Diagnostic Imaging

CAP Comments on Diagnostic Imaging

Please see attached comments from the College of American Pathologists (CAP) on ASTP's request for feedback on Diagnostic Imaging data elements in draft USCDI v6. 

CAP Draft USCDI v6 Comments_0.pdf

Inclusion of Diagnostic Imaging Report in USCDI v6

Consider including Diagnostic Imaging Report in the USCDI v6.  The report is essential to clinical research.   

https://build.fhir.org/diagnosticreport.html  (Maturity level: 3)

Diagnostic Imaging - Imaging Links in USCDI V6

The Radiological Society of North America (RSNA) supports the proposed addition of imaging links, including Imaging Reference and Accession Number, to the USCDI V6, as specified in the ASTP-ONC’s HTI-2 proposed final rule. The addition of these links will advance the critical goal of enabling secure, convenient access to medical imaging for patients and all providers of their healthcare team by enabling closer integration of medical imaging with a patient’s health records, including the Diagnostic Imaging Report, which is already a specified component of USCDI. These links will eliminate wasteful duplication of diagnostic imaging tests and fulfill patient expectation that their providers have access to all diagnostic imaging test results, which includes both the diagnostic imaging report and the images themselves.

Effective deployment of Imaging Reference and Accession Number will require specifying appropriate interoperability standards, such as the HL7 FHIR ImagingStudy resource and DICOM WADO-RS. Additionally, a module for diagnostic imaging as part of the ASTP-ONC HIT Certification program would allow more detailed specification and testing of workflow in linking EHRs and imaging systems. We look forward to working with ASTP-ONC and other stakeholders to develop these important supporting elements to achieve secure, convenient access to medical images for patients and providers.

NCQA Recommendations for USCDI v6- Diagnostic Imaging

Diagnostic Imaging Test:

Recommendation type: Modification to existing USCDI element.

Recommendation: Update vocabulary standards to include CPT, SNOMED CT, and ICD-PCS in addition to LOINC.

Rationale: These represent vocabulary standards routinely used to capture diagnostic imaging tests, such as CT scans and mammography. This data element should reflect the alignment occurring in the health IT space. 

Diagnostic Imaging Report

Recommendation type: Modification to existing USCDI element.

Recommendation: The current USCDI element includes both the structured and unstructured components of the report. We recommend adding RadLex and SNOMED CT as appropriate vocabulary standards representing the clinical findings from the structured components of an imaging report.

Rationale: The clinical conclusions or findings resulting from a diagnostic imaging study represent important information to be exchanged via standard terminology to support appropriate follow-up care and care coordination. NCQA is developing measures that require the findings from imaging reports, specifically for mammography and CTs for breast and lung cancer screenings, which routinely represent the clinical findings using the ACR Reporting and Data Systems (RADS).  Adding SNOMED CT and RadLex vocabulary standards to the USCDI element will support interoperable data exchange of these important clinical data.

ACR USCDI Draft V5 Comments - Diagnostic Imaging

The ACR continues its recommendation that ONC elevate USCDI Diagnostic Imaging Level 2 elements Accession Number, Imaging Reference, and Requested Procedure Identifier to USCDI v5. While the report availability is appreciated, the images themselves are critical to patient outcomes in many critical situations.  Even though the images themselves are stored in external systems such as PACS or VNAs, the EHR must store references to enable the continuum of care both within a healthcare system and across systems since patients inevitably transition institutions both by choice and in critical care situations. It is especially important during these critical care situations such as mass casualty or emergency transfer to facilities with acute care capabilities, where seconds matter and having imaging prior to arrival can be crucial in a patient’s outcome.  The Level 2 elements for Diagnostic Imaging provide these references and make immediate access possible both with a facility and outside the enterprise.

DICOMweb and IHE Web Based Image Access (WIA) are clear mechanisms that can be leveraged within a facility.  The addition of capabilities that have been specified and tested as part of the Argonaut Project SMART Imaging Access project as well as IHE Internet User Authorization (IUA) and Cross-Enterprise User Assertion (XUA) profiles, extend DICOMweb and WIA outside the enterprise. IUA has been tested by at least 26, and XUA at least 73 vendors at IHE Connectathons based on a web search of the testing results. These include vendors which represent both EHR and PACS/VNA space.  These standardized, profiled, and tested capabilities enable cross enterprise sharing of imaging which can be further advanced by the elevation of USCDI Diagnostic Imaging Level 2 elements Accession Number, Imaging Reference, and Requested Procedure Identifier to USCDI v5, thus immediately improving patient outcomes.

NCQA Comment on Diagnostic Imaging Test: for USCDI v5

1. Diagnostic Imaging Test:

NCQA recommends updating the terminology to include CPT, SNOMED, ICD-PCS in addition to LOINC, which represent terminology routinely used to capture diagnostic imaging tests, such as CT scans and mammography.

ACR Comment to Elevate USCDI Diagnostic Imaging Level 2 Elements

The ACR recommends ONC elevate USCDI Diagnostic Imaging Level 2 elements Accession Number, Imaging Reference, and Requested Procedure Identifier to USCDI v5.  While the Diagnostic Imaging Report and Diagnostic Imaging Test have existed since USCDI v2, the ability to reference the relevant DICOM image files themselves is critical to many patient care scenarios where the Diagnostic Imaging Report itself would be inadequate; these include scenarios such as surgical planning and disease staging. In the common scenario of Emergency Room patient transfers immediate access to images from another institution can prove critical to patient outcomes. It is especially critical when no contemporaneous radiologist reporting is available at and for transfer from the originating institution which is the case not infrequently.  This would include the relevant diagnostic imaging data and the ability to identify and track unique health concerns of veterans of military service from within the DOD and VA systems as well as in external care locations by providing appropriate comparative exams for radiologist interpretation. Facilitation of transfers of care with source images from the originating institution facilitates access to higher level care benefiting patients in relatively underserved areas who would otherwise not be able to access such higher-level care. The ability to seamlessly exchange complete health information including the actual images and associated relevant metadata facilitates patient transfers in the context of load balancing during emergency situations such as mass casualty events or future pandemic conditions. Under those conditions, swift transfer of patients with as complete a record as feasible is a major contributing factor to successful crisis management. USCDI-TF-2021-Phase 3 recommended ONC should prioritize and encourage the advancement of DICOM Image File as a high priority data element, which the elevation of these Level 2 elements would accomplish.  Additional patient care scenarios where these elements are critical include the prevention of duplicate exams and unnecessary radiation events, as well as the unnecessary rescheduling of patient appointments due to missing or unreadable data.  Duplicate exams also drive unnecessary costs for our healthcare system.  For patient safety, the creator maintains the image source of truth so that as appropriate updates are made, they become immediately available to providers.  The Accession Number and Requested Procedure Identifier provide the necessary context to link the Diagnostic imaging Report to the DICOM Images, while the Imaging Reference provides the retrieval path to indicate network services that can be used to access the DICOM studies, series, or instances content and metadata. Requisite technology for touchless image exchange exists today and is readily commercially available at achievable cost. Expansion of USCDI as proposed would compel institutions to develop the supporting workflows to leverage image exchange technology successfully. As such, we believe that the (technical) implementation burden of image exchange is modest at best.

In support of the elevated maturity the following references are available:

• DICOM Part 18 WADO-RS was added in 2011 and is well supported by the vendor community. (https://dicom.nema.org/medical/dicom/2019a/output/chtml/part18/chapter_6.html)
• IHE Web-based Image Access was published in 2018 and has been tested at IHE Connectathons successfully since. (https://www.ihe.net/uploadedFiles/Documents/Radiology/IHE_RAD_Suppl_WIA_Rev1.1_TI_2018-03-22.pdf)
• The FHIR ImagingStudy resource has been elevated to Maturity Level 4 providing the foundation for Imaging Reference. ( https://build.fhir.org/imagingstudy.html)
• The management of the ImagingStudy and ServiceRequest resources is a currently proposed workitem for IHE. (https://wiki.ihe.net/index.php/Enabling_SWF_on_FHIR)
• The integration of FHIR resources and DICOMweb for imaging is a focus of the HL7 Imaging Integration group at FHIR Connectathons. (https://confluence.hl7.org/display/FHIR/2023+-+09+Imaging)
• The Imaging ServiceRequest has been profiled in multiple FHIR IGs extending the v2 to FHIR HL7 message mapping directly to Imaging. (https://fhir.ch/ig/ch-rad-order/index.html) (https://confluence.hl7.org/display/IMIN/Draft+IG+Project+Proposal+-+Imaging+ServiceRequest+Profile) (https://confluence.hl7.org/display/IMIN/Imaging+Service+Request+Hierarchy)
• The Argonaut SMART Imaging project provided a unified solution for accessing imaging studies alongside clinical data using a single authorization flow, thus proving across multiple vendor platforms the maturity of the objects and their ability to be supported across the internet securely alongside all the other resources currently specified in USCDI v1-v4. (https://confluence.hl7.org/display/AP/SMART+Imaging+Access)

Rural and underserved areas would benefit greatly as the likelihood of patient transfer to larger institutions for care is increased: making images available prior to the patient’s physical arrival can be crucial to the determination of necessity of transfer, better clinical outcomes through expedition of transfer, and eliminates the need for unnecessary duplicate exams.  To address any concerns about gaps in transmission infrastructure that may have been previously raised, the ACR would like to point out the significant ongoing federal and state investment in rural infrastructure which would directly support these needed data elements.  USDA ReConnect has already invested more than $1 billion dollars to date and continues to fund the expansion of high-speed broadband infrastructure in unserved rural areas and tribal lands.  The Biden administration in August 2023 announced $667 million in new funding to build more broadband internet access in the U.S. The grants and loans are spread across 22 states, but will target rural, remote, and unserved communities.  The American Rescue Plan Act earmarked $350 billion for a variety of services for state and local agencies, including expanded broadband access.  The Infrastructure Investment and Jobs Act put another $65 billion in broadband funding on the table, $42 billion of which will be in the form of grants to states. The Rural Broadband Accountability Plan is a new effort by the FCC to monitor and ensure compliance for universal service high-cost programs including the Rural Digital Opportunity Fund and Connect America Fund Phase II Auction. The FCC’s high-cost universal service program is approving billions of new investment dollars to bring connectivity to consumers in areas lacking access to broadband.  All these investments in rural technology support the needs of these underserved areas when this technology is leveraged to provide DICOM Images as USCDI-TF-2021-Phase 3 recommended to ONC in 2021 and is supported by the technology advances of FHIR and DICOM, proven through IHE, HL7 and the Argonaut project efforts.  While these investments may take time to establish more equitable access to broadband transmission infrastructure, they are underway and the technology to leverage it exists.  Patients and their institutions who already have this in place should be enabled to reap the benefits and not be blocked by those using infrastructure as an excuse. Every day which passes by where we do not upgrade these USCDI data elements is a day when the benefits outlined in this comment text accrue to a more limited group of patients than is achievable and desirable.

CDC Unified Comment: California Department of Public Health

The California Department of Public Health recommends adding the following data elements for the Patient Profile in exchange between EHRs:

• Order Date (found in Orders or Imaging section) 

• Report Date (if available, would be found in one of these sections: Orders, Results, Imaging) 

Clinovations GovHealth USCDI Draft V2: Diagnostic Imaging

Consider future expansion to support automated adherence to clinical guidelines

• In finalizing the forthcoming version of USCDI, please clarify the difference in Diagnostic Imaging Report and Diagnostic Imaging Narrative. Both appear to contain the radiologist’s interpretation of the data and use the same LOINC code. Please also clarify the difference between these categories and the Diagnostic Studies and Results class added in Level 2.
• To support providers in following clinical guideline timelines, we raise the consideration of future expansion to include the date for orders and procedures and normal/abnormal value results. Knowing the study date, exam, date, and read date, for example, could facilitate automated tracking and follow-up to reduce provider burden. Knowing normal and abnormal value data could facilitate and prioritize follow-up.

Cerner Corporation USCDI Draft V2 Comments - Diagnostic Imaging

Provided below are Cerner's comments on the USCDI draft V2 proposals for the Diagnostic Imaging data class. Cerner's full public comments on the USCDI draft V2 have been posted on the USCDI general comments section.

Cerner appreciates and supports the proposal to add the new Diagnostic Imaging data class in the USCDI V2. The exchange of diagnostic imaging data is important to facilitate the best possible patient care and is mature enough within the industry for inclusion in the USCDI. However, we believe there are some important changes that must be made to the underlying data element proposals before adopting.

First and foremost, while we agree that the Diagnostic Imaging Narrative data element should be removed from the Clinical Notes data class as the content does not fit the intent of that data class, it represents a duplication of the newly proposed Diagnostic Imaging Report data element. As defined, the Diagnostic Imaging Report data element would actually encompass the Diagnostic Imaging Narrative (i.e., the narrative interpretation of the image). Furthermore, the standalone Diagnostic Imaging Narrative is limited in its value and usability without the additional supporting information from the full imaging report. Therefore, the Diagnostic Imaging Narrative data element should be eliminated from the Diagnostic Imaging data class in favor of the more complete Diagnostic Imaging Report data element.

Regarding the Diagnostic Imaging Report data element, there needs to be more clarity on the specific scope of a report. Most importantly, clarification is needed that the report does not include the diagnostic image itself. While in some cases the full report may contain replications or components of the diagnostic image (e.g., a .PDF report containing references to the image pointing out particular components for the interpretation), the diagnostic image itself is independent. Additionally, HL7 FHIR US Core does not currently provide a Profile supporting diagnostic images (only the reports), so it would be inconsistent with ONC’s stated intent with the USCDI V2 to include the actual diagnostic image in scope.

The proposal to adopt LOINC as the standard for the Diagnostic Imaging Report data element also appears problematic considering that the full report would consist of both structured and unstructured components. We would recommend that the standard be clarified as specific to distinct components of the full report to be clear that unstructured content is not expected to be codified by LOINC. Additionally, we note that if the re-classification and replacement of the Diagnostic Imaging Narrative data element is finalized in USCDI V2, ONC should include the same changes in USCDI V1 via an errata adoption. This is important for consistency as HIT developers and healthcare providers are currently working to align with USCDI V1. If changes other than additions are made for USCDI V2 it may create inconsistencies for exchange of the same data between those versions.

Finally, while we also support inclusion of the Diagnostic Imaging Order element to represent the service request from which the Diagnostic Imaging Report originated, there is a need to ensure that the definition sufficiently distinguishes the data element from a procedure that may also be associated with the Diagnostic Imaging Report.