Thanks for the reply. I have some more questions in the same direction.
Could someone please calrify these also?

Considering the example in my previous post, where I have a procedure having
CT series of different organs (body part examined).
Supposed there are 3 Hanging Protocols defined as follows: (I am questioning
only the selection of the Hanging Protocol - assuming the displaysets are

Now, could you please clarify which hanging protocol would match in the
following scenarios?
1. When a procedure having only CT-CHEST images is loaded?
2. When some series have CT-CHEST images and others have CT-ABDOMEN images?
3. When a procedure having only CT-ABDOMEN images is loaded?

In the above example I have only considered series where the body part
examined differs. But there could also be a possibility that a procedure has
series belonging to different modalities. For example (hypothetical example)
there could be a procedure having MR-HEAD series and a CT-CHEST series. So
how should the image selector sequence be defined to say that a particular
HP is for an image set that could have images belonging to MR-HEAD "and"
For this, if I specify one Selector Attribute as Modality and the Selector
Value CT\MR. And another Selector Attribute as Body part examined and the
Selector Value CHEST \HEAD. It would indicate that this particular selection
criteria holds good (ie the HP will match) for the image sets: CT-CHEST, or
CT-HEAD, or MR-CHEST, or MR-HEAD. But then this does not solve the problem
in our hypothetical example. Could you please give me some pointers in this
direction also?


wrote in message
news:1105709274.403642.82200@c13g2000cwb.googlegro ups.com...
> Jain wrote:
> >
> > I have some questions related to specifying image sets and selection
> > criteria in a Hanging Protocol(HP).
> >
> > 1. In DICOM terms, what does an image set equate to?
> > In our application, we are dealing with procedure steps (GPSPS/GPPS).

> So,
> > can it be said that a procedure equates to an image set?

> It depends on the application and the modality; the choice of image
> sets for hanging a series of single frame mammography images might be
> different from the image set organization for multiple series of MR
> images and different again for multiple instances of MR multiframe
> images.
> There is no restriction to any particular a priori organizing entuity
> such as a study, PPS or series.
> Further, the application needs to already have a pre-defined notion of
> what constitutues the current and prior "sets" from which it is
> filtering out the image sets in the first place ... this is not
> necessarily as broad as "all the patient's images", though could be.
> The standard allows you to select what goes into an image set as you
> see fit for the purpose of the particular hanging protocol.
> These are further mapped into display sets that go into the individual
> image boxes.
> > 2. For a "time based sequence", an "image set selector sequence" has

> to be
> > specified. Consider a procedure having CT series. Some of these have

> body
> > part examined as CHEST and others have it as ABDOMEN. That is my

> image set
> > has Modality = CT, and Organ = CHEST && ABDOMEN. The question I have

> here is
> > that how do I specify this as an item in the "Image Sets Sequence"?

> The text in Sup 60 is a little confusing with respect to whether one is
> matching a particular value in the images (e.g. the third value of
> Image Type), or any value in the image attribute against more than one
> possible value specified in the Hanging Protocol.
> The way I understand it is as follows:
> The Selector Attribute (0072,0026) would contain the attribute tag of
> the attribute Organ (there is no such DICOM attribute; perhaps you mean
> Body Part Examined). Let's say the VR of Organ is CS.
> Selector Attribute VR (0072,0050) would be CS.
> Selector CS Value (0072,0062) would be multi-valued, for example
> Selector Value Number (0072,0028) could be 1 (since Organ has only one
> value at a time) or 0 (match any value).
> In the images, a value of Organ of either CHEST or ABDOMEN would match
> this selector.
> I will check with those who wrote this to be sure that I am
> understanding it correctly.
> David