Detection of Sleep Apnea Based on ECG Recording, Pulse, and Heart Rate Variability Academic Article in Scopus uri icon

abstract

  • Sleep apnea, a respiratory disorder affecting a significant portion of the global population, is characterized by interruptions in breathing during sleep, leading to serious health issues like cardiovascular diseases and reduced quality of life. Early and accurate detection of sleep apnea is essential for effective treatment. This project aims to design a system for identifying sleep apnea events using Electrocardiogram (ECG) and Photoplethysmography (PPG) signals. The system involves stages of signal acquisition, data preprocessing, feature extraction, and apnea event identification. Using a database of ECG and PPG signals from subjects with varying severities of sleep apnea, preprocessing techniques were applied to remove noise and artifacts. Feature extraction methods based on frequency and time variability analysis, indicative of apnea events, were employed. Severity-based algorithms identified apnea events. Results: The patients with mild apnea show a significant decrease in the percentage amplitude of the PPG with median reduction (13.45%) vs recovery (11.85%) with p = 0.000002. When evaluating measurements of the RR intervals before, during and after the apnea event, showed the sustained decrease of said intervals in the moderate (p = 0.0094) and severe (p = 0.0057) groups. When evaluating the sympathetic and parasympathetic branches by analyzing the Heart Rate Variability (HRV) in the frequency domain, the increase in sympathetic activity evaluated in LF band was evident in the severe group (p = 0.0342). Conclusions: The noninvasive methods described here may be useful in identifying individuals at risk of developing apnea. This could lead to the development of a screening system to detect sleep apnea. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

publication date

  • January 1, 2025