Sleep Quality Detection Based on EEG Signals Using Transfer Support Vector Machine Algorithm.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Sleep disorders are increasingly prevalent due to modern lifestyle pressures, impacting quality of life and productivity.
• Effective monitoring and evaluation of sleep quality are crucial for public health.
• Sleep staging provides valuable insights for assessing sleep quality.

Purpose of the Study:

• To propose and validate a novel method for sleep quality detection.
• To analyze sleep quality using electroencephalography (EEG) signals.
• To improve the accuracy of sleep staging for better sleep assessment.

Main Methods:

• Preprocessing of electroencephalography (EEG) signals.
• Feature extraction using discrete wavelet transform (DWT).
• Classification of extracted features using a transfer support vector machine (TSVM) algorithm.

Main Results:

• The proposed method was evaluated on 60 datasets from the National Sleep Research Resource Library.
• Performance was assessed using sensitivity, specificity, and accuracy.
• The TSVM classifier exhibited significantly higher classification performance compared to other algorithms.

Conclusions:

• The developed method effectively detects and analyzes sleep quality.
• The transfer support vector machine (TSVM) approach proves highly effective for sleep staging.
• This research validates a robust approach for sleep quality assessment using EEG data.