Screen

Screen-based capture may help us break the EMR log jam.

Most enterprise systems remain closed environments. While they appear open, anyone who has switched from one EHR to another knows it usually requires us to manually copy data from one system to the next—a process that makes little sense in this day and age.

Take Epic, for example. It’s like a walled garden—except the garden can feel like a jungle, filled with wildly inconsistent experiences based on each client’s unique configuration.

As we design the next generation of Iguana, our focus is shifting. We’re moving past the debate of “HL7 vs APIs” and are now looking at ways to interact with Epic directly.

Our true requirement is simple: we need access to the information that users are seeing and entering. If we harness screen capture technology, analyze it with a local AI engine, and safely simulate user interactions, we believe we can engage with the actual workflow—rather than guessing what’s happening through partial HL7 feeds, fragile interfaces, or limited APIs.

Anyone who’s dealt with HL7, FHIR, or an Epic integration knows it’s like playing pin the tail on the donkey while blindfolded. The standards sound open, but in practice, crucial information is often obscured by local configuration choices, licensing, workflows, and institutional politics.

That’s why we believe screen-based workflow automation offers better longevity. Although it’s not a trivial engineering challenge, this approach brings us closer to the real source of truth: what users actually see, enter, and need to act on.