Imaging Interoperability Is Now Policy. Is Your Archive Ready to Share?

Federal policy has started moving medical images from a siloed asset toward shareable, standardized data. The health systems that feel it first are the ones whose images sit inside a proprietary archive.

For years, imaging interoperability was a goal that vendors talked about and few buyers pressed on. That has changed, and it changed because of policy, not marketing. As images and their reports become expected data elements across health information networks, the question every imaging leader faces is simple: can your archive produce and share that data on open standards, or does every exchange turn into a project with a fee attached?

What actually changed

On January 30, 2026, the federal health IT office, ASTP/ONC, issued a request for information on diagnostic imaging interoperability standards and certification. Comments closed on March 16, and the American Hospital Association filed a formal response the same day. The office has said the responses will inform potential future rulemaking, so nothing is binding yet.

What is settled is the data standard. On July 23, 2026, ASTP/ONC published version 7 of the United States Core Data for Interoperability. Standards Bulletin 2026-2 describes thirty new data elements plus one significantly revised element. Among the additions is Diagnostic Imaging Reference, described as the information that can be used to access a diagnostic imaging study, including an imaging study endpoint weblink, unique identifiers, and the contextual information needed to retrieve the study.

That is smaller than a headline makes it sound, and larger than it looks. A reference is a pointer. It does not move an image and it does not compel anyone to hand one over. What it does is establish that a diagnostic imaging study is expected to be reachable, and that where the study lives is now part of the standard record rather than a local implementation detail. A pointer is only worth what answers on the other end of it.

Why this lands hardest on proprietary archives

If your images live in a system that only speaks its own language, interoperability becomes a recurring cost. Every new exchange partner, every network you join, every AI tool you want to feed becomes a custom interface and an invoice. The organizations that will adapt quickly are the ones whose imaging data is already portable and already speaks open standards.

This is the practical case for data independence. It is not a slogan; it is a hedge against exactly the kind of policy shift now underway. When a patient record carries a reference to one of your studies, something has to answer at that endpoint. A vendor-neutral archive that supports DICOM and DICOMweb services, runs a configurable HL7 engine for orders and results, and handles patient matching through PIX and PDQ keeps your data ready to participate rather than trapped behind a gateway you do not control.

What makes that possible is the state of the data itself, not the software sitting in front of it. Imaging data at rest should be DICOM, not a proprietary derivative of it, and it should be current, so that the archive is the source of truth rather than a copy that has to be reconciled with an application before anyone can use it. Data entangled with the application that wrote it is only as available as that application chooses to be.

The harder test is whether your archive has to be in the path at all. Orchestration, normalization, aggregation and transformation are real work, and it is the work we do. But when a requester wants direct access on open standards and the archive itself can serve it, no intermediary should be the bottleneck. Managed cloud imaging services are starting to make the stored data the hub rather than the endpoint, and a neutral archive should be able to work alongside that model instead of competing with it for control of access.

Our take

We have been vendor-neutral since long before it was policy, so it would be easy for us to treat this as a victory lap. It is not one. A reference element within a data standard is a small, sensible step, and the hard part is still ahead for everyone.

Here is what we expect to happen, and what we would rather did not. Organizations will treat the reference as a field to populate. A pointer is written into the record, the pointer resolves to a gateway, and the gateway asks the requester for an agreement, a project, or a fee. The record appears interoperable and the image is still hard to obtain. It is worth naming the limit plainly: a reference does not create a right of access, and getting an image still depends on agreement between the organization requesting it and the organization holding the study. What the standard changes is where the friction shows up. It moves from we cannot find your images to we can see exactly where your images are and cannot get them, and the second version is much harder to explain to a referring physician or to a patient.

If we had one ask of the rulemaking that follows the request for information, it would be this. Specify what a reference has to resolve to, not only what it has to contain. A pointer with no floor under it is a pointer to a negotiation. Until that floor exists, the protection is the one it has always been. Keep your imaging data in a form you can serve on open standards, to a requester you did not anticipate, without opening a project. That is what data independence actually buys.

What to do now

Inventory where your images actually live, and which systems can and cannot export them on open standards without a custom project.

Test an export. Ask your current archive to produce a study set over DICOMweb and see what it costs in time and fees.

Find out what answers a reference. If a record points at one of your studies, ask which system responds, on which standard, and whether a new requester can be added without a project and an invoice.

Price your data exit. Understand egress, retrieval, and interface charges before they show up during a migration or a network onboarding.

Prefer neutrality. When you evaluate the next system, weight open standards and documented portability as heavily as the viewer.

The takeaway

Interoperability rewards the open. As imaging becomes standardized, shareable data by policy, the value of a neutral, portable archive stops being theoretical and starts showing up in what you can do and what you no longer have to pay for. The safest position is the open one.

Sources: ASTP/ONC, Request for Information: Diagnostic Imaging Interoperability Standards and Certification, Federal Register, January 30, 2026. American Hospital Association, comment letter to ASTP/ONC, March 16, 2026. ASTP/ONC, United States Core Data for Interoperability version 7 and Standards Bulletin 2026-2, July 23, 2026.