MediMenu is a powerful, dual-sided web application that bridges the gap between restaurants and diners with dietary restrictions. By leveraging OpenAI's GPT-4o vision capabilities, MediMenu digitizes restaurant menus and provides personalized safety assessments for users based on their specific allergies, medications, and dietary preferences.

• Personalized Profiles: Set up a detailed profile with specific allergies, medications, and dietary restrictions (e.g., Vegan, Gluten-Free, Halal).
• QR Code Menu Scanning: Scan a restaurant's MediMenu QR code to instantly view their menu.
• Personalized Risk Assessment: Every dish on the menu is automatically color-coded based on your profile:
  • 🟢 OK (Green): Safe to eat.
  • 🟡 Caution (Yellow): Possible cross-contact risks or minor dietary warnings.
  • 🔴 Avoid (Red): Contains allergens or heavily conflicts with your restrictions.
• Detailed Safety Insights: See exactly why a dish was flagged and what ingredients triggered the warning.

• AI-Powered Menu Digitization: Simply upload an image of your physical menu, and GPT-4o vision will extract the dishes and infer ingredients automatically.
• Menu Management: Review and edit the AI-generated menu, confirm specific allergens, and flag cross-contact risks for shared preparation areas.
• Instant Publishing: Once confirmed, publish the menu to immediately generate a scan-ready QR code for your diners to use.
• Analytics Dashboard: Track how many diners with specific allergies are scanning your menu to better tailor your offerings in the future.

• Frontend: React + Vite, React Router, vanilla CSS
• Backend: FastAPI (Python), SQLite
• AI Integration: OpenAI GPT-4o (Vision + Structured JSON outputs)

.
├─ backend/
│  ├─ app/
│  │  ├─ main.py           # FastAPI entry point & API endpoints
│  │  ├─ config.py         # Environment configuration
│  │  ├─ database.py       # SQLite database operations
│  │  ├─ prompts.py        # OpenAI system prompts
│  │  ├─ schemas.py        # Pydantic models & validation
│  │  └─ services/
│  │     └─ analyzer.py    # GPT-4o Vision API integration
│  ├─ requirements.txt     # Python dependencies
│  └─ .env.example
├─ frontend/
│  ├─ src/
│  │  ├─ App.jsx           # React app router & main B2C view
│  │  ├─ api.js            # API client for backend communication
│  │  ├─ styles.css        # Global styles and UI components
│  │  ├─ components/       # UI Components (Dashboards, QR Scanner, forms)
│  │  └─ utils/
│  │     └─ image.js       # Image processing utilities
│  ├─ package.json
│  ├─ vite.config.js
│  └─ .env.example
└─ .gitignore

You can try our app here: https://medimenu-frontend.onrender.com

For local set up, refer to below steps:

1. Navigate to the backend directory:

   cd backend

2. Create and activate a Python virtual environment (Optional but recommended):

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

4. Configure environment variables:
   • Copy the environment template:

     cp .env.example .env

   • Open backend/.env and add your OPENAI_API_KEY.
5. Run the API server:

   uvicorn app.main:app --reload --port 8000

   The backend will be available at http://localhost:8000.

1. Navigate to the frontend directory:

   cd frontend

2. Install dependencies:

   npm install

3. Configure environment variables:
   • Copy the environment template:

     cp .env.example .env

   • Make sure VITE_API_BASE_URL is set to your backend URL (usually http://localhost:8000).
4. Run the development server:

   npm run dev

   The frontend will be available at http://localhost:5173.

• POST /api/profile - Create a new user profile
• GET /api/profile/{id} - Fetch user profile details
• POST /api/analyze - Analyze an independent menu image against a profile
• GET /api/history/{profile_id} - View past scan history

• POST /api/restaurant - Register a new restaurant
• GET /api/restaurants - List available restaurants
• GET /api/restaurant/{restaurant_id} - Get restaurant details
• POST /api/restaurant/{restaurant_id}/menu - Upload and analyze a menu image via GPT-4o
• GET /api/restaurant/{restaurant_id}/menu - Get the current draft menu
• PUT /api/restaurant/{restaurant_id}/menu - Update dishes and confirm allergens
• POST /api/restaurant/{restaurant_id}/menu/confirm - Publish the menu for QR scanning
• GET /api/restaurant/{restaurant_id}/menu/personalized - Get a personalized view of the menu for a specific profile
• GET /api/restaurant/{restaurant_id}/analytics - View scan analytics

MediMenu is designed as a responsive, mobile-first web application (React). Here is how you can deploy and view it on a mobile device:

Go to https://medimenu-frontend.onrender.com on your phone for demo.

• B2B (Restaurants): A tiered subscription model. Basic menu digitization is free, but advanced features like cross-contact analysis, automated allergen tagging, and diner analytics are part of a premium tier.
• B2C (Diners): A freemium model. Basic profile matching and scanning are free to ensure accessibility for those who need it most. Premium features could include saving favorite safe meals across multiple restaurants and nutritional tracking.

• POS Integration: Direct integration with restaurant Point-of-Sale (POS) systems (like Toast or Square) to automatically sync menu updates and ingredient changes in real-time.
• Community-Driven Reviews: Allowing users to verify and review the allergen-safety of specific dishes.
• Multi-lingual Support: Translating menus dynamically for international travelers with dietary restrictions.
• Expanded Medical Database: Partnering with healthcare providers to maintain an up-to-date database of complex food-medication interactions.

• Jenny LinkedIn
• Sam samkwak188 (github)
• Sherry Sherry1247 (github)
• Mckenna mckenaz (github)

Built with ❤️ for CheeseHack 2026 at UW-Madison.