Real-time sleep quality assessment using single-lead ECG and multi-stage SVM classifier

Area of Science:

• Biomedical Engineering
• Sleep Medicine
• Signal Processing

Background:

• Objective sleep quality assessment is crucial for diagnosing sleep disorders.
• Current methods like Polysomnography (PSG) are intrusive, expensive, and require expert scoring.
• There is a need for accessible, automated, and non-intrusive sleep monitoring solutions.

Purpose of the Study:

• To develop and validate a home-based, automated system for objective sleep quality assessment.
• To utilize Electrocardiogram (ECG) signals for measuring key sleep indices.
• To offer a cost-effective and non-intrusive alternative to traditional polysomnography.

Main Methods:

• A multi-stage Support Vector Machines (SVM) classifier was employed.
• The system analyzes Electrocardiogram (ECG) data in 30-second segments.
• Three sleep quality indices were measured: Sleep Efficiency Index, Delta-Sleep Efficiency Index, and Sleep Onset Latency.

Main Results:

• The developed system provides objective sleep quality assessment using ECG.
• It successfully measures Sleep Efficiency Index, Delta-Sleep Efficiency Index, and Sleep Onset Latency.
• The method offers a viable alternative to conventional, more intrusive sleep studies.

Conclusions:

• A home-based, automated ECG system can effectively assess sleep quality.
• This approach offers a practical and less intrusive method for sleep monitoring.
• The system has the potential to improve accessibility to sleep quality assessment.