > worked
> > > > > > fine, also reading its output with ITK.
> > > However
> > > > > I'm
> > > > > > still undecided on how to proceed. Doing
> it
> > > this
> > > > > way
> > > > > > may well be an option, but on the other
> hand
> > > in
> > > > > order
> > > > > > to access additional DICOM info I'll still
> > > have to
> > > > > > read the DICOM header(s) of the original
> > > file(s).
> > > > > But
> > > > > > first I have to see how many DICOM header
> > > > > information
> > > > > > is actually needed in the project.
> > > > > >
> > > > > > Best regards,
> > > > > > Mathias
> > > > > >
> > > > > >  --- Jolinda Smith
> > > <jolinda@darkwing.uoregon.edu>
> > > > > > schrieb: > Hi Mathias,
> > > > > > >
> > > > > > > Is the reason you don't want to write
> meta
> > > image
> > > > > > > headers for each volume
> > > > > > > due to the number of files you'd have to
> > > create?
> > > > > Or
> > > > > > > some other reason? I
> 
=== message truncated === 

__________________________________________________________________

Gesendet von Yahoo! Mail - http://mail.yahoo.de
Bis zu 100 MB Speicher bei http://premiummail.yahoo.de