Eyefinity & ModMed EHR Integration | Merry AI

A Documentation Workflow Built for Eyefinity / ModMed

Clinical documentation inside Eyefinity / ModMed rarely fails because clinicians lack diligence. It fails because the physical act of ophthalmic examination is fundamentally incompatible with keyboard-and-mouse data entry. A clinician seated at the slit lamp, both hands on the joystick and focus wheel, cannot simultaneously click through the OD cornea field, toggle laterality, and type an intraocular pressure reading into a flowsheet grid. The result is a documentation debt paid later, from memory, at the end of a long clinic day.

Eyefinity and ModMed EMA were architected around structured, eye-specific grids precisely because ophthalmic care demands discrete data. Eyefinity's Simple Eye Exam workflow exposes an Exam Controls bar for laterality, dilation, slit lamp, discs, and fundus, with Eye Exam Details screens that break each anterior segment into its own clickable region. ModMed EMA embeds MIPS quality capture directly inside those same structured fields. This structural rigor is exactly why Merry AI's discrete injection model aligns so cleanly with how these platforms were designed to be used.

Merry AI closes the gap between spoken examination and structured chart entry. As you narrate findings at the slit lamp, our browser-native extension resolves the clinical entities, laterality, and measurement method, then writes discrete values directly into the native exam grids. No switching from the slit lamp to the keyboard. No navigating across segments to click OD and OS fields. No copying values from a narrative note into an IOP flowsheet after the fact.

Why Closed Ophthalmic EMRs Resist Traditional Integration

Both Eyefinity and ModMed operate as largely closed clinical ecosystems. While each advertises equipment interfaces and diagnostic image imports, neither exposes an open, granular FHIR write API that a third party can use to populate individual slit-lamp segment fields or push a numeric IOP reading with method and laterality into the proprietary grid. Claiming otherwise would be technically dishonest. The reality of ophthalmic EMR integration is that the richest, most codable fields live behind vendor-controlled interfaces that were never designed for external structured writes.

This constraint has historically forced clinicians into one of two poor choices: pay for a marketplace integration that only handles coarse-grained data and lags behind the exam in real time, or fall back to manual entry and copy-paste from a separate transcription tool. Both approaches reintroduce the exact keyboard-tethered friction that ophthalmic workflows cannot tolerate.

The DOM Injection Architecture

Merry AI takes a different engineering path entirely. Our Chrome Extension operates inside the clinician's already-authenticated Eyefinity or ModMed browser session. Rather than negotiating a server-to-server API handshake, the extension identifies the DOM input selectors that render the exam grids, IOP flowsheet rows, and Eye Exam Details fields, then writes recognized values directly into those input nodes. To the EHR, the data arrives through the same input elements a technician would populate by hand, so all downstream logic, coding, problem lists, and MIPS calculations treat it as native structured data.

Field-Level Mapping and Laterality

The precision of this model depends on rigorous field mapping. When you say "trace nuclear sclerosis OU," the system resolves the segment (lens), the finding, and bilateral laterality, then populates both the OD and OS lens fields. When you say "pressure 16 OD, 18 OS by Goldmann," the extension writes the numeric values into the correct IOP grid rows while preserving method and eye side. Internally these observations are represented with structured semantics, valueQuantity in mmHg, bodySite for laterality, method for the tonometry technique, so the vendor-specific injection layer maps cleanly to each platform's proprietary field names.

Integration Architecture Comparison

The distinction between approaches becomes clear when you evaluate deployment overhead, real-time performance, and cost across the full lifecycle of an integration.

Because the extension inherits the clinician's existing session, there is no separate authentication surface, no marketplace onboarding queue, and no API rate limit throttling the flow of exam data. Deployment collapses from weeks to minutes. You can Access 1,000+ EHR Clinical Templates and be documenting a glaucoma follow-up in the same session you install.

Regulatory Synergy: MIPS, G2211, and Info Blocking

Ophthalmic quality reporting depends heavily on structured data that documentation fatigue routinely leaves incomplete. Glaucoma and ocular hypertension measures require IOP values, optic nerve C:D ratios, and visual field results. Because Merry AI injects these into the same native fields ModMed EMA uses for MIPS capture, our workflow enhances measure completeness rather than bypassing it. No extra clicks, no separate re-entry, no missing data due to end-of-day recall.

Capturing G2211 for Longitudinal Care

Ophthalmology and optometry are among the specialties most defined by continuity, glaucoma, macular degeneration, and diabetic retinopathy are longitudinal conditions requiring ongoing focal management. HCPCS add-on code G2211 exists precisely to recognize this visit complexity when reported alongside office and outpatient E/M codes 99202 through 99215. Merry AI surfaces documentation supporting the continuing-focal-point rationale so eligible visits capture the complexity add-on appropriately. Clinicians should always validate coding against current AMA CPT Evaluation and Management guidance before submission.

Info Blocking and Data Portability

Because Merry AI writes discrete data into the EHR's own certified fields rather than trapping it in a proprietary side channel, the resulting record remains fully accessible for patient export and interoperability under the Cures Act info-blocking provisions. Slit-lamp findings and IOP values are stored as the standard structured data the platform already supports, reinforcing existing regulatory infrastructure instead of undermining it. Ready to see it in your own environment? Schedule a 15-Minute Workflow Audit.

Security, Attribution, and the Author of Record

Ambient capture inevitably touches PHI, so security governance is not an afterthought in our architecture. Under executed Business Associate Agreements, Merry AI retains zero audio and zero transcript after the session concludes. Transport and storage are encrypted, and because injection happens inside the clinician's authenticated session, the entire record stays within the EHR's native access-control and audit-trail framework. There is no intermediate transcription database accumulating sensitive ophthalmic data outside the vendor's boundary.

The clinician remains the author of record at every step. Merry AI produces a structured draft, but nothing commits to the chart until you review the populated grids, correct any values, and sign off. This preserves both clinical integrity and the audit attribution that quality programs and legal defensibility require. The AI accelerates the mechanics of documentation; it never replaces clinical judgment or authorship.

For practices evaluating scale, the economics favor the DOM injection model decisively. There is no marketplace revenue share, no per-integration API fee, and no dependency on vendor roadmap timelines to expose new endpoints. When Eyefinity or ModMed updates a screen, we update our selector mapping, and clinicians keep documenting. To align a rollout with your specialty mix and payer program obligations, Explore Merry AI Practice Partner Plans and bring your existing exam workflow to the audit.