Sleeping has a significant role in refreshing the brain and body from fatigue. Due to the aging, however, obtaining the sleep of high quality gradually becomes difficult. In Japan, which now faces the problem of aging society, it is important to promote the maintenance and enhancement of the health by securing the sleep of high quality. For that end, unconstrained monitoring of sleeping conditions on a daily basis is expected to be helpful. In this study, comb filters were applied to bio-signal obtained by pneumatic method to extract with high accuracy the heartbeats and the body movements, which then were used to define an index that shows the condition of a REM sleep and another that shows the depth of a sleep. Also determined was a function used for obtaining the incidence ratio and the standard deviation of the extracted elements for each sleep stage. The function was determined for each age-group. Using these indices and the functions, an algorithm to surmise the sleeping stage of a subject was proposed. The mean rate of agreement between sleep stage categorized into 6 sleep stages or 5 sleep stages or 3 sleep stages determined by the proposed method and that by the R-K method, the world standard method, were 51.6%, 56.2% and 77.5%. Mean of kappa statistics were 0.29, 0.39 and 0.48, respectively.
抄録全体を表示