SymptoSafe - AI-Powered Symptom Checker

A production-grade, failure-aware AI symptom checker built with Next.js 14, TypeScript, and OpenAI.

🌟 Features

Safety-First Architecture

• Rule-Based Emergency Detection: Instant detection of critical symptoms before AI analysis
• Dual Confidence Scoring: Transparent diagnostic confidence and information completeness metrics
• Risk Stratification: Clear risk levels (Critical, High, Medium, Low) with actionable recommendations
• Emergency Escalation: Full-screen alerts with direct links to emergency services

AI-Powered Analysis

• OpenAI GPT-4 Integration: Advanced symptom analysis with structured responses
• Differential Diagnosis: Multiple possible conditions with probability estimates
• Follow-up Questions: Intelligent questioning to gather more information
• Uncertainty Transparency: Always expresses limitations and recommends professional care

Privacy & Security

• Firebase Authentication: Secure Google Sign-In and Email/Password authentication
• Conditional Data Storage: Anonymous users have no data stored; logged-in users can save history
• Encrypted Storage: All data encrypted in transit and at rest
• No Third-Party Sharing: Health information stays private

User Experience

• Multi-Language Support: English, Hindi, and Marathi
• Caregiver Mode: Describe symptoms for someone else
• Hospital Finder: Quick access to nearby hospitals
• Responsive Design: Works seamlessly on desktop and mobile
• Glass Morphism UI: Modern, accessible design with smooth animations

🚀 Quick Start

Prerequisites

• Node.js 18+ installed
• OpenAI API key
• Firebase project (optional, for authentication)

Installation

1. Clone and navigate to the project:

cd symptosafe

2. Install dependencies:

npm install

3. Set up environment variables:

cp .env.local.example .env.local

4. Edit .env.local and add your keys:

OPENAI_API_KEY=your_openai_api_key_here

# Firebase (optional - app works without auth)
NEXT_PUBLIC_FIREBASE_API_KEY=your_firebase_api_key
NEXT_PUBLIC_FIREBASE_AUTH_DOMAIN=your_project_id.firebaseapp.com
NEXT_PUBLIC_FIREBASE_PROJECT_ID=your_project_id
NEXT_PUBLIC_FIREBASE_STORAGE_BUCKET=your_project_id.appspot.com
NEXT_PUBLIC_FIREBASE_MESSAGING_SENDER_ID=your_sender_id
NEXT_PUBLIC_FIREBASE_APP_ID=your_app_id

5. Run the development server:

npm run dev

6. Open http://localhost:3000

🧪 Testing

Try these test scenarios:

1. Normal Symptoms: "I have headache and fever"
2. Emergency: "Severe chest pain and breathing problem"
3. Vague Symptoms: "I don't feel well"

Use the "Test Scenarios" button in the bottom-right corner for quick testing.

🏗️ Architecture

Safety Engine (lib/safety-engine.ts)

• Runs BEFORE AI analysis
• Detects emergency keywords instantly
• Returns structured emergency results
• Covers: chest pain, stroke, breathing issues, bleeding, suicide, seizures, overdose, burns

AI Engine (lib/ai-engine.ts)

• OpenAI GPT-4 integration
• Structured JSON responses
• Confidence scoring (0-100)
• Information completeness tracking
• Risk level assessment
• Follow-up question generation

State Management (lib/store/useAppStore.ts)

• Zustand for global state
• User authentication state
• Message history
• Analysis results
• Emergency status
• Language preferences

Firebase Integration

• lib/firebase/config.ts: Firebase initialization
• lib/firebase/auth-service.ts: Authentication methods
• lib/firebase/firestore-service.ts: Data persistence

API Routes

• app/api/analyze/route.ts: Main analysis endpoint
  1. Validates request
  2. Runs safety check
  3. Performs AI analysis if safe
  4. Returns structured response

📦 Tech Stack

• Framework: Next.js 14 (App Router)
• Language: TypeScript
• Styling: Tailwind CSS
• AI: OpenAI GPT-4
• Authentication: Firebase Auth
• Database: Firebase Firestore
• State Management: Zustand
• Charts: Recharts
• Animations: Framer Motion
• Icons: Lucide React

🔒 Security Best Practices

1. API Key Protection: OpenAI key stored server-side only
2. Input Validation: All user inputs validated
3. Error Handling: Graceful error messages without exposing internals
4. Rate Limiting: Consider implementing rate limiting in production
5. HTTPS Only: Always use HTTPS in production
6. Firebase Rules: Configure Firestore security rules

🌍 Internationalization

Supported languages:

• English (en)
• Hindi (hi)
• Marathi (mr)

Add more languages in lib/i18n/translations.ts

📱 Responsive Design

• Mobile-first approach
• Breakpoints: sm (640px), md (768px), lg (1024px), xl (1280px)
• Touch-friendly interface
• Optimized for all screen sizes

🚀 Deployment

Vercel (Recommended)

1. Push code to GitHub
2. Import project in Vercel
3. Add environment variables
4. Deploy

Other Platforms

Build the production bundle:

npm run build
npm start

⚠️ Important Disclaimers

1. Not Medical Advice: This tool is for informational purposes only
2. Emergency Situations: Always call emergency services for serious symptoms
3. Professional Care: Always consult healthcare professionals for diagnosis
4. AI Limitations: AI can make mistakes and miss rare conditions
5. Data Accuracy: Results depend on information quality provided

📄 License

This project is for educational and demonstration purposes.

🤝 Contributing

Contributions welcome! Please ensure:

• Code follows TypeScript best practices
• All features are tested
• Documentation is updated
• Security best practices are maintained

📞 Support

For issues or questions:

1. Check the documentation
2. Review test scenarios
3. Verify environment variables
4. Check browser console for errors

🎯 Future Enhancements

• Voice input for symptoms
• Image upload for visual symptoms
• Integration with telemedicine platforms
• Symptom tracking over time
• Family health profiles
• Medication interaction checker
• Health tips and preventive care

---

Built with ❤️ for better healthcare accessibility