EEG Seizure Detection System

A real-time epileptic seizure detection system using EEG signals. The application processes EEG data via MQTT, uses a machine learning model for prediction, and displays results through an interactive web interface.

System Overview

This system consists of three main components:

1. Node-RED Flow: Processes incoming EEG data from MQTT
2. Python Flask Backend: Hosts the web application and ML model
3. Web Interface: Real-time visualization dashboard

Prerequisites

• Python 3.8+
• Node.js and npm
• Node-RED
• MQTT broker (Mosquitto recommended)

Installation Guide

1. MQTT Broker Setup

First, install and start the Mosquitto MQTT broker:

# Ubuntu/Debian
sudo apt-get install mosquitto mosquitto-clients
sudo systemctl start mosquitto
sudo systemctl enable mosquitto  # Start on boot

# For macOS with Homebrew
brew install mosquitto
brew services start mosquitto

2. Node-RED Setup

Install Node-RED globally:

npm install -g node-red

Start Node-RED:

node-red

Access the Node-RED editor at http://localhost:1880.

Import the Flow into Node-RED:

1. In the Node-RED interface, click on the menu (≡), then "Import"
2. Click "select a file to import"
3. Choose the flows.json file from this repository
4. Click "Import"
5. Deploy the flow by clicking the "Deploy" button

3. Python Application Setup

Create and activate a virtual environment:

# Create virtual environment
python3 -m venv venv

# Activate virtual environment
# On Windows
venv\Scripts\activate
# On Unix or MacOS
source venv/bin/activate

Install all required dependencies:

pip install flask flask-socketio paho-mqtt numpy pandas scikit-learn tensorflow

Start the application:

python3 webapp/app.py

Access the web application at http://localhost:5000.

Using the Application

1. Start the Simulator: Click "Start the EEG Simulation" on the web interface to begin generating test data
2. Adjust Parameters:
   • Use the interval dropdown to set the speed of data generation
   • Adjust the threshold slider to change the seizure detection sensitivity
3. View Results: Monitor real-time EEG signals, detection status, and history
4. Stop Simulation: Click "Stop Simulation" when finished