Skip to article frontmatterSkip to article content
Site not loading correctly?

This may be due to an incorrect BASE_URL configuration. See the MyST Documentation for reference.

JupyterHealth

Open infrastructure for health care. JupyterHealth is a project that connects wearable, clinical, and patient-generated data to computational and AI tools for health care. All of the tools it builds are open source, built on Jupyter and open standards like FHIR and Open mHealth, and made for researchers, clinicians, and patients.

JupyterHealth’s two main components are the Exchange, which stores patient-consented data, and the Hub, where people analyze that data and build data products for clinicians and patients. Those data products can open inside an EHR with SMART on FHIR, so clinicians use them from a patient’s chart.

Thinking about what you could build with JupyterHealth tools, or whether it fits your organization? Start with the use cases and How JupyterHealth works, or talk to the team.

Here’s the workflow it supports.

What do you want to do?

Each page covers one step of the workflow.

Try the Berkeley demo

You can try JupyterHealth with the Berkeley demo Exchange and Hub. The Exchange has sample data that you can explore in a notebook on the Hub. You’ll need an invite code to create an account. To request one, ask in the #jupyterhealth Zulip channel.

  1. Create an account on the Berkeley demo Exchange with your invite code.

  2. Open the Berkeley demo Hub to use JupyterLab, where you can create notebooks and use the AI chat interface.

To get started exploring some data, you can:

These links are specific to the Berkeley demo. Other institutions may run their own Exchange and Hub, or connect their Exchange to a different computing environment. If you use JupyterHealth through another institution, use the links and instructions it gives you.