🏥 Smart Medicine Box

An AI-driven IoT healthcare solution that bridges the gap between medication adherence and remote clinical supervision through real-time monitoring, intelligent reminders, and caregiver connectivity.

📋 Table of Contents

Overview
Features
System Architecture
Tech Stack
Installation
Usage
Project Structure
API Endpoints
Team
Future Enhancements

🌟 Overview

Medication non-adherence affects nearly 50% of chronic disease patients globally, leading to treatment failures and preventable hospitalizations. The Smart Medicine Box addresses this critical healthcare challenge by combining:

IoT Hardware: ESP32-based device with push buttons, LCD display, and buzzer
Cloud Infrastructure: Firebase Realtime Database for secure data synchronization
AI Analytics: Grok-powered intelligent log analysis and conversational insights
Web Portal: Next.js caregiver dashboard for remote monitoring and communication

Key Statistics

✅ 100% data synchronization accuracy
✅ 100% successful reminder delivery (4-second latency)
✅ 7 seconds average AI response time
✅ 99% system reliability in experimental testing

✨ Features

Patient-Side Features

🔔 Automated Medication Reminders: Visual (LCD) + audio (buzzer) alerts at scheduled times
✅ Medicine Intake Confirmation: One-button confirmation with timestamp logging
🆘 Emergency Alert System: Instant emergency notification to caregivers
💬 Real-time Messaging: Receive messages from caregivers on the device screen
📊 Event Logging: Automatic cloud logging of all interactions

Caregiver Features

📱 Multi-Device Dashboard: Monitor multiple patients from a single interface
📈 Visual Analytics: Interactive charts showing adherence trends and patterns
⏰ Remote Reminder Scheduling: Configure medication schedules with custom repeat cycles
💬 Communication: Send messages directly to patient devices
🤖 AI Assistant: Automated log summaries and conversational query interface
🔐 Secure Authentication: Google OAuth with whitelisted access control

🏗 System Architecture

┌─────────────────────────────────────────────────────────────┐
│                     CLOUD LAYER                             │
│  ┌──────────────────────────────────────────────────────┐  │
│  │         Firebase Realtime Database                    │  │
│  │  • /devices    • /events    • /messages              │  │
│  │  • /reminders  • /allowedDoctors                     │  │
│  └──────────────────────────────────────────────────────┘  │
└─────────────────────────────────────────────────────────────┘
            ▲                              ▲
            │                              │
   ┌────────┴────────┐          ┌────────┴──────────┐
   │  ESP32 DEVICE   │          │  WEB APPLICATION  │
   ├─────────────────┤          ├───────────────────┤
   │ • 16×2 LCD      │          │ • Next.js 15.5.2  │
   │ • Push Buttons  │          │ • Material-UI v6  │
   │ • Piezo Buzzer  │          │ • Recharts        │
   │ • Wi-Fi Module  │          │ • AI Integration  │
   └─────────────────┘          └───────────────────┘

💻 Tech Stack

Hardware

Microcontroller: ESP32 DevKit v1
Display: 16×2 LCD with I2C backpack
Audio: Piezo buzzer (2-4 kHz)

Embedded Software

IDE: Arduino IDE 2.3.6
Libraries:
- Firebase ESP32 Client
- LiquidCrystal I2C
- WiFi.h
- time.h (NTP sync)

Web Application

Framework: Next.js 15.5.2 (React 19.1.0)
Styling: Material-UI v7.3.2, Tailwind CSS 4.1.14
Database: Firebase Realtime Database 12.4.0
Charts: Recharts 3.1.2
Carousel: Swiper 11.2.10

Cloud & AI

Backend: Firebase (Authentication, Realtime Database)
AI Model: Grok 4.1 Fast (via OpenRouter API)
Deployment: Vercel (Web), Firebase Hosting (Database)

🚀 Installation

Prerequisites

Node.js 20.x or higher
Arduino IDE 2.x
Firebase project with Realtime Database enabled
OpenRouter API key

1. Hardware Setup

# Install Arduino ESP32 board support
# Tools → Board → Boards Manager → Search "esp32" → Install

# Install required libraries via Library Manager:
# - Firebase ESP32 Client by Mobizt
# - LiquidCrystal I2C by Frank de Brabander

2. Configure Firebase

Create a Firebase project at console.firebase.google.com
Enable Realtime Database
Set database rules:

{
  "rules": {
    "devices": {
      "$deviceId": {
        ".read": "auth != null",
        ".write": "auth != null"
      }
    },
    "events": {
      "$deviceId": {
        ".read": "auth != null",
        ".write": "auth != null"
      }
    },
    "messages": {
      "$deviceId": {
        ".read": "auth != null",
        ".write": "auth != null"
      }
    },
    "reminders": {
      "$deviceId": {
        ".read": "auth != null",
        ".write": "auth != null"
      }
    },
    "allowedDoctors": {
      ".read": "auth != null"
    }
  }
}

Add authorized caregivers to /allowedDoctors path (replace . with , in email)

3. ESP32 Configuration

Open Arduino sketch
Update Wi-Fi credentials:

#define WIFI_SSID "your-wifi-ssid"
#define WIFI_PASSWORD "your-wifi-password"

Update Firebase credentials:

#define API_KEY "your-firebase-api-key"
#define DATABASE_URL "your-database-url" // e.g., https://project-id.firebaseio.com

Upload to ESP32

4. Web Application Setup

# Clone repository
git clone https://github.com/yourusername/smart-medicine-box.git
cd smart-medicine-box

# Install dependencies
npm install

# Create .env.local file
cp .env.example .env.local

5. Configure Environment Variables

Edit .env.local:

OPENROUTER_API_KEY3=your_openrouter_api_key
NEXT_PUBLIC_FIREBASE_API_KEY=your_firebase_api_key
NEXT_PUBLIC_FIREBASE_AUTH_DOMAIN=your-project.firebaseapp.com
NEXT_PUBLIC_FIREBASE_DATABASE_URL=https://your-project.firebaseio.com
NEXT_PUBLIC_FIREBASE_PROJECT_ID=your-project-id
NEXT_PUBLIC_FIREBASE_STORAGE_BUCKET=your-project.appspot.com
NEXT_PUBLIC_FIREBASE_MESSAGING_SENDER_ID=your-sender-id
NEXT_PUBLIC_FIREBASE_APP_ID=your-app-id
NEXT_PUBLIC_FIREBASE_MEASUREMENT_ID=G-XXXXXXXXXX

6. Run the Application

# Development mode
npm run dev

# Production build
npm run build
npm start

Open http://localhost:3000 in your browser.

📖 Usage

For Caregivers

Login: Access /login and authenticate with Google
Dashboard: View all registered devices
Select Patient: Choose a device to monitor
View Logs: See complete event history with timestamps
Analyze Data: View charts and AI-generated summaries
Send Messages: Communicate directly with the patient
Set Reminders: Schedule medication alerts with custom repeat cycles

For Patients

Power On: Connect the Smart Medicine Box to power
Wait for Connection: LCD shows "Smart Medicine Box" → Wi-Fi connected
Take Medicine: Press left button when taking medication
Emergency: Press right button for urgent caregiver notification
Follow Reminders: Respond to buzzer alerts and LCD messages

📁 Project Structure

smart-medicine-box/
├── arduino/
│   └── smart_medicine_box.ino      # ESP32 firmware
├── src/
│   ├── app/
│   │   ├── api/
│   │   │   ├── ai-chat/route.ts    # Conversational AI endpoint
│   │   │   └── ai-summary/route.ts # Log summarization endpoint
│   │   ├── analyse/page.tsx        # Analytics dashboard
│   │   ├── dashboard/page.tsx      # Caregiver dashboard
│   │   ├── login/page.tsx          # Authentication page
│   │   ├── logs/page.tsx           # Event history viewer
│   │   ├── messages/page.tsx       # Messaging interface
│   │   ├── reminders/page.tsx      # Reminder scheduler
│   │   ├── layout.tsx              # Root layout
│   │   └── page.tsx                # Landing page
│   ├── components/
│   │   └── analyse/
│   │       └── AnalyseChat.tsx     # AI chat component
│   └── lib/
│       ├── firebase.ts             # Firebase configuration
│       └── theme.ts                # Material-UI theme
├── .env.example                     # Environment template
├── firebase.json                    # Firebase config
├── next.config.ts                   # Next.js config
├── package.json                     # Dependencies
└── README.md                        # This file

🔌 API Endpoints

`/api/ai-summary`

Generates automated adherence summaries from patient logs.

Request:

{
  "timeRange": "Week",
  "medicineTaken": 42,
  "emergencies": 3,
  "dailyBreakdown": [...],
  "reminders": [...]
}

Response:

{
  "summary": "Patient showed 85% adherence this week with 3 emergency alerts..."
}

`/api/ai-chat`

Conversational AI for querying patient data.

Request:

{
  "messages": [
    {"role": "user", "content": "How many doses were missed this week?"}
  ],
  "stats": {...}
}

Response:

{
  "reply": "Based on the logs, the patient missed 6 doses this week..."
}

🔮 Future Enhancements

Short-term

SMS/Email notifications (Twilio/SendGrid integration)
Mobile application (React Native)
Battery backup system for power outages

Long-term

Automated pill dispensing mechanism
Weight sensors for pill counting
Fall detection and activity monitoring

🙏 Acknowledgments

Firebase for cloud infrastructure and real-time database
OpenRouter for AI model access
Material-UI and Next.js communities for excellent documentation

⭐ Star this repository if you find it helpful!

Name		Name	Last commit message	Last commit date
Latest commit History 5 Commits
arduino-code		arduino-code
src		src
.env.example		.env.example
.firebaserc		.firebaserc
.gitignore		.gitignore
README.md		README.md
eslint.config.mjs		eslint.config.mjs
firebase.json		firebase.json
next.config.ts		next.config.ts
package-lock.json		package-lock.json
package.json		package.json
tsconfig.json		tsconfig.json

Folders and files

Latest commit

History

Repository files navigation

🏥 Smart Medicine Box

📋 Table of Contents

🌟 Overview

Key Statistics

✨ Features

Patient-Side Features

Caregiver Features

🏗 System Architecture

💻 Tech Stack

Hardware

Embedded Software

Web Application

Cloud & AI

🚀 Installation

Prerequisites

1. Hardware Setup

2. Configure Firebase

3. ESP32 Configuration

4. Web Application Setup

5. Configure Environment Variables

6. Run the Application

📖 Usage

For Caregivers

For Patients

📁 Project Structure

🔌 API Endpoints

/api/ai-summary

/api/ai-chat

🔮 Future Enhancements

Short-term

Long-term

🙏 Acknowledgments

About

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

`/api/ai-summary`

`/api/ai-chat`

Packages