JupyterHealth Exchange is a Django web application that facilitates the sharing of user-consented medical data with authorized consumers through a web UI, REST, and FHIR APIs.
In the context of JupyterHealth, data producers are typically study participants (FHIR Patients) using the CommonHealth Android App linked to personal devices (e.g., Glucose Monitors), and data consumers are typically researchers (FHIR Practitioners).
The Problem: Healthcare Data Silos¶
Healthcare data today is fragmented across proprietary systems. Wearables, medical devices, and apps each collect valuable health information, but this data remains locked in vendor-specific formats and isolated databases. Researchers struggle to access real-time patient data, and patients have limited control over sharing their own health information.
Traditional healthcare operates on:
Proprietary systems with vendor lock-in
Siloed data that can’t easily be shared between systems
All-or-nothing consent where patients must share everything or nothing
Delayed insights due to manual data collection and batch processing
The Solution: Open Platform for Consent-Based Data Sharing¶
JupyterHealth Exchange provides a secure, open-source middleware platform that enables:
For Patients:
Granular control over what data to share
Scope-level consent (e.g., share blood glucose but not weight)
Study-specific consent (different permissions for different research projects)
Ability to revoke consent at any time
For Researchers and Practitioners:
Real-time access to consented patient data
Standardized data formats (FHIR R5, Open mHealth)
REST and FHIR APIs for programmatic access
Role-based access control for team collaboration
Integration with analysis tools (JupyterHub, Python notebooks)
For Institutions:
HIPAA-compliant framework for data management
Hierarchical organization structure (institution → department → lab)
Audit trails for data access
Flexible deployment (cloud or on-premises)
Part of a Broader Vision¶
JupyterHealth Exchange is a key component of the Berkeley Agile Metabolic Health initiative, a collaboration between UC Berkeley, UCSF, Project Jupyter, 2i2c, and The Commons Project. The initiative aims to transform diabetes and metabolic health treatment by creating an open-source, data-driven healthcare platform.
The goal is to enable:
Personalized medicine based on individual patient data rather than population averages
Faster medical innovation through open data sharing and collaboration
Lower healthcare costs by reducing inefficiencies in data collection
Individual empowerment with global reach where patients control their data while contributing to research
Who Should Use JupyterHealth Exchange?¶
Research Institutions¶
Conducting clinical trials or observational studies that require:
Real-time wearable data collection
Patient consent management
FHIR-compliant data storage
Integration with analysis pipelines
Healthcare Providers¶
Running patient monitoring programs that need:
Remote patient monitoring capabilities
Integration with personal health devices
Secure data sharing with researchers
Compliance with healthcare regulations
Individual Researchers¶
Working on studies involving:
Diabetes and metabolic health
Cardiovascular monitoring
Sleep and activity tracking
Digital biomarker development
Personalized health interventions
Platform Developers¶
Building on the JupyterHealth ecosystem to:
Develop custom data collection apps
Create analysis dashboards and visualizations
Integrate new data sources or wearables
Extend functionality for specific use cases
What Makes JupyterHealth Exchange Different?¶
Granular Consent Management¶
Unlike all-or-nothing consent models, patients can choose exactly which types of data to share with specific studies. This scope-level consent respects patient autonomy while enabling research.
Built on Healthcare Standards¶
FHIR R5 for interoperability with healthcare systems
Open mHealth (IEEE 1752) schemas for normalized device data
OAuth 2.0/OIDC for secure authentication
SMART on FHIR OAuth provider for research applications
Lightweight and Accessible¶
Vanilla JavaScript SPA (no complex build tools required)
Single container deployment option
Minimal dependencies
Serves as a reference implementation for others to build upon
Open Source and Community-Driven¶
Developed under the JupyterHub community
Apache 2.0 license
Active development on GitHub
Contributions welcome from researchers, developers, and healthcare professionals
Core Capabilities¶
JupyterHealth Exchange provides three main interfaces:
Web User Interface¶
A single-page application for practitioners to:
Create and manage organizations and studies
Invite and enroll patients
Configure consent requests
View patient data and consents
Generate patient invitation links
Admin REST API¶
A comprehensive API for programmatic management:
Organization hierarchy management
Study creation and configuration
Patient enrollment workflows
Consent management
Data source configuration
FHIR API¶
Standards-based API for healthcare interoperability:
Patient resource queries
Observation resource queries (health data)
FHIR Bundle support for batch operations
FHIR-compliant search parameters
Compatible with FHIR tooling and libraries
Technology Choices¶
JupyterHealth Exchange makes deliberate technology choices to balance functionality, accessibility, and maintainability:
Django: Mature Python web framework with strong healthcare adoption
PostgreSQL: Rich JSON support for flexible FHIR data storage
Vanilla JavaScript: Avoids npm/webpack complexity for simpler deployment
FHIR Resources: Standard-based data models for healthcare compatibility
Open mHealth: Normalized schemas for wearable and device data
For a deeper understanding of these architectural decisions, see Architecture in the reference documentation.
Current Status and Roadmap¶
Current State: Proof of Concept
JupyterHealth Exchange is actively developed and suitable for:
Research prototypes
Pilot studies
Development and testing
Reference implementation for similar systems
You can monitor progress in our GitHub Repository.
Getting Started¶
Ready to explore JupyterHealth Exchange?
For Developers: See the deployment tutorial to set up JupyterHealth Exchange.
For Understanding: Continue reading the explanation documentation to understand the technologies and design decisions behind JupyterHealth Exchange.
Learn More¶
External Resources:
Explanation Documentation: For deeper understanding of JupyterHealth Exchange, see:
Consent Management - How patient consent controls data access
Role-Based Access and Governance - User roles and permissions
System Architecture - Component architecture and deployment
Data Flow - How data moves through the system
Security Overview - HIPAA compliance and security controls
Features¶
OAuth 2.0, OIDC, and SMART on FHIR Identity Provision using django
-oauth -toolkit FHIR schema validation using fhir.resources
REST APIs using Django Rest Framework
Built-in, lightweight Vanilla JS SPA UI (npm not required) using oidc-client-ts, handlebars, and bootstrap
Status¶
This project is currently in a Proof of Concept stage. You can monitor progress in our GitHub Repository.