AI-driven sleep apnea screening with overnight blood oxygen saturation: current practices and future directions

Frontiers in Digital Health, Journal Year: 2025, Volume and Issue: 7

Published: April 9, 2025

Sleep apnea is one of the most common sleep disorders, which, if left untreated, may have severe health consequences in long term. Many patients remain non-diagnosed due to lacking access medical tests. In recent years, portable and wearable sensors that measure blood oxygen saturation (SpO2) are becoming affordable for daily use, they open door accessible screening context everyday life. To learn about advancement SpO2-based screening, we conducted a survey published studies. We searched databases including Springer, Science Direct, Web Science, ACM Digital Library, IEEE Xplore using keywords "sleep apnea" AND ("SpO2" OR "blood saturation") ("machine learning" "deep learning"). After 835 results, included 31 publications full-text review. Analysis shows studies consist three main categories: (1) individual events detection, (2) apnea-hypopnea index prediction, (3) severity classification. found two significant research gaps: lack sufficient diverse publicly available datasets, absence standardized protocols data collection, signal preprocessing, model bench marking. Future should focus on addressing these gaps enhance effectiveness reliability AI-driven methods SpO2 signals.

Language: Английский

Implementing deep learning on edge devices for snoring detection and reduction

Computers in Biology and Medicine, Journal Year: 2024, Volume and Issue: 184, P. 109458 - 109458

Published: Nov. 23, 2024

Language: Английский