trove-fhir-viewer

KO | EN

gitlite — search

#python #react #tailwindcss #reactjs #api #android #mysql #rust #php #docker #typescript #unity

trove-fhir-viewer

★ 26 Open GitHub ↗

In-browser HL7 FHIR Bundle viewer. Read, diff, and quality-check FHIR R4/STU3/R5 bundles entirely client-side — zero server, zero login, zero PHI ever leaves your device.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Trove FHIR Viewer

> **Read any FHIR bundle in your browser. Zero upload. Zero login. Zero PHI leaves your device.**

[![License: Apache 2.0](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
[![Live demo](https://img.shields.io/badge/demo-fhir.trovehealth.io-419958?style=for-the-badge)](https://fhir.trovehealth.io)
[![Built by Trove](https://img.shields.io/badge/built_by-Trove_Health-1b3954)](https://www.trovehealth.io)
[![LinkedIn](https://img.shields.io/badge/LinkedIn-Trove_Health-0a66c2)](https://www.linkedin.com/company/trovehealth)

---

## 🔗 Try it live → **[fhir.trovehealth.io](https://fhir.trovehealth.io)**

No install. No account. Drop a FHIR bundle into the page and you're reading it in two seconds.

---

![Trove FHIR Viewer preview](https://raw.githubusercontent.com/trovehealth/trove-fhir-viewer/main/assets/og-image.png)

A free, in-browser HL7 FHIR Bundle viewer. The kind of tool you can open on an air-gapped laptop and still trust with production PHI.

## Why this exists

Every other "FHIR viewer" we tried was a FHIR *server* with a UI bolted on — HAPI, Medplum, Aidbox, Smile CDR. All serious tools, all of them requiring you to upload your bundle to their server before you can look at it. For most providers, payers, and converter teams, that alone makes them unusable for real production data.

So we built one that parses the bundle entirely in the browser. Drop a 20 MB Synthea-style bundle in — it parses in a Web Worker without freezing the page. Nothing crosses a network boundary.

## What's in the box

- **In-browser parsing** of FHIR R4, STU3, and R5 bundles. No backend.
- **Patient Summary** with KPIs, smart highlights (severe allergies, abnormal labs, conformance gaps), and an encounter timeline.
- **Patient Story** — a multi-lane timeline of every dated event in the bundle, with zoom and chip filters.
- **Charts** — auto-generated views for bundle composition, encounter timelines, and lab trends.
- **Friendly view** for every resource — grouped property cards driven by a per-type spec, with a smart auto-grouper for uncurated types.
- **JSON view** — code-editor with line numbers, fold triangles, syntax highlighting, and a Copy button.
- **Bundle diff** — drop two bundles in, get a side-by-side diff with two-pass natural-key matching, run in a Web Worker so big bundles don't freeze.
- **CCDA section tracking** via `meta.tag` LOINC codes — pivot a bundle by "show me everything from the Problem List section."
- **Data Quality dashboard** — missing required fields, unresolved references, junk codes, date sanity. Built for CCDA→FHIR converter QA.
- **Mobile responsive**. Sidebar collapses to a slide-in drawer; tables reflow.
- **Light and dark mode.**
- **Document-wide search** that surfaces matches across every resource type.
- **CSV export** per resource type.
- **Zero telemetry.** We don't know who's using it, and that's the point.

## Quick start

The viewer is a single HTML file with no build step.

### Use the hosted version

Just open **[fhir.trovehealth.io](https://fhir.trovehealth.io)**. Drop a bundle in.

### Run it locally

```bash
git clone https://github.com/trovehealth/trove-fhir-viewer.git
cd fhir-viewer
open index.html        # macOS
xdg-open index.html    # Linux
```

No `npm install`, no build, no server. If you want it served (e.g. for `file://` security reasons in Chrome):

```bash
python3 -m http.server 8000
# Visit http://localhost:8000/
```

### Self-host on your own domain

There's nothing to compile and no environment variables to set. Drop `index.html` plus the `assets/` folder onto any static host:

- **GitHub Pages** — fork this repo, enable Pages on `main` / `/`.
- **Cloudflare Pages / Netlify / Vercel** — drag-and-drop the folder into the dashboard.
- **S3 + CloudFront** — `aws s3 sync . s3://your-bucket/` then point CloudFront at it.
- **nginx / Apache** — `cp -r . /var/www/your-site/`.

The `<head>` references `assets/og-image.png` and `assets/favicon.svg` relatively, so any host that serves the folder as a directory works.

## How it compares

| | Trove FHIR Viewer | HAPI FHIR server | Medplum / Aidbox | Inferno |
|---|---|---|---|---|
| Free | ✅ | ✅ | Freemium | ✅ |
| Server required | ❌ | ✅ | ✅ | ✅ |
| PHI leaves device | Never | Yes | Yes | Yes |
| Login required | ❌ | Optional | ✅ | ✅ |
| Bundle diff | ✅ | ❌ | Partial | ❌ |
| Data quality view | ✅ | ❌ | ❌ | (conformance only) |
| CCDA section tracking | ✅ | ❌ | ❌ | ❌ |
| Built for | Reading | Storing | Storing + clinical apps | Conformance testing |

None of those projects are bad — they just solve different problems. If you need to *store* FHIR, use HAPI or Aidbox. If you need to *certify* conformance, use Inferno. If you need to *read* a bundle quickly and safely, use this.

## Project layout

```
fhir-viewer/
├── index.html         # The entire application — single file
├── assets/            # Static binary assets
│   ├── og-image.png   # Social-card preview (1200×630)
│   ├── og-image.svg   # Vector source for the preview
│   ├── favicon.svg
│   ├── favicon-32.png
│   └── apple-touch-icon.png
├── docs/
│   └── architecture.md  # Tour of the single-file source
├── .github/           # Issue and PR templates
├── LICENSE            # Apache 2.0
├── NOTICE             # Third-party attribution + trademark notice
└── SECURITY.md        # How to report vulnerabilities
```

The whole application is `index.html`. ~440 KB at v12. Vanilla JavaScript, no build, no bundler, no framework. Tailwind CSS is loaded from CDN at runtime; the FHIR parser, charts, diff engine, code-editor view, and grouped property-card renderer are all hand-written and live in that one file.

If you're contributing, see [docs/architecture.md](docs/architecture.md) for a tour.

## Contributing

Bug reports and pull requests are welcome — see the issue templates in [`.github/`](.github/).

Particularly valuable contributions right now:

- Edge-case bundles that break the parser
- Friendly-view specs for resource types we haven't curated yet (we have ~23; FHIR R4 has ~150)
- Performance work on the bundle-diff worker for bundles > 50 MB

By submitting a pull request you agree to license your contribution under [Apache 2.0](LICENSE), the same as the rest of the project.

## A note on the brand

This project is open source under Apache 2.0 — you can fork it, modify it, and ship it. But the "Trove" name and atom-cluster logo are trademarks of [Trove Health Inc.](https://www.trovehealth.io/) The trademark itself is **not** licensed.

If you fork and distribute, please use your own brand. You're encouraged to attribute the original (`Built with the Trove FHIR Viewer`), but please don't ship a fork that calls itself "Trove" or uses the logo.

---

## Built by Trove Health

[Trove Health](https://www.trovehealth.io/) builds tools for healthcare data interoperability. The FHIR Viewer is one of several open-source artifacts we release in the spirit of "show, don't tell" — if you like how this tool feels, our commercial products extend the same approach to data ingestion, harmonization, and longitudinal records.

If you're at a payer, provider, or digital-health company wrestling with FHIR + CCDA at scale, we'd love to talk.

See also our companion [Trove CCDA Viewer](https://github.com/trovehealth/trove-ccda-viewer) for HL7 C-CDA documents.

## License

Apache License, Version 2.0. See [LICENSE](LICENSE) for the full text.

trove-fhir-viewer

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README