カオス解析による心拍変動からの睡眠段階の推定

概要

論文の詳細を見る
In this paper, chaos analysis was used and examined for the possibility of sleep stage estimation from a non-stationary element included in long time heart rate variability. The experiments were conducted using nine normal, healthy males aged between 21 and 38. Each subject's EEC, EOG, EMG, and ECG were measured during six-hour periods between 12 midnight and 6 a.m. Furthermore, by identifying the sleep stage of each subject based on the international standard, and extracting the most remarkable REM and SWS stages, chaos analysis was conducted from the RRI in each sleep stage. As a result, the max lyapunov exponent and KS entropy in REM increased in a statistically significant manner (p < 0.01). In addition, through the results of statistical analysis between all sleep stages and a max lyapunov exponent, KS entropy indicated the possibility of estimating the sleep stage by using a non-stationary element in the heart rate variability. Future work will improve the accuracy of sleep stage estimation using multiple parameters including a non-stationary element in order to realize a sleep stage monitoring device that can be used easily in normal daily life.
社団法人日本生体医工学会の論文
2003-12-10