Getting Started Guide

A step-by-step walkthrough for standing up your first instrument interface in Vorian Connect — from connecting an analyzer to routing live results to your LIS. Written for lab operators and interface engineers.

What Vorian Connect does

Vorian Connect sits between your lab instruments and your LIS (Epic Beaker, Cerner, or another). It captures each analyzer's real messages, shows you where those messages diverge from the vendor's spec, lets you validate the interface with evidence before anything goes live, and then routes results to your LIS — with autoverification rules, QC checks, and a full evidence trail.

You can run it alongside your existing middleware (like Data Innovations) in shadow mode, so nothing touches live patient results until you say so.

The core concept: Shadow Mode

When an instrument is in Shadow Mode, Vorian ingests and analyzes its messages but does not deliver results anywhere. That means you can point a live analyzer at Vorian, watch everything it does, prove the interface is correct — all with zero risk to patient care or to your existing middleware setup. When you're satisfied, you flip that instrument from Shadow to Live. Nothing is irreversible until then.

The end-to-end workflow

Step 1 — Connect your first instrument. Go to Sources and add the analyzer you want to interface. This tells Vorian's on-prem Edge Agent to listen for that instrument's messages over MLLP, the standard analyzer wire protocol. Start the instrument in Shadow Mode and leave it there for the whole setup.
Step 2 — Confirm it's actually talking. Go to Topology and find the instrument. Its Last Heartbeat should be under ~60 seconds. If it's Offline or stale, check that the instrument's outbound IP:port matches your Edge Agent and that the agent is running: docker logs vorian-agent --tail 20.
Step 3 — Watch the real traffic. As the analyzer runs, its messages flow into the Live Feed on the home dashboard. Open any message in the Inspector to see exactly what the instrument sent, field by field — PHI-safe. This is the raw truth of what your analyzer emits.
Step 4 — Check the spec against reality. In Interface Builder, compare the instrument's real messages against its vendor implementation guide. Vorian flags every place they diverge — required fields the instrument leaves empty, values landing in the wrong field, undocumented segments it sends, documented segments it never sends. These are the mismatches that break go-lives. Resolve them here, before you're live.
Step 5 — Clear the Unmapped queue. If results show up in Unmapped Results but not in the Live Feed, the instrument's test codes need mapping. Map codes individually or bulk-map with a CSV upload via ML → Catalog Import — upload the vendor's test catalog or fill in instrument-code to LOINC mappings.
Step 6 — Add autoverification and QC. Once results flow cleanly, layer in Rules (autoverification with comment, flag, notification, and result logic), Delta Checks, and QC. Every rule is inspectable and validated before it activates — nothing changes a result silently. You can skip this on day one and add it as you go.
Step 7 — Produce the proof. Go to Validation to generate the evidence packet: what was examined, the spec-vs-reality findings, the honest limitations, and a PHI-safe report you can hand to compliance or an inspector. This is the artifact that proves the interface is sound.
Step 8 — Go Live. When the interface is validated and you're satisfied, return to Sources, find the instrument, and switch it from Shadow Mode to Live. Now results deliver for real. Configure where they go in Delivery Routing and Destinations (your LIS / Beaker), including STAT priority, failover, and load-balancing.
Step 9 — Operate day-to-day. Watch the Live Feed, review Delivery Routing for what's delivering or blocked, work exceptions in the Worklist, and monitor system health in Observability.

Getting help

Most operational pages have a Help panel (the ? icon) with step-by-step runbooks for that page's common issues. The in-product help covers: Live Feed, QC Evidence, Topology, Rules, Unmapped Results, Migration Mode, Destinations, Sources, Interface Builder, Validation Evidence, Delivery Routing, and the Message Inspector.

For anything the runbooks don't cover, open a case under Support → Cases.

The 30-second version

Connect the instrument in Shadow Mode → confirm the heartbeat → watch its real traffic → check spec-vs-reality in the Interface Builder → clear the Unmapped queue → generate the validation proof → flip Shadow to Live → route to your LIS. Nothing touches live results until Step 8.

Last updated: July 29, 2026