An approach to absence epileptic seizures detection using Approximate Entropy

Area of Science:

• Neurology
• Biomedical Signal Processing
• Computational Neuroscience

Background:

• Epilepsy is a common chronic neurological disorder, particularly in children.
• Electroencephalogram (EEG) signals are crucial for diagnosing and analyzing epileptic seizures.
• Approximate Entropy (ApEn) quantifies time series regularity in physiological signals.

Purpose of the Study:

• To apply Approximate Entropy (ApEn) for detecting the onset and end of epileptic seizures.
• To evaluate the efficacy of ApEn in analyzing EEG signals for seizure detection.
• To investigate the influence of ApEn parameters on detection performance.

Main Methods:

• Utilized Approximate Entropy (ApEn) analysis on electroencephalogram (EEG) data.
• Developed a method for detecting epileptic seizure onset and termination using ApEn.
• Performed a preliminary analysis to determine optimal ApEn parameters.

Main Results:

• The ApEn-based method demonstrated promising results in detecting epileptic seizure events.
• Analysis of EEG signals using ApEn effectively identified seizure onsets and endings.
• The study highlighted the significant impact of ApEn parameters on the method's behavior.

Conclusions:

• Approximate Entropy (ApEn) shows potential for efficient epileptic seizure detection from EEG.
• Further investigation into ApEn parameter determination is necessary for robust clinical application.
• Establishing a consistent methodology for parameter selection will enhance detection accuracy.