QuickCheck - AI-Powered Health Diagnosis App

QuickCheck is a Flutter-based mobile application with a Python Flask backend that uses AI to analyze facial features and provide health assessments.

Features

• AI Health Diagnosis: Capture images of eyes, teeth, and tongue for health analysis
• Multi-condition Detection:
  • Anemia (conjunctival pallor + glossitis)
  • Jaundice (scleral icterus using CIELAB + fuzzy logic)
  • Dental Health (gingivitis and caries)
  • Tongue Health (coating and texture analysis)
  • Conjunctivitis (pink eye detection)
  • Chikungunya (combined eye + tongue symptoms)
• AI Medical Chat: Chat with Groq-powered Llama 3 AI for medical guidance
• Multi-language Support: English and Hindi (हिंदी)
• Emergency Detection: Keyword-based emergency symptom detection
• Symptom Analysis: Voice and text input for symptom description

Project Structure

JiloHealth/
├── mobile/                 # Flutter mobile app
│   ├── lib/
│   │   ├── screens/       # App screens
│   │   ├── services/      # API and business logic services
│   │   └── main.dart
│   ├── assets/
│   │   └── translations/  # Localization files (en.json, hi.json)
│   └── pubspec.yaml
│
├── server/                # Python Flask backend
│   ├── app.py            # Main Flask application
│   ├── calculate_probability.py  # Disease detection algorithms
│   ├── feature_extractor.py      # MediaPipe facial feature extraction
│   ├── requirements.txt
│   └── venv/             # Python virtual environment
│
└── README.md

Setup Instructions

Prerequisites

• Flutter SDK (>=3.0.0)
• Python (>=3.8)
• Firebase Account (for authentication)
• Groq API Key (for AI chat and diagnosis)

Backend Setup (Flask Server)

1. Navigate to server directory:

   cd server

2. Create and activate virtual environment:

   # Windows
   python -m venv venv
   venv\Scripts\activate
   
   # macOS/Linux
   python3 -m venv venv
   source venv/bin/activate

3. Install dependencies:

   pip install -r requirements.txt

4. Setup environment variables:

   # Copy the example file
   cp .env.example .env
   
   # Edit .env and add your API keys
   # GROQ_API_KEY=your-actual-groq-api-key

5. Run the Flask server:

   python app.py

   Server will start on http://localhost:5000

Mobile App Setup (Flutter)

1. Navigate to mobile directory:

   cd mobile

2. Install dependencies:

   flutter pub get

3. Setup Firebase:

   • Create a Firebase project at Firebase Console
   • Add Android/iOS apps to your Firebase project
   • Download and place configuration files:
     • Android: google-services.json → android/app/
     • iOS: GoogleService-Info.plist → ios/Runner/

4. Configure API URLs:

   • Edit lib/services/api_service.dart
   • Update baseUrl and flaskServerUrl to match your backend

5. Run the app:

   # For Android
   flutter run
   
   # For iOS
   flutter run -d ios
   
   # For web (if enabled)
   flutter run -d chrome

Environment Variables

Backend (.env in server/)

GROQ_API_KEY=your-groq-api-key
FLASK_ENV=development
FLASK_DEBUG=True
SECRET_KEY=your-secret-key
PORT=5000

Frontend (hardcoded in services for now)

• Backend API URL
• Flask Server URL
• Firebase configuration

API Endpoints

Flask Server (http://localhost:5000)

• POST /api/detect - Complete health analysis (extract + detect)
• POST /api/diagnose - Get AI diagnosis from health analysis
• POST /api/chat - Medical chat with AI
• POST /api/extract - Extract facial features from images

Technologies Used

Frontend

• Flutter - Cross-platform mobile framework
• Camera - Image capture
• Google ML Kit - Face detection
• Easy Localization - Multi-language support
• Provider - State management
• Flutter Markdown - Markdown rendering

Backend

• Flask - Web framework
• MediaPipe - Facial feature extraction
• OpenCV - Image processing
• LangChain + Groq - AI integration with Llama 3
• NumPy - Numerical computations

AI & ML

• Groq Llama 3 (70B) - Medical diagnosis and chat
• CIELAB Color Space - Perceptually uniform color analysis
• Fuzzy Logic - Intelligent symptom combination
• MediaPipe Face Mesh - 468-point facial landmarks

Health Detection Algorithms

Anemia Detection

• CIELAB a* channel analysis of conjunctiva
• Glossitis texture analysis (smooth/beefy red tongue)
• Weighted fusion: Eyes (70%) + Tongue (30%)

Jaundice Detection

• CIELAB b* channel analysis
• Fuzzy logic combining eye and face scores
• Advanced region segmentation (luminance + YCrCb)

Conjunctivitis Detection

• CIELAB a* channel for sclera redness
• Threshold: a* > 135 indicates inflammation

Chikungunya Detection

• Fuzzy logic combining conjunctivitis + white tongue coating
• Eye as primary indicator, tongue as secondary

App Screens

1. Login/Signup - Firebase authentication
2. Home - Start diagnosis or AI chat
3. Symptom Input - Describe symptoms (text/voice)
4. Camera Capture - 4-stage guided capture (face, eyes, teeth, tongue)
5. Review & Analysis - View captured images and health analysis
6. Medical Chat - AI-powered medical Q&A

Localization

• English (en)
• Hindi (हिंदी) (hi)

All UI elements and AI responses support both languages.

Security Notes

• DO NOT commit .env files to git
• DO NOT commit API keys or secrets
• DO NOT commit Firebase configuration files
• Always use .env.example as a template

License

This project is for hackathon/educational purposes.

Contributors

Built for a hackathon with ❤️

Troubleshooting

Flask server won't start

# Make sure you're in the virtual environment
cd server
source venv/bin/activate  # or venv\Scripts\activate on Windows
pip install -r requirements.txt

Flutter build fails

# Clean and rebuild
flutter clean
flutter pub get
flutter run

MediaPipe errors

# Reinstall with correct versions
pip install mediapipe==0.10.11 numpy==1.24.4

Support

For issues and questions, please check the code documentation or create an issue in the repository.

---

Medical Disclaimer: This app is for educational and informational purposes only. Always consult a qualified healthcare professional for medical advice, diagnosis, or treatment.