Comments on DICOM Data Dictionary 2006

[Mathieu Malaterre]

Hello there,

I have some questions about the latest version of the DICOM
dictionary. Please correct anything you think is a wrong assertion:

1)
Two tags are missing: ACR NEMA, Tag: 0020 1003/1005
1. 0020 1003 IS 1 Images in Series (RET)
2. 0020 1005 IS 1 Images in Study (RET)

2)
(4000,0010) should be "Text Arbitrary" instead of simply "Arbitrary"

3)
0018 0060 DS 1 KVP should be kVp

4)
Tags are not ordered correctly:

* 0018 9295 FD 1 Chemical Shifts Minimum Integration Limit in ppm
* 0018 9296 FD 1 Chemical Shifts Maximum Integration Limit in ppm

5)
* 2020 0010 US 1 Image Position
should be:
* 2020 0010 US 1 Image Box Position

6)
* 0012 0062 CS 1 Patient Identify Removed
should be
* 0012 0062 CS 1 Patient Identity Removed

7)
* 5000 2610 US 1 Referenced Overlay Group (RET)
should be
* 5000 2610 US 1 Curve Referenced Overlay Group (RET)

Thanks,
Mathieu

Ref:
http://gdcm.sourceforge.net/wiki/index.php/Data_Dictionary

[David Clunie]

Hi Mathieu

Thanks for the thorough review.

Mathieu Malaterre wrote:

> I have some questions about the latest version of the DICOM
> dictionary. Please correct anything you think is a wrong assertion:
>
> 1)
> Two tags are missing: ACR NEMA, Tag: 0020 1003/1005
> 1. 0020 1003 IS 1 Images in Series (RET)
> 2. 0020 1005 IS 1 Images in Study (RET)

Actually, these do not seem to have ever been in ACR-NEMA
(1985 or 1988) but were for some reason present in DICOM
1993 but marked as retired in the dictionary. I will add
them to 2007 to be sure that we do not re-use them.

> 2)
> (4000,0010) should be "Text Arbitrary" instead of simply "Arbitrary"

In ACR-NEMA this actually was just "Arbitrary", so there is
probably no reason to change it in the standard (even though
in my own dicom3tools dictionary I entered it as Text Arbitrary
for some reason).

> 3)
> 0018 0060 DS 1 KVP should be kVp

Even though "kVP" is a better abbreviation from a physics
perspective, it has been "KVP" since ACR-NEMA 1985 and is
consistent throughout the standard, so I don't see any
pressing reason to change this.

> 4)
> Tags are not ordered correctly:
>
> * 0018 9295 FD 1 Chemical Shifts Minimum Integration Limit in ppm
> * 0018 9296 FD 1 Chemical Shifts Maximum Integration Limit in ppm

You mean should be after (0018,9241) ?

Fixed for 2007.

> 5)
> * 2020 0010 US 1 Image Position
> should be:
> * 2020 0010 US 1 Image Box Position

"Image Position" is used consistently throughout the current
standard, but unfortunately the name conflicts with the retired
ACR-NEMA element (0020,0030) - in my dicom3tools dictionary I call
it Image Box Position to prevent this conflict but to change it
in the standard would require a CP; I will submit one.

> 6)
> * 0012 0062 CS 1 Patient Identify Removed
> should be
> * 0012 0062 CS 1 Patient Identity Removed

Fixed for 2007.

> 7)
> * 5000 2610 US 1 Referenced Overlay Group (RET)
> should be
> * 5000 2610 US 1 Curve Referenced Overlay Group (RET)

There is no conflict in the standard here that I can find,
but it would be more consistent to rename it as you suggest,
and it is retired so it doesn't really matter.

> Ref:
> http://gdcm.sourceforge.net/wiki/index.php/Data_Dictionary

The "extra" tags that you mention as being from dicom3tools
are from ACR-NEMA PS 2 and from various CPs and Supplements
to the 2006 files.

David