Multimedia Monitoring System of Obstructive Sleep Apnea via a Deep Active Learning Model

Obstructive sleep apnea (OSA) is one of the most commonsleep-related breathing disorders. Nearly 1 billion people worldwide suffer from it, causing serious health effects and social burden. However, traditional monitoring systems often fall short in terms of cost and accessibility. In this article, we first propose a deep active learning model to detect OSA events from electrocardiogram (ECG). We then designed and developed a prototype of OSA monitoring system using an ECG sensor and smartphone, in which our OSA detection algorithm is implemented and validated. Experiments show that we achieve accuracy of 92.15% while using 40% of labeled data, significantly reducing the cost of labeling and maximizing the performance. According to detection results and health-related multimedia signals, we provide OSA risk level and medical advice to the users. We believe that the multimedia monitoring system can efficiently help diagnose OSA, which could lead to effective intervention strategies and better sleep care.

Obstructive sleep apnea (OSA) is a prevalent sleep-related breathing disorder, estimated to affect nearly one billion people worldwide. Patients with OSA commonly suffer from fatigue and daytime sleepiness, which not only affects their life quality, but also increases the risk for traffic accidents and workplace accidents. Polysomnography (PSG) is the golden standard for diagnosis of OSA, which records 16 physical signals and is carried out in the exclusive sleep laboratory. PSG not only requires specific hospital settings, but is also invasive and may interrupt with sleep, factors that make OSA diagnosis a costly and uncomfortable procedure. Consequently, 80%–90% of the OSA cases are estimated to be undiagnosed and untreated. Therefore, reliable alternative methods that use fewer sensors and provide better availability for OSA detection should be developed.