Skip to content

Live demo: record viewer

← Back to docs home Open it full screen

Below is the record viewer from the core pipeline, loaded with an invented ALS cohort (Jane and John Doe shells, no real records) that the site build generates and runs through the pipeline. It runs entirely in your browser: no server, no upload, no install. The same viewer ships with the pipeline and works the same way with your own bundle inside an approved environment.

What you can try

  • Browse Patients (default): the invented participants are listed by pseudonym (PT-NNNN) in the left sidebar. Click one to load the per-patient view, then click through the Docs, Visits, Problems, Meds, Labs/Obs and Sections tabs.
  • Keyword Search: toggle the mode in the sidebar header and search every record and document, for example ALSFRS, riluzole, BiPAP, PEG, El Escorial or a LOINC code such as 82953-1. Each hit shows a snippet with the match highlighted, and clicking a hit opens that patient and tab.
  • Format filters on Docs: filter to C-CDA, HTML, RTF, PDF or text attached to FHIR resources.
  • Export buttons: "Export CSV" on any tab and "Export hits" in keyword search download what the viewer shows. With identifiers hidden, the files hold pseudonyms, but the text columns can still hold identifiers written in notes.

What you are seeing

Each invented participant has a motor neuron disease diagnosis, medicines (riluzole, edaravone, symptom management and comorbidity medicines), coded ALSFRS-R and FVC percent predicted observations, laboratory results, procedures such as PEG placement and BiPAP, an EMG report and clinic notes attached to FHIR resources, and C-CDA, RTF, HTML and PDF documents from the export. Some participants have records from two invented health systems, linked through the mapping file. The other live demos, built from the same run, are on the home page.

Identifiers are hidden by default. Names, MRNs and full dates of birth are replaced before anything is shown or exported; text inside documents and notes is shown as written.

How the embedding works

The viewer checks for a window.EMBEDDED_BUNDLE variable when it starts. If one is present, it loads that data with identifiers hidden in place of the file picker. Otherwise the file picker works as it does in the standalone viewer. The demo page is dashboard.html with the invented bundle in a <script> block in the page head, written by the site build.

For your own deployments you can:

  • Use the standard viewer. Keep dashboard.html and dashboard_data.json inside the approved environment and load the JSON with the file picker. This is the lowest-friction option and what most teams use.
  • Embed a bundle. Add <script>window.EMBEDDED_BUNDLE = { ... }</script> to the head of dashboard.html and save it as one file. The result holds the full bundle, so it needs the same protection as dashboard_data.json and stays inside the approved environment.