JupyterHealth is made of several projects that can be used on their own or together as a combined JupyterHealth platform. This page links to the documentation for each component.
JupyterHealth Exchange: stores patient-consented data and serves it over REST and FHIR, with an optional MCP server. The Exchange also handles login for the Hub.
JupyterHealth client: Python library for reading Exchange data.
JupyterHealth Hub: hosted JupyterHub for analyzing Exchange data and prototyping dashboards. Its configuration is public.
Demos: example notebooks, each with a researcher view in JupyterLab and a clinician view as a Voilà dashboard. The demos only run on the demo deployment.
Helm charts: Helm chart for deploying the Exchange and its MCP server to Kubernetes.
Single-user image: the software environment each Hub user gets.
SMART on FHIR tools: for dashboards that clinicians open from their EHR. See Share data products with clinicians and patients for how they fit together.
Provider template: a starter dashboard app to copy and edit.
jhe-smart-demo: a mock EHR and Exchange for testing a launch locally.
jupyter
-smart -on -fhir: the Jupyter extension that handles the SMART launch.
OMH shim: converts wearable data from vendor schemas to IEEE 1752 and Open mHealth schemas.