XDR-I - Detailed Proposal

1. Proposed Workitem: XDR-I

• Proposal Editor: Chris Lindop
• Editor: Chris Lindop
• Date: N/A (Wiki keeps history)
• Version: N/A (Wiki keeps history)
• Domain: Radiology

2. Summary

Currently, Imaging has no payload which would allow for the push of images outside of an enterprise environment. IHE ITI has a transport protocol called XDR which serves well for non-imaging payloads. It is compatible with the Security and Privacy protocols developed by the IHE ITI domain. The XDS-I.b protocol breaks the content profile into 2 components. The first component is designed essential as a notification of where to pull the content. Usage of the XDR transport protocol should not require this notification/pull architecture, but just the push model. It should remove the requirement for a source to be a cross enterprise source of Images.

The value for this capability would allow for a less sophisticated system to be a source of images. An XDR-I image source could be as low cost as a workstation. Additional value could be to use the XDR-I web-services as an XDS-I.b proxy with a regional XDS-I.b source. Currently, when an XDS-I.b Image Document Source provides & register images on behalf of a local facility, there is no mechanism to provide feedback to the originating source. This would allow for an XDS-I.b Image document source to use the response of the XDR transport services to provide success or failure status back to the originating source of the images.

3. Key Use Case

The potential use cases can be fairly robust.

1) A Clinical outpatient facility providing images to a shared image repository.

This is the case where a clinical facility with diagnostic imaging systems would like to provide an acquired image set to a shared image repository via reliable web-services messaging.

2 Sending images and other clinical documents to a referring clinician without the use of CD media.

In this case, a clinical facility with diagnostic imaging systems may have additional documents (a Radiology Report, for example)and would like to provide the images and documents to a a referring physician via reliable web-services messaging.

3 Connecting a Hospital or Clinical practice PACS to a Regional Repository that is an XDS-I.b Image Document Source.

As in the Canadian National Health Infoway Blueprint and other regional projects around the globe, the hospitals and other clinical institutions share a regional image repository that also functions as an XDS-I.b Image Document Source. The source which acquires the images, provides the images to the regional image repository with a reliable message transaction. The regional image repository would provide and register the manifest to an XDS.b Registry/Repository. As part of the response back to the source. the regional repository would include the XDS-I.b transaction results.

4 Submitting images for a clinical trial

A clinical facility enrolled in a clinical trial would transfer the images and additional clinical information using reliable messaging.

5 Submitting Dose reports to a National registry.

A clinical facility enrolled in a National Dose Registry would transfer the dose reports and additional clinical information using reliable messaging.

4. Standards and Systems

Existing Standards

1. ITI XDR
2. RAD XDS-I.b
3. RAD REM
4. RAD TCE

Existing Systems

1. Workstations
2. Acquisition Modalities
3. PACS and Enterprise Archives
4. Regional PACS and Archives
5. Regional and Enterprise RIS
6. Dose and Clinical Trial Registries

5. Technical Approach

Existing actors

Image Document Source

Similar to XDM, keep the existing actor, "Image Document Source" as the source of the image document set.

New actors

Image Document Recipient

Similar to XDM, add a new actor, "Image Document Recipient" as the recipient of the image document set.

Existing transactions

RAD-68 Provide & Register Imaging Document Set

This transaction will be modified in an equivalent manner as to how XDM specifies the ITI-41 Provide and register transaction. The transaction will be modified to include the DICOM images as part of the payload for this Transaction. The payload should include the manifest. However, to provide the maximum flexibility of this profile, consideration does need to identify conditions where the manifest may be optional as part of the payload. For instance, if this transaction is reused with TCE, the KOS object should use the appropriate KOS manifest specified in that profile. In cases where a manifest is not specified like RAD-63, the manifest could be optional.

New transactions (standards used)

None.

Impact on existing integration profiles

Other profiles which could benefit for an XDM-I transaction specifically include TCE and REM where DICOM instances are exported.

This transaction could be re-used with REM for the RAD-63, Submit Dose Information transaction.

This transaction could also be re-used with the TCE Integration Profile for the RAD-53 Export Instances.

New integration profiles needed

New profile is XDR-I.

Breakdown of tasks that need to be accomplished

1. Update Volume 1 with the XDM-I Profile and use cases
2. Change RAD-68 to include the DICOM Instances for XDR-I
3. Evaluate Manifest as an option
4. Evaluate impact to other profiles eq REM & TCE
5. review the patient reconciliation XDR and make sure it works with us
6. Review the potential configuration of multiple destinations.
7. Evaluate to use as a push bridge between domains (to parallel XCA-I pull bridge)

6. Support & Resources

Two key groups who have expressed interest in this proposal is the Canadian SCWG-5 and DICOM WG-27. Additionally, there are several regional deployment groups, including IHE Australia, who have expressed interest. It is potentially an important component to the RANZCR CT Dose project.

7. Risks

Very low risk technically. This profile meets a known gap in our transport capability of DICOM instances using web services.

8. Open Issues

Questions:

1. What would the number of Manifests to be included per transaction?
2. In certain cases, do we allow the manifest to be optional or inversely, required?
3. Is there a need to specify an XDM-I profile? Or is the guidance sufficiently adequate for an XDM to include the PDI profile as the DICOM instance content? Would this be out of scope for this profile?

9. Tech Cmte Evaluation

Effort Evaluation (as a % of Tech Cmte Bandwidth):

The proposed profile is based on the existing XDR profile. Expected effort is anticipated as editorial.

However, there is additional evaluation of the open items as specified.

• 20% with good preperation

Responses to Issues:

See italics in Risk and Open Issue sections

Candidate Editor:

Chris Lindop