🥗 NutriVision AI - Smart Recipe Assistant for Modern Nutrition

AI-powered nutrition analysis and recipe assistant specialized in Thai cuisine. Helping Thai people make informed food choices through evidence-based nutrition science.

อ่านภาษาไทย: README-TH.md

🌟 Features

Core Capabilities

🔍 AI Food Recognition: Snap a photo and instantly identify ingredients using Google Gemini Vision API
📊 8-Dimension Nutrition Scoring: Comprehensive health analysis including:
- Blood Sugar Impact (crucial for Thai diet)
- Gut Health Score
- Inflammation Score
- Nutrient Density
- Processing Level
- Protein Quality
- Micronutrient Coverage
- Overall Health Score
🍜 Thai Food Specialization:
- 1000+ Thai recipes with nutrition data
- Common Thai ingredients database
- Cultural context and traditional wisdom
- Thai measurement conversions
💬 AI Nutrition Coach:
- Chat with AI nutritionist in Thai or English
- Evidence-based recommendations
- Personalized advice based on your goals
📱 Mobile-First PWA:
- Works like a native app
- Offline mode for saved recipes
- Quick camera access

Unique Thai Market Features

Blood Sugar Management: Specialized for Thai rice-heavy diet
Eating Sequence Education: Vegetables → Protein → Fat → Carbs → Sweets (70% blood sugar reduction)
Thai Payment Integration: PromptPay, TrueMoney, Rabbit Line Pay
Line Login: Popular Thai messaging app integration
Thai Language AI: Natural Thai conversation with nutrition AI

🏗️ Architecture

Tech Stack

Backend (Python)

FastAPI - Modern async API framework
PostgreSQL - Primary database
Redis - Caching and sessions
Google Gemini AI - Image recognition and chat
SQLAlchemy - ORM
Alembic - Database migrations

Frontend (TypeScript)

Next.js 14 - React framework with App Router
Tailwind CSS - Utility-first styling
next-intl - Internationalization (Thai + English)
TanStack Query - Server state management
Zustand - Client state management
Recharts - Data visualization

Infrastructure

Docker & Docker Compose - Containerization
Nginx - Reverse proxy
GitHub Actions - CI/CD

🚀 Quick Start

Prerequisites

Docker and Docker Compose
Node.js 18+ and npm 9+
Git
Google Gemini API key (Get one here)

Installation

Option 1: Clone from GitHub (Recommended for Development)

Clone the repository

git clone https://github.com/bejranonda/Nutri-Vision-AI.git
cd Nutri-Vision-AI

Option 2: Install via npm (For Quick Setup)

npx degit bejranonda/Nutri-Vision-AI my-nutrivision-app
cd my-nutrivision-app
npm install

Set up environment variables

cp .env.example .env

Edit .env and add your credentials:

GEMINI_API_KEY=your_gemini_api_key_here
SECRET_KEY=your_secret_key_here  # Generate with: openssl rand -hex 32
JWT_SECRET_KEY=your_jwt_secret_here

Start the application

docker-compose up -d

Initialize database

# Run migrations
docker-compose exec backend alembic upgrade head

# Seed Thai food data
docker-compose exec backend python app/db/seed_data.py

Access the application

Frontend: http://localhost:3000
Backend API: http://localhost:8000
API Documentation: http://localhost:8000/docs

📖 Usage

For Users

Scan Food

Navigate to "สแกนอาหาร" (Scan Food)
Take a photo or upload an image
Get instant nutrition analysis with 8-dimension scores
View detailed ingredient breakdown

Ask AI Nutritionist

Go to "ปรึกษาผู้เชี่ยวชาญ" (Ask Expert)
Type your question in Thai or English
Get evidence-based answers
Chat history is saved

Browse Recipes

Visit "สูตรอาหาร" (Recipes)
Filter by dietary needs
View detailed nutrition scores
Save favorites

Track Progress

Access "แดชบอร์ด" (Dashboard)
View scan history
Track nutrition trends
Set health goals

For Developers

Run Backend Only

cd backend
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install -r requirements.txt
uvicorn app.main:app --reload

Run Frontend Only

cd frontend
npm install
npm run dev

Run Tests

# Backend tests
docker-compose exec backend pytest

# Frontend tests
docker-compose exec frontend npm test

🎯 Nutrition Scoring System

Our proprietary 8-dimension scoring system evaluates food across multiple health aspects:

1. Blood Sugar Impact (0-100)

Calculates Glycemic Load
Accounts for fiber content
Critical for Thai high-carb diet
Higher score = Lower blood sugar spike

2. Gut Health (0-100)

Fiber content (soluble/insoluble)
Prebiotic and probiotic presence
Fermented foods bonus
Higher score = Better gut health

3. Inflammation (0-100)

Omega-3 to Omega-6 ratio
Antioxidant ORAC values
Anti-inflammatory ingredients
Higher score = Anti-inflammatory

4. Nutrient Density (0-100)

Nutrients per calorie (ANDI score)
Protein and fiber density
Vitamin/mineral richness
Higher score = More nutrients per calorie

5. Processing Level (0-100)

NOVA classification (1-4)
Additive detection
Whole food percentage
Higher score = Less processed

6. Protein Quality (0-100)

Complete amino acid profile
Protein quantity
Source quality (PDCAAS)
Higher score = Better protein

7. Micronutrient Coverage (0-100)

% RDI for 12 key vitamins/minerals
Breadth of nutrient variety
Bioavailability factors
Higher score = More complete nutrition

8. Overall Health Score (0-100)

Weighted average of all dimensions
Personalized to user goals
Thai dietary context adjustments
Higher score = Healthier overall

🌐 API Documentation

Authentication Endpoints

POST /api/auth/register      # Register new user
POST /api/auth/login         # Login with email/password
GET  /api/auth/me            # Get current user
POST /api/auth/refresh       # Refresh access token

Food Scanning Endpoints

POST /api/scan/analyze       # Analyze food image
GET  /api/scan/history       # Get scan history
GET  /api/scan/{id}          # Get specific scan

Recipe Endpoints

GET  /api/recipes/           # List recipes with filters
GET  /api/recipes/{id}       # Get recipe details
GET  /api/recipes/featured   # Get featured recipes
POST /api/recipes/suggest    # Generate recipe suggestions

Chat Endpoints

POST /api/chat/ask           # Ask AI nutritionist
GET  /api/chat/history       # Get chat history

User Profile Endpoints

GET  /api/user/profile       # Get user profile
PUT  /api/user/profile       # Update profile
GET  /api/user/stats         # Get usage statistics

See full API documentation at: http://localhost:8000/docs

📊 Database Schema

users
├── id (PK)
├── email (unique)
├── hashed_password
├── subscription_tier (free/premium/family)
├── language (th/en)
├── health_info (age, weight, height, goals)
└── usage_tracking

food_scans
├── id (PK)
├── user_id (FK)
├── image_url
├── detected_items (JSON)
├── nutrition_summary (JSON)
└── 8 dimension scores

recipes
├── id (PK)
├── name_th, name_en
├── ingredients (JSON)
├── instructions_th, instructions_en
├── nutrition_per_serving
├── dietary_flags (vegetarian, halal, etc.)
└── 8 dimension scores

ingredients
├── id (PK)
├── name_th, name_en
├── category
├── nutrition_per_100g
├── glycemic_index
└── special_properties

chat_messages
├── id (PK)
├── user_id (FK)
├── role (user/assistant)
├── content
└── language

meal_plans
├── id (PK)
├── user_id (FK)
├── plan_date
├── breakfast, lunch, dinner (JSON)
└── daily_totals

🔒 Security Features

✅ No API keys in code (environment variables only)
✅ Password hashing with bcrypt
✅ JWT token authentication
✅ CORS protection
✅ Rate limiting on all endpoints
✅ SQL injection prevention (parameterized queries)
✅ XSS protection
✅ CSRF tokens
✅ GDPR compliant from day 1
- Data export functionality
- Right to be forgotten
- Consent management
- Data minimization

🌍 Internationalization

Supports Thai (default) and English:

// Access translations
const t = useTranslations('namespace');

// Thai
<h1>{t('title')}</h1>  // "NutriVision AI"

// Language switch
<Link href="/en/dashboard">English</Link>
<Link href="/th/dashboard">ไทย</Link>

Translation files:

frontend/src/messages/th.json
frontend/src/messages/en.json

🤝 Contributing

We welcome contributions! Please see CONTRIBUTING.md for guidelines.

Development Setup

Fork the repository
Create a feature branch (git checkout -b feature/amazing-feature)
Make your changes
Write tests
Commit with descriptive messages
Push to your fork
Open a Pull Request

Code Style

Python: PEP 8 (use black formatter)
TypeScript: ESLint + Prettier
Commits: Conventional Commits format

📄 License

This project is licensed under the MIT License - see LICENSE file for details.

📞 Contact & Support

Website: https://nutrivision.app
Email: [email protected]
Issues: https://github.com/bejranonda/Nutri-Vision-AI/issues
Discussions: https://github.com/bejranonda/Nutri-Vision-AI/discussions
npm Package: https://www.npmjs.com/package/@nutrivision/nutrivision-ai

Thai Support

Line Official: @nutrivisionai
Facebook: facebook.com/nutrivisionai
Email (Thai): [email protected]

🙏 Acknowledgments

Thalay Community - For nutrition education insights and content inspiration
Google Gemini - For powerful AI capabilities
Thai Food Community - For recipe contributions and cultural context
Early Beta Testers - For valuable feedback

📈 Roadmap

Phase 1 (Current) ✅

Phase 2 (Q2 2024) 🚧

Phase 3 (Q3 2024) 📋

Phase 4 (Q4 2024) 🔮

⭐ Star History

If you find this project helpful, please consider giving it a star!

Made with ❤️ for the Thai community

"อยู่ได้นานพร้อมกับกินอาหารอร่อย" - Live long while eating well

Name		Name	Last commit message	Last commit date
Latest commit History 18 Commits
Initial		Initial
backend		backend
docker		docker
frontend		frontend
research		research
scripts		scripts
.env.example		.env.example
.gitignore		.gitignore
.npmignore		.npmignore
CHANGELOG.md		CHANGELOG.md
CONTRIBUTING.md		CONTRIBUTING.md
DEPLOYMENT.md		DEPLOYMENT.md
LICENSE		LICENSE
PROJECT_PLAN.md		PROJECT_PLAN.md
PUBLISHING.md		PUBLISHING.md
README-TH.md		README-TH.md
README.md		README.md
RELEASE_NOTES.md		RELEASE_NOTES.md
docker-compose.yml		docker-compose.yml
package-lock.json		package-lock.json
package.json		package.json

License

bejranonda/Nutri-Vision-AI

Folders and files

Latest commit

History

Repository files navigation