🧠 Dtwin – All-in-One Health Management & Disease Prediction App

Dtwin is an AI-powered, full-stack health management application designed to deliver personalized real-time health insights, early disease predictions, and custom meal and fitness recommendations. Leveraging data from wearables like Fitbit and other real-time sources, Dtwin empowers users to take control of their health with AI-driven guidance—not generic advice.

---

🚀 Live Demo

🔗 Live Web App: https://dtwin-health.app

NOTE: This website is only mobile compatible.

🔗 Backend API: https://api.dtwin-health.app

🔗 DTwin APK: Download APK

---

📸 App Screenshots

🫀 Heart Analytics	🤖 Personalized Diet	🔎 Gut Analysis

📊 Main Dashboard	🩺 Mental Wellness Chatbot	📈 Variable Data

---

🧬 Features

• 🩺 Disease Prediction
  Real-time risk prediction for:

  • Heart Disease
  • Diabetes
  • Gut Health Disorders
  • More conditions coming soon

• 📊 Real-Time Health Insights
  Get instant health feedback based on:

  • Fitbit data
  • Heart rate variability
  • Step count, sleep patterns, calorie burn

• 🍽️ Personalized Meal Plans
  Custom diet suggestions tailored to each user's:

  • Blood sugar trends
  • Activity level
  • Gut microbiome profiles

• 🏃‍♂️ Custom Exercise Recommendations
  Dynamic fitness plans based on:

  • Energy expenditure
  • Weight trends
  • Recovery and fatigue scores

• 📈 Smart Analytics Dashboard
  Visualizations and predictions to help you make smarter decisions about your health.

---

🛠️ Tech Stack

🔷 Frontend

• React.js
• TailwindCSS
• Chart.js & D3.js (for visualizations)

🔶 Backend

• Node.js
• Express.js
• REST API (with token-based authentication)
• MongoDB (for user data)

🧪 Machine Learning

• Python (Scikit-learn, TensorFlow)
• Trained on real-time + public datasets
• Served via FastAPI/Python microservices

📱 Mobile App

• Capacitor.js
• Android & iOS builds

---

📂 Datasets Used

• Heart Disease Prediction: Garmin Real-Time Sensor Dataset
• Diabetes Prediction: IICMBC Clinical Dataset
• Gut Health: American Gut Project (AGP)

All models are trained with preprocessing pipelines and continuously updated using anonymized wearable data.

---

⚙️ Installation (Development)

Prerequisites

• Node.js & npm
• Python 3.9+
• MongoDB running locally or Atlas DB

Clone and Run

git clone https://github.com/yourusername/dtwin.git
cd dtwin

# Install frontend
cd client
npm install
npm start

# Install backend
cd ../server
npm install
npm run dev

# Run ML services
cd ../ml-service
pip install -r requirements.txt
uvicorn app:main --reload