Automatic Wake and Deep-Sleep Stage Classification Based on Wigner-Ville Distribution Using a Single...

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Accurate sleep stage classification is crucial for diagnosing sleep disorders.
• Manual scoring of polysomnography (PSG) data is time-consuming and subjective.
• Automated methods are needed to improve the efficiency and objectivity of sleep analysis.

Purpose of the Study:

• To develop and validate an automated method for classifying wakefulness and deep sleep (N3) stages.
• To utilize single-channel EEG signals and time-frequency analysis for sleep classification.
• To compare the automated method's performance against expert scoring based on American Academy of Sleep Medicine (AASM) standards.

Main Methods:

• Employed Wigner-Ville Distribution (WVD) for time-frequency analysis of EEG signals.
• Calculated EEG energy in specific frequency bands (δ, θ, α).
• Utilized Particle Swarm Optimization (PSO) to determine optimal thresholds for distinguishing sleep stages.

Main Results:

• The automated classification achieved high sensitivity, accuracy, and kappa coefficient.
• The method demonstrated reliable differentiation between wakefulness and N3 sleep stages.
• Results closely aligned with manual scoring by sleep technicians according to AASM criteria.

Conclusions:

• The proposed automated method offers an intuitive and effective approach for sleep stage classification.
• The algorithm shows promise for reliable sleep staging, potentially improving diagnostic efficiency.
• Future work aims to extend the algorithm for classifying all sleep stages.