Overview
The EEG-Based Seizure Detection System is a wearable health device powered by Machine Learning and IoT that continuously analyzes EEG brain signals to detect seizure patterns. Upon detection, it automatically notifies hospitals and emergency contacts through an integrated Android application.
Objective
- Develop a real-time wearable EEG monitoring system.
- Detect seizures using ML-based pattern recognition.
- Transmit alerts via Bluetooth to Android devices.
- Enable instant hospital notification and location tracking.
Core Features
- EEG signal acquisition via non-invasive sensors.
- Preprocessing with Fast Fourier Transform (FFT).
- Machine learning-based classification for seizure detection.
- Android app integration for emergency alerts.
Machine Learning Implementation
EEG data is preprocessed, segmented, and classified using ML algorithms such as Random Forest and SVM. Frequency-domain features from FFT and statistical features are extracted for classification.
import pandas as pd
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
data = pd.read_csv("eeg_features.csv")
X = data.drop("label", axis=1)
y = data["label"]
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
model = RandomForestClassifier(n_estimators=120, random_state=42)
model.fit(X_train, y_train)
pred = model.predict(X_test)
print("Accuracy:", accuracy_score(y_test, pred))
Android App Integration
The EEG sensor communicates via Bluetooth Serial with an Android app that visualizes EEG readings and triggers emergency alerts.
// Bluetooth data listener (pseudo)
if (input.contains("SEIZURE")) {
showAlert();
sendSMS("Emergency detected at " + location);
}
Results & Accuracy
Achieved 96.3% accuracy in seizure classification. Real-time alerts triggered within 2 seconds of event detection.
- Low latency Bluetooth communication
- Fast classification using optimized features
- Reliable emergency alerts in real-time
Future Scope
- Integrate deep learning (CNN/LSTM) for predictive seizure forecasting.
- Cloud dashboard for neurologists to monitor patients remotely.
- Smartwatch integration for 24/7 EEG tracking.
- Improved comfort through flexible electrode materials.