Automatic seizure detection using diffusion distance and BLDA in intracranial EEG.

Area of Science:

• Neurology
• Biomedical Engineering
• Signal Processing

Background:

• Epilepsy affects approximately 1% of the global population, necessitating effective monitoring and diagnostic tools.
• Automatic seizure detection systems are crucial for managing epilepsy, improving patient care, and facilitating research.
• Intracranial electroencephalogram (EEG) recordings provide high-resolution data for epilepsy analysis.

Purpose of the Study:

• To propose a novel method for automatic seizure detection in intracranial EEG recordings.
• To develop an accurate and reliable system for identifying epileptic seizures.
• To evaluate the performance of the proposed method using extensive clinical data.

Main Methods:

• EEG recordings were segmented into 4-second epochs.
• Wavelet decomposition was applied to EEG epochs across five scales.
• Diffusion distances were extracted as features, and Bayesian linear discriminant analysis (BLDA) was employed as the classifier.

Main Results:

• The system achieved an average sensitivity of 94.99% and specificity of 98.74%.
• A low false-detection rate of 0.24 per hour was recorded.
• The method demonstrated high performance in detecting seizures from long-term intracranial EEG data.

Conclusions:

• The proposed diffusion distance-based seizure detection system is effective for long-term EEG monitoring.
• The system offers a promising tool for the automatic detection of epileptic seizures.
• This approach contributes to improved epilepsy diagnosis and patient management through accurate automated analysis.