A self-hosted, privacy-first web application for centralizing, monitoring, and analyzing health and wellness data using open healthcare standards.

Note: Health Assistant is an independent, open-source project inspired by the privacy-first, local-control philosophy of Home Assistant. It is not affiliated with, endorsed by, or connected to Home Assistant or Nabu Casa.

Development Status: This project is currently in Beta. The core platform is fully functional, featuring AI-powered OCR, dynamic visualizers, and multi-tenant patient management, but is still undergoing active development.

💖 Support the Project

If Health Assistant has helped you take control of your health data, or if you believe in privacy-first healthcare infrastructure, please consider supporting the project.

Your sponsorship directly supports the continued development, maintenance, and improvement of the platform.

Buy Me A Coffee

📖 Documentation

• Technical Architecture
• AI System & Extensibility
• Integrations SDK & Developer Guide
• Development Guide
• CI/CD Deployment & Setup
• API Documentation
• Visual Tour

For a detailed breakdown of the codebase, see the Project Structure document.

Key Features

• Multi-Tenancy & Household-Ready: Zero-config setup for home users. The first user becomes the System Admin, and households are auto-provisioned upon registration.
• Data Sovereignty & No Vendor Lock-in: Perfect for healthcare startups and clinics. Deploy Health-Assistant on your own HIPAA-compliant infrastructure. You own the code and the data, avoiding the trap of BAA-locked proprietary platforms.
• Identity & Record Linking: Seamlessly link login accounts to clinical Patient or Doctor records, perfect for individual users managing their own family's health.
• HL7 FHIR Standard: Full compliance with Fast Healthcare Interoperability Resources, including validated-on-write storage (invalid FHIR is rejected at the API) and a dedicated FHIR Server integration that connects to external hospital/health-record systems via SMART-on-FHIR.
• Examinations Platform: Native grouping of medical documents, clinical notes (Rich-Text/Markdown), and analytics exclusively by distinct clinical visits.
• Clinical Events & Longitudinal Tracking: Specialized system for tracking clinical events like pregnancies, chronic pain, or surgical recoveries. Features a Metadata-Driven Architecture allowing for dynamic, specialty-specific fields and categorized filtering.
• Bi-directional Visit Mapping: Seamlessly link examinations to clinical events with specific clinical reasons. Associations can be managed directly from the Examination preview or detail pages.
• Clinical Dashboard: Global and patient-specific views of long-term medical events.
• Unified AI Architecture & Agentic Chatbot: Provider-agnostic AI factory supporting OpenAI, Anthropic, and Local LLMs via LangChain. Configure models per task type and tenant. Features an advanced, context-aware AI Assistant that acts as an agentic chatbot. It is equipped with tools to fetch biomarker history, explain complex medications, query raw document contents, and provide interactive clinical insights.
• AI-Powered "Magic Fill" & Copilot: Specialized AI workflows that auto-generate clinical biomarker and medication definitions, generate custom SVG UI icons, and provide a "Magic Fill" capability to instantly map unstructured text into structured examination records.
• Dynamic Category Visualizations: Auto-tagging mechanism routing processed results directly to specialized Clinical Components (Lab Results, Imaging Gallery, Procedure Reports).
• Patient Context Control: Robust top-level tracking system controlling the flow of views and data filtering across the entire application.
• TimescaleDB: Native time-series support for tracking patient biomarker progression over time.
• Pluggable Integration Framework & SDK: A modular system for connecting third-party platforms (wearables, labs, webhooks, and cloud OAuth sources via SMART-on-FHIR with Dynamic Client Registration) with a schema-driven UI and automated background syncing. Includes a dedicated Integrations SDK for developers to rapidly build custom data providers with built-in connection pooling, rate limit handling, cursor-based delta syncing, and reusable OAuth2+PKCE auth.
• Async Processing: Fast background tasks powered by Redis + Celery.

Quick Start

For manual setup and production deployment, please see the Installation Guide. For developers, refer to the Development Guide.

License

Apache License 2.0 - See LICENSE file for details.

Disclaimer

This software, including its AI chatbot and agentic features that may offer health-related information, guidance, or medication explanations, is for informational and wellness purposes only. It does NOT provide medical diagnosis or act as a substitute for professional medical care. Always consult certified medical professionals for health advice, diagnoses, or before making any medical decisions based on the software's outputs.