A Node.js backend for the FallGuardian system - a healthcare application that connects patients with nurses, featuring fall detection and communication tools.

• User Management: Patient, nurse, and admin roles
• Patient-Nurse Communication: Secure message threads through forms
• Fall Detection: Alert system for patient falls
• AI Integration: Text generation and support

• Node.js (v18 or higher)
• MongoDB (local or MongoDB Atlas)
• Google OAuth credentials
• Docker (for containerized deployment)

Create a .env file in the root directory with:

# Server Configuration
PORT=3000
NODE_ENV=development  # Use 'production' in production environments

# Database
MONGODB_URI=mongodb+srv://username:password@cluster.mongodb.net/database

# Session
SESSION_SECRET=your_session_secret_key_here

# Google OAuth
GOOGLE_CLIENT_ID=your_google_client_id_here
GOOGLE_CLIENT_SECRET=your_google_client_secret_here
GOOGLE_CALLBACK_URL=http://localhost:3000/api/auth/google/callback
LOGOUT_REDIRECT_URL=/

# SMS/Notifications
TELESIGN_CUSTOMER_ID=your_telesign_customer_id
TELESIGN_API_KEY=your_telesign_api_key

# AI API
NEBIUS_API_KEY=your_nebius_api_key
DEEPGRAM_API_KEY=your_deepgram_api_key

1. Install dependencies:

   npm install

2. Run in development mode:

   npm run dev

# Build the Docker image
docker build -t fallguardian-backend .

# Run the container
docker run -p 3000:3000 --env-file .env fallguardian-backend

The project includes GitHub Actions workflows for:

• Automated testing
• Building and pushing Docker images
• Deployment to Azure Container Instances

See DEPLOYMENT.md for detailed deployment instructions.

Swagger API documentation is available at /api-docs when the server is running.

{
  name: String,            // User's full name
  email: String,           // Email address (unique)
  googleId: String,        // Google OAuth ID
  age: Number,             // User's age (0-120)
  role: String,            // 'patient', 'nurse', or 'admin'
  phoneNumber: String,     // Phone number (E.164 format)
  nurseId: ObjectId,       // For patients: assigned nurse
  assignedPatients: [ObjectId], // For nurses: list of patients
  createdAt: Date          // Account creation timestamp
}

{
  title: String,           // Form title
  patient: ObjectId,       // Patient reference
  nurse: ObjectId,         // Nurse reference
  status: String,          // 'pending', 'in-progress', 'resolved', 'cancelled'
  resolved: Boolean,       // Whether form is resolved
  resolvedBy: ObjectId,    // User who resolved the form
  resolvedAt: Date,        // When the form was resolved
  messages: [{
    sender: ObjectId,      // Message sender
    body: String,          // Message content
    attachment: String,    // Optional attachment URL
    createdAt: Date        // Message timestamp
  }],
  createdAt: Date,         // Form creation timestamp
  updatedAt: Date          // Last update timestamp
}

• GET /api/auth/google: Google OAuth login
• GET /api/auth/dashboard: Authentication dashboard
• GET /api/auth/logout: Logout

• GET /api/users: Get all users
• GET /api/users/me: Get current user profile
• PUT /api/users/me: Update current user profile

• GET /api/nurse/me/patients: Get nurse's patients
• POST /api/nurse/me/patients/:patientId/assign: Assign patient to nurse
• DELETE /api/nurse/me/patients/:patientId: Remove patient from nurse

• GET /api/patient/me/nurse: Get patient's nurse
• POST /api/patient/me/fall: Alert nurse about patient fall

• POST /api/forms: Create new form
• GET /api/forms/me: Get user's forms
• GET /api/forms/:id: Get specific form
• POST /api/forms/:id/messages: Add message to form
• POST /api/forms/:id/resolve: Mark form as resolved

The application follows a layered architecture:

• Models: Database schemas (MongoDB/Mongoose)
• Services: Business logic
• Controllers: Request handling
• Routes: API endpoint definitions
• Middleware: Authentication, validation, error handling

MIT License