Analysis of Features Extracted from EEG Epochs by Discrete Wavelet Decomposition and Hilbert Transform for Sleep...

Area of Science:

• Biomedical Signal Processing
• Neuroscience
• Sleep Medicine

Background:

• Sleep apnea (SA) is a prevalent sleep disorder characterized by abnormal breathing.
• Electroencephalogram (EEG) signals exhibit changes during apneic events, offering potential for diagnostic analysis.
• Previous research explored automatic sleep apnea classification using EEG, but further refinement is needed.

Purpose of the Study:

• To analyze statistical features from EEG epochs using combined Discrete Wavelet Transform (DWT) and Hilbert Transform (HT).
• To develop an automated classifier for differentiating normal breathing, obstructive sleep apnea (OSA), and central sleep apnea (CSA) based on EEG features.

Main Methods:

• Extraction of statistical features from EEG epochs.
• Application of Discrete Wavelet Transform (DWT) and Hilbert Transform (HT) for feature extraction.
• Utilizing a feedforward neural network with 17+7 neurons across two hidden layers for classification.
• Selection of the 30 most discriminative features for the classification model.

Main Results:

• The developed classifier achieved 73.9% accuracy on the training set and 77.3% accuracy on the testing set.
• Analysis highlighted the significance of theta, beta, and gamma brain waves in distinguishing sleep apnea types.
• The combined DWT and HT approach proved effective in extracting relevant EEG features.

Conclusions:

• Automated classification of sleep apnea using EEG-derived features is feasible.
• The study demonstrates the utility of DWT and HT in analyzing EEG for sleep apnea detection.
• Theta, beta, and gamma brain wave activity are important indicators for classifying sleep apnea subtypes.