Hi team!

Tissue Microarray support is not completely defined in DICOM Standard, so there are some questions that we want to clarify.

In SS4.5 of Supplement 122, it is specified the following:

“If the TMA slide is imaged as a whole, e.g., at low resolution as an index, it must be given a “pseudopatient” identifier (since it does not relate to a single patient). Images created for each spot should be assigned to the real patients.”

Therefore, an OVERVIEW image with the tissue microarray as a whole can contain all the specimen identifiers.

However, it is not so clear how VOLUME levels of the TMA can be organized according to the standard. We think about two posible options:

1. The first option is to create a DICOM pyramid for each specimen (spot) with the correct specimen ID. All these pyramids would have a common OVERVIEW image. We think that this option can have visualization problems to display all specimens as a
whole image.

2. The second option is to treat the TMA as a whole and create only one DICOM pyramid containing all specimens. This option makes harder to identify each specimen with its ID in all pyramid levels but it is easyer to generate a single image for all
specimens

So, the question is which is the correct option to generate a TMA Whole Slide Image?


Thanks in advance

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

Hi Adriana

First of all, I would not refer to ancient supplements, but rather the current standard, into which Sup 122 has long since been folded, and later supplements and CPs applied, specifically relevant sections for specimen identification being:

http://dicom.nema.org/medical/dicom/current/output/chtml/part03/sect_C.7.6.22.html
http://dicom.nema.org/medical/dicom/current/output/chtml/part17/chapter_NN.html

As for TMAs, WG 26 has not done any more formal work on this, but the situation (different subjects in the same acquisition/image) is analogous to that which we addressed for mouse hotels for pre-clinical research. As such, I have suggested to TMA folks
that they might want to consider using the Patient Group Macro solution described in:

http://dicom.nema.org/medical/dicom/current/output/chtml/part03/sect_C.7.html#sect_C.7.1.4

to address:

- single TMA images of the entire slide with spots from different subjects (Group of Patients Identification Sequence)
- cut out images of individual spots that reference the original dummy identifier (Source Patient Group Identification Sequence)

This proposed application to TMA is hypothetical, and addresses the Patient ID, not the Specimen Identifier, and locations are relative, not absolute.

There is discussion of localization of a specific specimen (which can be listed as one of many items of Specimen Description Sequence, in which Specimen Identifier and Specimen UID are nested) in:

http://dicom.nema.org/medical/dicom/current/output/chtml/part03/sect_C.7.6.22.html#sect_C.7.6.22.1.4
http://dicom.nema.org/medical/dicom/current/output/chtml/part16/chapter_C.html#sect_TID_8004

I don't know if anyone has any experience actually implementing Specimen Localization Content Item Sequence and TID 8004 for TMAs though.

And there has been no attempt, AFAIK, to harmonize the approach for patient (subject) and specimen identification localization yet.

If you are interested in pursuing this, I suggest that you join WG 26.

David

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

El lunes, 5 de octubre de 2020, 14:17:55 (UTC+2), David Clunie escribió:

Hi Adriana

First of all, I would not refer to ancient supplements, but rather the current standard, into which Sup 122 has long since been folded, and later supplements and CPs applied, specifically relevant sections for specimen identification being:

http://dicom.nema.org/medical/dicom/current/output/chtml/part03/sect_C.7.6.22.html
http://dicom.nema.org/medical/dicom/current/output/chtml/part17/chapter_NN.html

As for TMAs, WG 26 has not done any more formal work on this, but the situation (different subjects in the same acquisition/image) is analogous to that which we addressed for mouse hotels for pre-clinical research. As such, I have suggested to TMA

folks that they might want to consider using the Patient Group Macro solution described in:

http://dicom.nema.org/medical/dicom/current/output/chtml/part03/sect_C.7.html#sect_C.7.1.4

to address:

- single TMA images of the entire slide with spots from different subjects (Group of Patients Identification Sequence)
- cut out images of individual spots that reference the original dummy identifier (Source Patient Group Identification Sequence)

This proposed application to TMA is hypothetical, and addresses the Patient ID, not the Specimen Identifier, and locations are relative, not absolute.

There is discussion of localization of a specific specimen (which can be listed as one of many items of Specimen Description Sequence, in which Specimen Identifier and Specimen UID are nested) in:

http://dicom.nema.org/medical/dicom/current/output/chtml/part03/sect_C.7.6.22.html#sect_C.7.6.22.1.4
http://dicom.nema.org/medical/dicom/current/output/chtml/part16/chapter_C.html#sect_TID_8004

I don't know if anyone has any experience actually implementing Specimen Localization Content Item Sequence and TID 8004 for TMAs though.

And there has been no attempt, AFAIK, to harmonize the approach for patient (subject) and specimen identification localization yet.

If you are interested in pursuing this, I suggest that you join WG 26.

David

Thanks for your answer, David.

We will look at the possibility of making a proposal for the Standard. At the moment, I am already a member of the WG-26.

Thanks. Best regards.

Adriana

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)