Classification of Sleep Apnea Severity by Electrocardiogram Monitoring Using a Novel Wearable Device.

Area of Science:

• Cardiology
• Sleep Medicine
• Biomedical Engineering

Background:

• Sleep apnea (SA) diagnosis relies on polysomnography (PSG), which is costly and complex.
• Simpler, long-term monitoring solutions are needed beyond initial diagnosis.
• The diagnostic value of single-parameter wearable devices for SA severity remains under investigation.

Purpose of the Study:

• To evaluate a novel wearable electrocardiogram (ECG) acquisition system (ECG belt) for classifying sleep apnea severity.
• To assess the suitability of heart rate variability (HRV) analysis from the ECG belt for SA classification.
• To compare the performance of the ECG belt against traditional patched ECG during PSG.

Main Methods:

• Developed a wearable ECG belt for continuous, whole-night signal acquisition.
• Analyzed heart rate variability (HRV) using data from the ECG belt, with and without pre-filtering.
• Employed and compared various classification algorithms, selecting Support Vector Machine (SVM) for its performance.
• Recorded simultaneous ECG signals from 241 patients using both the ECG belt and standard patched ECG during PSG.

Main Results:

• The ECG belt demonstrated classification accuracy (72%) comparable to patched ECG (74%).
• Sensitivity and specificity for the ECG belt were 70% and 74%, respectively.
• The highest accuracy was achieved in differentiating between no/mild SA and moderate/severe SA.
• Obstructive sleep apnea was present in 65% of the study cohort, with a median AHI of 21.

Conclusions:

• The wearable ECG belt effectively captures signals comparable to traditional patched ECG.
• The ECG belt is a viable tool for assessing sleep apnea severity.
• This technology holds potential for convenient, long-term sleep apnea monitoring and follow-up.