Visualizing Harmony Pipelines

Harmony is designed to be the flexible glue between your healthcare data systems. Whether you're routing DICOM images, proxying FHIR requests, or transforming JSON payloads, pipelines are at the core of how data moves through Harmony.

To help visualize these flows, we've introduced a new diagramming system. Let's explore some common Harmony patterns using these new visualizations.

DICOM SCP Pipeline

One of the most common use cases for Harmony is acting as a DICOM Service Class Provider (SCP). In this configuration, Harmony accepts incoming DICOM associations (like C-STORE or C-FIND) and routes them to storage or other systems.

Here is a visualization of a standard DICOM SCP workflow where images are received, validated, and stored to disk:

In this example, the Modality initiates the connection. Harmony accepts the association via its DICOM Listener, validates the AE Title, and then streams the received DICOM files to the Storage backend.

FHIR Proxy with Authentication

Modern healthcare interoperability often relies on FHIR, but legacy systems might not support modern authentication standards. Harmony can act as a secure gateway, adding authentication layers to existing FHIR servers.

The pipeline above demonstrates a request where the Auth Guard middleware first validates the credentials. Only successful requests are passed to the Harmony Proxy core, which forwards them to the upstream Firely Server.

JOLT Transformation Pipeline

Data rarely arrives in the exact format you need. Harmony includes a powerful transformation engine based on JOLT (JSON of Large Transform), allowing you to restructure payloads on the fly without writing custom code.

This pipeline takes a Raw Input payload, passes it through the JOLT Engine to apply a transformation spec (like patient_to_fhir.json), performs final Normalization, and delivers a compliant FHIR Bundle to the destination.

Building Your Own

These diagrams are just a glimpse of what's possible with Harmony. By chaining middleware and configuring flexible backends, you can build complex integration pipelines that solve real-world interoperability challenges.

Check out the examples directory in our repository to try these configurations yourself.