A Novel Sleep Staging Method Based on EEG and ECG Multimodal Features Combination.

Area of Science:

• Neuroscience
• Cardiology
• Biomedical Engineering
• Sleep Medicine

Background:

• Accurate sleep staging is crucial for diagnosing sleep disorders and related conditions.
• Existing methods often overlook the combined dynamics of brain and heart activity during sleep.
• Integrating electroencephalography (EEG) and electrocardiography (ECG) data offers a more comprehensive approach to sleep analysis.

Purpose of the Study:

• To develop and evaluate a generalized EEG and ECG multimodal feature combination for efficient and accurate sleep stage classification.
• To investigate the contributions of distinct physiological signals (EEG vs. ECG) to classifying different sleep stages.
• To compare the performance of the proposed multimodal approach against existing state-of-the-art methods.

Main Methods:

• Utilized hybrid features from multichannel EEGs (multiscale entropy, intrinsic mode function) and ECGs (heart rate variability, sample entropy).
• Applied dimensionality reduction techniques: max-relevance/min-redundancy and principal component analysis.
• Classified features using four traditional machine learning models, evaluating performance with Macro-F1, macro-geometric mean, and Cohen kappa on the ISRUC-S3 dataset.

Main Results:

• EEG features were more influential for classifying wake stages, while ECG features were more critical for deep sleep stages.
• The multimodal feature combination achieved a peak accuracy of 84.3% and a Cohen kappa value of 0.794 using a support vector machine classifier.
• The proposed method demonstrated superior accuracy and efficiency compared to other state-of-the-art sleep staging techniques.

Conclusions:

• The integration of multimodal EEG and ECG features provides a promising and effective strategy for accurate sleep staging.
• This approach enhances the understanding of sleep dynamics by leveraging both brain and heart activity.
• The developed method offers a significant advancement in sleep disorder diagnosis and management.