🚀 OpenHealth

AI Health Assistant | Powered by Your Data, Running Locally

🌍 Choose Your Language

English | Français | Deutsch | Español | 한국어 | 中文 | 日本語

---

🌟 Overview

OpenHealth helps you take charge of your health data. By leveraging AI and your personal health information, OpenHealth provides a private and locally-run assistant that helps you better understand and manage your health.

✨ Project Features

Core Features

• 📊 Centralized Health Data Input: Easily consolidate all your health data in one place.
• 🛠️ Smart Parsing: Automatically parses your health data and generates structured data files.
• 🤝 Contextual Conversations: Use the structured data as context for personalized interactions with GPT-powered AI.

📥 Supporting Data Sources & Language Models

Data Sources You Can Add	Supported Language Models
• Blood Test Results • Health Checkup Data • Personal Physical Information • Family History • Symptoms	• LLaMA • DeepSeek-V3 • GPT • Claude • Gemini

🤔 Why We Built OpenHealth

• 💡 Your health is your responsibility.
• ✅ True health management combines your data + intelligence, turning insights into actionable plans.
• 🧠 AI acts as an unbiased tool to guide and support you in managing your long-term health effectively.

🗺️ Project Diagram

graph LR
    subgraph Health Data Sources
        A1[Clinical Records<br>Blood Tests/Diagnoses/<br>Prescriptions/Imaging]
        A2[Health Platforms<br>Apple Health/Google Fit]
        A3[Wearable Devices<br>Oura/Whoop/Garmin]
        A4[Personal Records<br>Diet/Symptoms/<br>Family History]
    end

    subgraph Data Processing
        B1[Data Parser & Standardization]
        B2[Unified Health Data Format]
    end

    subgraph AI Integration
        C1[LLM Processing<br>Commercial & Local Models]
        C2[Interaction Methods<br>RAG/Cache/Agents]
    end

    A1 & A2 & A3 & A4 --> B1
    B1 --> B2
    B2 --> C1
    C1 --> C2

    style A1 fill:#e6b3cc,stroke:#cc6699,stroke-width:2px,color:#000
    style A2 fill:#b3d9ff,stroke:#3399ff,stroke-width:2px,color:#000
    style A3 fill:#c2d6d6,stroke:#669999,stroke-width:2px,color:#000
    style A4 fill:#d9c3e6,stroke:#9966cc,stroke-width:2px,color:#000
    
    style B1 fill:#c6ecd9,stroke:#66b399,stroke-width:2px,color:#000
    style B2 fill:#c6ecd9,stroke:#66b399,stroke-width:2px,color:#000
    
    style C1 fill:#ffe6cc,stroke:#ff9933,stroke-width:2px,color:#000
    style C2 fill:#ffe6cc,stroke:#ff9933,stroke-width:2px,color:#000

    classDef default color:#000

Loading

Note: The data parsing functionality is currently implemented in a separate Python server and is planned to be migrated to TypeScript in the future.

Getting Started

⚙️ How to Run OpenHealth

Installation Instructions

1. Clone the Repository:

   git clone https://github.com/OpenHealthForAll/open-health.git
   cd open-health

2. Setup and Run:

   # Copy environment file
   cp .env.example .env
   
   # Add API keys to .env file:
   # UPSTAGE_API_KEY - For parsing (You can get $10 credit without card registration by signing up at https://www.upstage.ai)
   # OPENAI_API_KEY - For enhanced parsing capabilities
   
   # Start the application using Docker Compose
   docker compose --env-file .env up

   For existing users, use:

   docker compose --env-file .env up --build

3. Access OpenHealth: Open your browser and navigate to http://localhost:3000 to begin using OpenHealth.

Note: The system consists of two main components: parsing and LLM. Currently, parsing utilizes Upstage and OpenAI APIs (which showed the best performance in our testing) with a local parser coming soon, while the LLM component can run fully locally using Ollama.

Note: If you're using Ollama with Docker, make sure to set the Ollama API endpoint to: http://docker.for.mac.localhost:11434/

---

🌐 Community and Support