Hi,

Can someone please tell me where can i find the ACR-NEMA 1.0 file
format standards. I am writing a utility which will convert the ACR-
NEMA 1.0 files into DICOM 3 format using C++. If there are any utility
doing the same task then please do let me know.

Thanks
Santosh S.B

ACR-NEMA 1.0: ftp://medical.nema.org/medical/dicom/1985/ACR-NEMA_300-1985.pdf
ACR-NEMA 2.0: ftp://medical.nema.org/medical/dicom/1988/ACR-NEMA_300-1988.pdf

Nils

On Jan 12, 7:08=A0am, santosh s b <santoshbalma...@gmail.com> wrote:
> Hi,
>
> Can someone please tell me where can i find the ACR-NEMA 1.0 file
> format standards. I am writing a utility which will convert the ACR-
> NEMA 1.0 files into DICOM 3 format using C++. If there are any utility
> doing the same task then please do let me know.

If you are aiming only at ACR-NEMA 1.0 and not 2.0 (ie. no
compression), then most DICOM toolkit out-there do import them:

- GDCM, http://gdcm.sf.net
- DCMTK, http://dcmtk.org
- dicom3tools, http://www.dclunie.com/dicom3tools.html

I do not know of any utility directly converting ACR-NEMA 1.0 directly
to DICOM, since this would imply not-only finding the proper SOP Class
automagically, but also fixing any missing/incomplete attribute (as
described by IODs).

Good luck
-Mathieu

Hi Santosh,

Long ago and far away I wrote some perl scripts to convert a bunch of
MR images from ACR-NEMA 2.0 to DICOM 3.0.  (I was recovering images
from a failed archive.  The project was successful).

Here are the main issues I remember:

1) I didn't have to worry about mapping the SOP Class (they were all
MR's).  You could probably base the DICOM SOP Class on ACR-NEMA
"Modality".  If your images are mainly CT's and/or MR's, it would
probably work.  If they're Ultrasound, well....  (find another job...
whatever they're paying you isn't enough).

2) You have to generate UID's, and you have to keep consistent Study
and Series UID's for images which are in the same study and series.  I
did this by matching patient id, study id, and series id.  I was
continually building a database, so I could check if an image seemed
to be in an already existing Study and/or Series.  This implies that
the conversion is a "batch" conversion and not a standalone "image"
conversion.  (Actually, I can't imagine taking on such a project
without using a database, which is populated as the conversion
proceeds).

3) Back in the ACR-NEMA days, vendors were not too particular about
including 3D positioning information.  You'll be lucky if you get good
"Image Orientation" and "Image Position"  (these are just like "Image
Orientation Patient" and "Image Position Patient" except that the axes
aren't patient relative (and they have different tag numbers).
There's also no "Frame of Reference", so you'll need some logic to
decide which images are spatially related (usually just if they're in
the same study) and generate a Frame of Reference UID.  You won't be
able to use the resulting CT's or MR's without this information.
There may be ways to fabricate such information in the absence of
"Image Orientation" and "Image Position" (if you believe "Slice
Location" is accurate).

4) There's this thing called "acquisition id" which is in the ACR-NEMA
"identification hierarchy".  Two (different) images may have the same
Study Id, Same Series Id, Same Image Id, but different Acquisition
Id.  This makes them different images.  If I recall correctly, this is
more a problem for CT's than MR's (but I could be misremembering).
You'll have to look at the data to figure out what the best way of
mapping these images into DICOM Series.

If I were in your position, I'd just write a perl script, not a C++
program.  The reason being that this project sounds to me like it has
"one off" written all over it, and I'd want to lower my overhead by
using a scripting language.  There's enough pain and suffering in the
world already.

If you are interested in looking at doing it in Perl, there is an open
source perl package called Posda (http://posda.com) which I support
(Posda is actually the great grandchild of the collection of perl
scripts I wrote for that conversion).

If you have religious reasons for not using Perl (which many people
do, and I understand), I believe that there are packages for dealing
with DICOM in both of the more politically correct languages Python
and Ruby.  Its just that I'm not familiar with these packages, being
an unrepentant Perl user.

I hope some of this information is helpful to you.

Good Luck,

Bill Bennett

On Jan 12, 12:08=A0am, santosh s b <santoshbalma...@gmail.com> wrote:
> Hi,
>
> Can someone please tell me where can i find the ACR-NEMA 1.0 file
> format standards. I am writing a utility which will convert the ACR-
> NEMA 1.0 files into DICOM 3 format using C++. If there are any utility
> doing the same task then please do let me know.
>
> Thanks
> Santosh S.B

On Mon, 11 Jan 2010 22:08:02 -0800 (PST), santosh s b
<santoshbalmatti@gmail.com> wrote:

>Hi,
>
>Can someone please tell me where can i find the ACR-NEMA 1.0 file
>format standards. I am writing a utility which will convert the ACR-
>NEMA 1.0 files into DICOM 3 format using C++. If there are any utility
>doing the same task then please do let me know.
>
>Thanks
>Santosh S.B

Bonjour Santosh,

You do know that DICOMatic does that conversion do you?

Yves

On Jan 14, 8:06=A0pm, Yves Martel <Martel(at)TomoVision.com> wrote:
> On Mon, 11 Jan 2010 22:08:02 -0800 (PST), santosh s b
>
> <santoshbalma...@gmail.com> wrote:
> >Hi,
>
> >Can someone please tell me where can i find the ACR-NEMA 1.0 file
> >format standards. I am writing a utility which will convert the ACR-
> >NEMA 1.0 files into DICOM 3 format using C++. If there are any utility
> >doing the same task then please do let me know.
>
> >Thanks
> >Santosh S.B
>
> Bonjour Santosh,
>
> You do know that DICOMatic does that conversion do you?
>
> Yves

Thanks for all the replies. Yes Yves we are working on using the
DICOMatic tool for the convertion. We will be testing the DICOMatic at
our site as soon as possible.

Thanks
Santosh S.B