Noncontact Detection of Sleep Apnea Using Radar and Expectation-Maximization Algorithm

Sleep apnea syndrome requires early diagnosis because this syndrome can lead to a variety of health problems. If sleep apnea events can be detected in a noncontact manner using radar, we can then avoid the discomfort caused by the contact-type sensors that are used in conventional polysomnography. This study proposes a novel radar-based method for accurate detection of sleep apnea events. The proposed method uses the expectation-maximization algorithm to extract the respiratory features that form normal and abnormal breathing patterns, resulting in an adaptive apnea detection capability without any requirement for empirical parameters. We conducted an experimental quantitative evaluation of the proposed method by performing polysomnography and radar measurements simultaneously in five patients with the symptoms of sleep apnea syndrome. Through these experiments, we show that the proposed method can detect the number of apnea and hypopnea events per hour with an error of 4.8 times/hour; this represents an improvement in the accuracy by 1.8 times when compared with the conventional threshold-based method and demonstrates the effectiveness of our proposed method.

Noncontact Respiratory Measurement for Multiple People at Arbitrary Locations Using Array Radar and Respiratory-Space Clustering

In this study, we develop a noncontact measurement system for monitoring the respiration of multiple people using millimeter-wave ultrawideband array radar. To separate the radar echoes of multiple people, conventional techniques cluster the radar echoes in the time, frequency, or spatial domain. Focusing on the measurement of the respiratory signals of multiple people, we propose a method called respiratory-space clustering in which individual differences in the respiratory rate are effectively exploited to accurately resolve the echoes from human bodies. The proposed respiratory-space clustering can separate echoes, even when people are located close to each other. In addition, the proposed method can be applied when the number of targets is unknown and can accurately estimate the number of people and their positions. We perform measurements under two scenarios involving five and seven participants to verify the performance of the proposed method, and quantitatively evaluate the estimation accuracy of the number of people and the respiratory intervals. The experimental results show that the root-mean-square error in estimating the respiratory interval is 172 ms on an average. The proposed method improves the estimation accuracy of the number of people by 85.0% compared to the conventional method, demonstrating the high-precision measurement of the respiration of several people.