IHE CodeSystems in AuditMessages

68 views
Skip to first unread message

Dave Franken

unread,
Jun 19, 2018, 9:25:14 AM6/19/18
to IHE MHD Implementors
Dear all,

I was wondering about mapping IHE-based audit messages to FHIR AuditMessage resources.

FHIR dictates that for coded values, URNs are used as code systems, but IHE / DICOM uses codeSystemName.
For instance the following code system names in IHE need to be mapped to URNs:
  • DCM
  • RFC-3881
  • IHE Transactions
We have found/chosen URNs for the first 2, but are wondering what kind of URN to pick for the "IHE Transactions" code system.

Any ideas?

Dave Franken
Meddex

Dave Franken

unread,
Jun 20, 2018, 2:28:15 AM6/20/18
to Robert Lichtenberger, ihe-mhd-im...@googlegroups.com

Robert,

 

For RFC-3881, you can use something based on https://tools.ietf.org/html/rfc2648 which defines URNs for RFCs

Which would mean urn:ietf:rfc:3881 for RFC-3881 codes.

 

I guess we can leave it empty for IHE Transactions, or choose something like urn:oid:1.3.6.1.4.1.19376.1.2 where 1.3.6.1.4.1.19376.1.2 is the root OID for the IHE IT Infrastructure domain, see http://wiki.ihe.net/index.php/ITI_-_OID_assignment_1.3.6.1.4.1.19376.1.2

 

Dave Franken

Meddex

From: Robert Lichtenberger <r.licht...@gmail.com>
Sent: woensdag 20 juni 2018 07:29
To: Dave Franken <dfra...@meddex.nl>
Subject: Re: IHE CodeSystems in AuditMessages

 

I was having the same problem a few months ago. I chose:
* https://tools.ietf.org/html/rfc3881 for RFC3881 codes
* http://dicom.nema.org/resources/ontology/DCM for DICOM codes

and left the system-field empty for IHE codes, which clearly is suboptimal but seems to be allowed by HL7 FHIR, since http://hl7.org/fhir/datatypes.html#Coding has:

system     Σ    0..1    uri    Identity of the terminology system

I therefore translate e.g.:

    <EventIdentification EventActionCode="R" EventDateTime="2018-04-10T10:41:58.959+02:00" EventOutcomeIndicator="0">
        <EventID csd-code="110106" originalText="Export" codeSystemName="DCM"/>
        <EventTypeCode csd-code="ITI-42" originalText="Register Document Set-b" codeSystemName="IHE Transactions"/>
    </EventIdentification>

To:
    "type": {
        "system": "http://dicom.nema.org/resources/ontology/DCM",
        "code": "110106",
        "display": "Export"
    },
    "subtype": [
        {
            "code": "ITI-42",
            "display": "Register Document Set-b"
        }
    ],
    "action": "R",
    "recorded": "2018-04-10T10:41:58.959+02:00",
    "outcome": "0"

--
You received this message because you are subscribed to the Google Groups "IHE MHD Implementors" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implemen...@googlegroups.com.
To post to this group, send email to ihe-mhd-im...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ihe-mhd-implementors/b08bb16d-bb4d-4986-8253-adc99660e027%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

 

John Moehrke

unread,
Jun 20, 2018, 7:59:52 AM6/20/18
to Dave Franken, Mauro Zanardini, IHE MHD Implementors
Dave,

Where do you still see references to RFC-3881? Those should all be DICOM now.

There is an effort in IHE to do a better job at IHE defined vocabulary. Anything you do will likely need to change.

John Moehrke
Principal Engineering Architect: Standards - Interoperability, Privacy, and Security
CyberPrivacy – Enabling authorized communications while respecting Privacy
HITRUST Certified CSF Practitioner
M +1 920-564-2067
JohnM...@gmail.com
https://www.linkedin.com/in/johnmoehrke
https://healthcaresecprivacy.blogspot.com
Postel's Law: Be strict when sending and tolerant when receiving


--

Robert Lichtenberger

unread,
Jun 20, 2018, 9:03:10 AM6/20/18
to ihe-mhd-im...@googlegroups.com
Am 20.06.2018 um 13:59 schrieb John Moehrke:
Dave,

Where do you still see references to RFC-3881? Those should all be DICOM now.
I have this in an audit record:
<ParticipantObjectIdentification ParticipantObjectID="ASCPATID20180424" ParticipantObjectTypeCode="1" ParticipantObjectTypeCodeRole="1">
    <ParticipantObjectIDTypeCode csd-code="2" originalText="Patient Number" codeSystemName="RFC-3881"/>
</ParticipantObjectIdentification>
and will translate it to:
"identifier": {

    "use": "secondary",
    "type": {
        "coding": [
            {
                "system": "https://tools.ietf.org/html/rfc3881",
                "code": "2",
                "display": "Patient Number"
            }
        ]
    },
    "value": "ASCPATID20180424"

},
(I will probably change to urn:ietf:rfc:3881 as suggested by Dave Franken).

I have thought of changing the AuditRecord itself to use "DCM" instead of "RFC-3881", however, since http://dicom.nema.org/resources/ontology/DCM will not list "2" as a valid code for "Patient Number" I wonder if this is a good idea.

John Moehrke

unread,
Jun 20, 2018, 10:25:13 AM6/20/18
to Robert Lichtenberger, IHE MHD Implementors
I think we need a CP entered to drive for a formal fix.

The Issue you observe is a mismatch between dicom that followed the RFC and defined these as attributes with fixed enumeration, where FHIR tries to be more flexible. This needs a formal solution.

John 

To view this discussion on the web visit https://groups.google.com/d/msgid/ihe-mhd-implementors/acc207bd-4447-2257-048a-fa66800cae47%40gmail.com.

Dave Franken

unread,
Jun 21, 2018, 2:21:15 AM6/21/18
to IHE MHD Implementors
I found RFC-3881 also referenced in the following paragraph in IHE ITI Vol 2a ITI-20:

3.20.7.1.3 ParticipantObjectIDTypeCode
The DICOM schema mandates codeSystemName and originalText for all coded types. When standard codes are not available and the Audit Event Report is still using the integer values that were specified in RFC3881, IHE actors shall use the integer from RFC3881 as the codeValue, the description from RFC3881 in the originalText attribute and "RFC-3881" in the codeSystemName attribute. Where codes from other coding systems are available, those codes should be used because RFC3881 has been deprecated.

 I couldn't find any other code + system for "patient number".

Dave Franken
Meddex
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implementors+unsub...@googlegroups.com.
--
You received this message because you are subscribed to the Google Groups "IHE MHD Implementors" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implementors+unsub...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "IHE MHD Implementors" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implementors+unsub...@googlegroups.com.

John Moehrke

unread,
Jun 26, 2018, 9:59:52 AM6/26/18
to Dave Franken, IHE MHD Implementors
Hi all, I am back in the office and able to look more closely at this.

When I made AuditEvent Resource in FHIR, I created FHIR specific vocabulary for all of these enumerated codes. These are referenced in the FHIR specification under AuditEvent. 

Thus when an enumeration is used in DICOM AuditMessage and you are converting to FHIR AuditEvent, you have a defined vocabulary for the very same concept with a codeing system identifier.

When you are converting a FHIR AuditEvent to a DICOM AuditMessage, where these codes are used there is no loss of information... but where some other codeing system is used in FHIR AuditEvent, you will need to handle this with extended elements in the DICOM AuditMessage.

Your specific question about ParticipantObjectTypeCodeRole of "1" for Patient. This is covered by 


John Moehrke
Principal Engineering Architect: Standards - Interoperability, Privacy, and Security
CyberPrivacy – Enabling authorized communications while respecting Privacy
HITRUST Certified CSF Practitioner
M +1 920-564-2067
JohnM...@gmail.com
https://www.linkedin.com/in/johnmoehrke
https://healthcaresecprivacy.blogspot.com
Postel's Law: Be strict when sending and tolerant when receiving

To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implemen...@googlegroups.com.
--
You received this message because you are subscribed to the Google Groups "IHE MHD Implementors" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implemen...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "IHE MHD Implementors" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implemen...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "IHE MHD Implementors" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implemen...@googlegroups.com.

To post to this group, send email to ihe-mhd-im...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ihe-mhd-implementors/0d084a7a-8c86-4090-adb6-32e30d4ead5f%40googlegroups.com.

Dave Franken

unread,
Jun 27, 2018, 2:57:43 AM6/27/18
to IHE MHD Implementors
Thanks for your reply, John,

This leaves me with one last translation, that is the IHE EventTypeCode:
EventTypeCode M EV(“ITI-18”, “IHE Transactions”, “Registry Stored Query”)

How would I encode this in the FHIR AuditMessage, what code system do I use for "IHE Transactions".

I looked to MHD for guidance, but MHD just refers back to the regular XDS counterparts for Audit messages.

Dave Franken
Meddex
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implementors+unsub...@googlegroups.com.
--
You received this message because you are subscribed to the Google Groups "IHE MHD Implementors" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implementors+unsub...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "IHE MHD Implementors" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implementors+unsub...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "IHE MHD Implementors" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implementors+unsub...@googlegroups.com.

John Moehrke

unread,
Jun 27, 2018, 7:53:31 AM6/27/18
to Dave Franken, IHE MHD Implementors
Hi Dave,

IHE defined "urn:ihe:event-type-code" for that purpose
urn:ihe:event-type-codeUsed to identify a IHE Transaction numbers when used in ATNA messages. For example in ITI-43 the audit message EventTypeCode is defined as EV(“ITI-43”, “IHE Transactions”, “Retrieve Document Set”). urn:ihe:event-type-code is the URI/URN equivalent of "IHE Transactions" for the codesystem.

See here on the codes page

I started to make a FHIR codeSystem, but don't have it filled completely. 

I would be happy if someone wants to help fill this codeSystem so that we can publish it.

John Moehrke
Principal Engineering Architect: Standards - Interoperability, Privacy, and Security
CyberPrivacy – Enabling authorized communications while respecting Privacy
HITRUST Certified CSF Practitioner
M +1 920-564-2067
JohnM...@gmail.com
https://www.linkedin.com/in/johnmoehrke
https://healthcaresecprivacy.blogspot.com
Postel's Law: Be strict when sending and tolerant when receiving

To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implemen...@googlegroups.com.
--
You received this message because you are subscribed to the Google Groups "IHE MHD Implementors" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implemen...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "IHE MHD Implementors" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implemen...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "IHE MHD Implementors" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implemen...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "IHE MHD Implementors" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implemen...@googlegroups.com.

To post to this group, send email to ihe-mhd-im...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ihe-mhd-implementors/2daebd30-dab7-4a11-a629-cb3be419ef08%40googlegroups.com.

Hyeonjun Seo

unread,
Jun 18, 2021, 4:32:31 AM6/18/21
to IHE MHD Implementors
Dear all,


I have one question for this discussion, too.

As occured in IHE_ITI_TF_Vol2b.pdf, 
the audit message for ITI-41 and ITI-42 shall contain ParticipantObjectIDTypeCode; EV(“urn:uuid:a54d6aa5-d40d-43f9-88c5-b4633d873bdd”, “IHE XDS Metadata”, “submission set classificationNode”) in AuditMessage/ParticipantObjectIdentification.

Which URI/URN shall I use for "IHE XDS Metadata"?


Hyeonjun, Seo
ezCaretech Co., Ltd

John Moehrke

unread,
Jun 18, 2021, 7:35:28 AM6/18/21
to IHE MHD Implementors, IHE XDS Implementors, Hyeonjun Seo
Question to implementers, for the purpose of improving the IHE specification. The following issue has been brought forward
As occured in IHE_ITI_TF_Vol2b.pdf, 
the audit message for ITI-41 and ITI-42 shall contain ParticipantObjectIDTypeCode; EV(“urn:uuid:a54d6aa5-d40d-43f9-88c5-b4633d873bdd”, “IHE XDS Metadata”, “submission set classificationNode”) in AuditMessage/ParticipantObjectIdentification.
Which URI/URN shall I use for "IHE XDS Metadata"?

Good catch. I suspect there will be others that we need to fixup. These are problems in the XDS space too.

A couple of alternatives that I would like to have some discussion from implementers about

What is in use today? I suspect there will be some that literally have used that string. Others left the system blank.

Some alternatives that I can think of that all might be useful 
a) don't specify a system. It is not required, it is just highly recommended - the unfortunate part is that we do indicate to use "IHE XDS Metadata", so we clearly are not recommending it be left blank.
b) "urn:ietf:rfc:3986" -- because this is what FHIR says one should use when the value is globally unique
c) "http://ihe.net" -- as this is the closest thing IHE has to an http root
d) "http://profiles.ihe.net" -- as this is likely the root where any new codesystem would be created
e) IHE should define a formal codesystem as has been done with FormatCode
    e.1) This should be done inside MHD as it is most urgently needed there

I tend to favor (b)

John Moehrke 🔥 Architect: Healthcare Informatics Standards - Interoperability, Privacy, and Security
IHE Co-Chair IT Infrastructure Planning and Technical
HL7 Co-Chair Security WG, FHIR FMG, FHIR facilitator, and FHIR Foundation founding member
Employee of By Light -- Contractor to VHA MyHealtheVet
JohnM...@gmail.com  |  M +1 920-564-2067  |  John.M...@bylight.com
 https://healthcaresecprivacy.blogspot.com


--
You received this message because you are subscribed to the Google Groups "IHE MHD Implementors" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implemen...@googlegroups.com.

Oliver Egger

unread,
Jun 18, 2021, 8:58:40 AM6/18/21
to John Moehrke, IHE MHD Implementors, IHE XDS Implementors, Hyeonjun Seo
Hi  John


Question to implementers, for the purpose of improving the IHE specification. The following issue has been brought forward
As occured in IHE_ITI_TF_Vol2b.pdf, 
the audit message for ITI-41 and ITI-42 shall contain ParticipantObjectIDTypeCode; EV(“urn:uuid:a54d6aa5-d40d-43f9-88c5-b4633d873bdd”, “IHE XDS Metadata”, “submission set classificationNode”) in AuditMessage/ParticipantObjectIdentification.
Which URI/URN shall I use for "IHE XDS Metadata"?

Good catch. I suspect there will be others that we need to fixup. These are problems in the XDS space too.

A couple of alternatives that I would like to have some discussion from implementers about

What is in use today? I suspect there will be some that literally have used that string. Others left the system blank.
On the connectathon floor this week we have seen empty ones, but also some that used the string which does not go through
the FHIR Validator, I think a solution/recommendation would be very helpful. 

Some alternatives that I can think of that all might be useful 
a) don't specify a system. It is not required, it is just highly recommended - the unfortunate part is that we do indicate to use "IHE XDS Metadata", so we clearly are not recommending it be left blank.
b) "urn:ietf:rfc:3986" -- because this is what FHIR says one should use when the value is globally unique
c) "http://ihe.net" -- as this is the closest thing IHE has to an http root
d) "http://profiles.ihe.net" -- as this is likely the root where any new codesystem would be created
e) IHE should define a formal codesystem as has been done with FormatCode
    e.1) This should be done inside MHD as it is most urgently needed there
I tend to favor (b)
I tend to favor e.1 then it could be validated with a conformance profile directly and  could be mapped  directly
from the "system string" like DCM.

Best regards,
Oliver


John Moehrke 🔥 Architect: Healthcare Informatics Standards - Interoperability, Privacy, and Security
IHE Co-Chair IT Infrastructure Planning and Technical
HL7 Co-Chair Security WG, FHIR FMG, FHIR facilitator, and FHIR Foundation founding member
Employee of By Light -- Contractor to VHA MyHealtheVet
JohnM...@gmail.com  |  M +1 920-564-2067  |  John.M...@bylight.com
 https://healthcaresecprivacy.blogspot.com


On Fri, Jun 18, 2021 at 3:32 AM Hyeonjun Seo <leonar...@gmail.com> wrote:
Dear all,


I have one question for this discussion, too.

As occured in IHE_ITI_TF_Vol2b.pdf, 
the audit message for ITI-41 and ITI-42 shall contain ParticipantObjectIDTypeCode; EV(“urn:uuid:a54d6aa5-d40d-43f9-88c5-b4633d873bdd”, “IHE XDS Metadata”, “submission set classificationNode”) in AuditMessage/ParticipantObjectIdentification.

Which URI/URN shall I use for "IHE XDS Metadata"?


Hyeonjun, Seo
ezCaretech Co., Ltd

--
You received this message because you are subscribed to the Google Groups "IHE MHD Implementors" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implemen...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ihe-mhd-implementors/319913c8-82f0-48f7-8808-6e5010ca06b8n%40googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "IHE MHD Implementors" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ihe-mhd-implemen...@googlegroups.com.

Hyeonjun Seo

unread,
Jun 22, 2021, 1:47:34 AM6/22/21
to IHE MHD Implementors
Thank you everyone.


How about this URN, for example, "urn:ihe:participant-object-type-code" or "urn:ihe:type-code:{...}", according to this below.
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
IHE defined "urn:ihe:event-type-code" for that purpose

urn:ihe:event-type-code / Used to identify a IHE Transaction numbers when used in ATNA messages. For example in ITI-43 the audit message EventTypeCode is defined as EV(“ITI-43”, “IHE Transactions”, “Retrieve Document Set”). urn:ihe:event-type-code is the URI/URN equivalent of "IHE Transactions" for the codesystem.
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------


Best regards,
Hyeonjun, Seo

qligier #

unread,
Jun 29, 2021, 12:01:33 PM6/29/21
to IHE MHD Implementors
Hi,

One of FHIR AuditEvent examples (https://www.hl7.org/fhir/audit-event-example-media.json.html) uses urn:uuid:a54d6aa5-d40d-43f9-88c5-b4633d873bdd as system for this. Even though it's probably one of the last to consider, it's valid nonetheless.
Also, shouldn't IHE ask HL7 to register its systems in https://www.hl7.org/fhir/terminologies-systems.html?

Kind regards,
Quentin
Reply all
Reply to author
Forward
0 new messages