A Novel Framework for Epileptic Seizure Detection Using Electroencephalogram Signals Based on the Bat Feature...

Area of Science:

• Neuroscience
• Biomedical Engineering
• Data Science

Background:

• Accurate electroencephalogram (EEG) signal classification is crucial for brain-computer interfaces (BCI).
• EEG signals present complexity and non-stationarity, necessitating advanced feature extraction and data mining.
• Existing methods require improvement for effective epilepsy diagnosis.

Purpose of the Study:

• To develop a novel pipeline for EEG signal feature construction and dimension reduction using Bat and genetic algorithms.
• To accurately classify epilepsy EEG signals for improved BCI applications.
• To evaluate the proposed method's performance against established classifiers like k-Nearest Neighbors and naïve Bayes.

Main Methods:

• Wavelet extraction and segmentation of EEG signals.
• Bat algorithm for identifying relevant features.
• Genetic algorithm combined with a neural network for automated classification of epilepsy EEG segments.

Main Results:

• The proposed framework achieved superior accuracy and runtime compared to existing methods.
• Minimum accuracies of 100% for balanced classes and 75.9% for unbalanced classes were recorded.
• The system successfully distinguished signals from healthy volunteers and epilepsy patients during various conditions.

Conclusions:

• The novel Bat and genetic algorithm pipeline offers a highly accurate and efficient approach for epilepsy EEG signal analysis.
• This method has significant potential for direct clinical application in rapid and precise diagnosis.
• The developed framework advances BCI capabilities in neurological disorder detection.