System for discriminating sleep state

• US 5,101,831 A
• Filed: 07/12/1991
• Issued: 04/07/1992
• Est. Priority Date: 07/07/1989
• Status: Expired due to Term

• Current Assignee: Matsushita Electric Works Ltd (Panasonic Holdings Corporation)
• Original Assignee: Matsushita Electric Works Ltd (Panasonic Holdings Corporation)
• Inventors: Hagiwara, Hiroshi, Michimori, Akihiro, Koyama, Emi
• Primary Examiner(s): Kamm, William E.
• Assistant Examiner(s): PONTIUS, KEVIN LANE
• Application Number:
  US07/729,843
• Time in Patent Office: 270 Days
• Field of Search: 128/731, 128/732, 128/733, 128/687, 128/689, 128/690, 128/721, 128/736, 128/716, 600/26, 340/575
• US Class Current: 600/500
• CPC Class Codes: A61B 5/024 Detecting, measuring or recording pulse rate or heart rate A61B5/0205, A61B5/021 take precedence
• CPC Class Codes: A61B 5/0255 Recording instruments specially adapted therefor
• CPC Class Codes: A61B 5/4812 Detecting sleep stages or cycles
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: A61M 2021/0083 especially for waking up
• CPC Class Codes: A61M 2230/06 Heartbeat rate only
• CPC Class Codes: A61M 2230/42 Rate
• CPC Class Codes: G04G 13/02 at preselected times, e.g. alarm clocks
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

A system for discriminating sleep state of human body provides variation indexes representing variation of the biological signal on the basis of first variation amount denoting a tendency of increment in time series of measured biological signal from the starting time of the measurement and second variation amount denoting the temporal variation of the biological signal, and discriminates different sleep state on the basis of distribution density of the variation indexes exceeding a predetermined threshold, whereby NREM and REM sleep periods in particular can be reliably discriminated, to be sufficiently contributive to a transmission of sleep information that ensures a comfortable awakening.

Claims

1. A system for discriminating a sleep state of and selectively waking a subject, the system comprising means for setting a measuring time, means for measuring, as a biological signal of the subject, at least one biological signal per unit time selected from the group consisting of pulse rate and respiration rate, a variation index computing means for providing variation indexes C(t) representing variation of the biological signal H(t) in the form of a linear combination of first variation amount I(t) denoting a tendency of increment in time series of said biological signal H(t) from a starting time of the measuring to a time when the biological signal is obtained with second variation amount D(t) denoting a standard variation of the biological signal H(t) during said measuring time, and a sleep index operating means for providing sleep indexes allowing a NREM sleep state and all other sleep states to be discriminated from one another on the basis of said variation indexes C(t) continuously exceeding a predetermined threshold C(th).

2. A system according to claim 1, which further comprises means for setting a reference value H(r) which can be regarded as said biological signal at a stably resting and awakening state on the basis of the biological signal H(t) at initial stage of the measurement, and means for presuming a sleep-on set time when the biological signal is below said predetermined threshold C(th) set on the basis of said reference value H(r).

3. A system according to claim 2, which further comprises a sleep state discriminating means which discriminates, when said variation index C(t) denotes a period other than said NREM sleep period when the biological signal H(t) is above said reference value H(r).

4. A system according to claim 1, wherein said variation index computing means comprises means for obtaining a trend line Tr(t) denotes a trend of the variation with elapsing time on the basis of a moving average A(t) of said biological signal H(t), said trend line being obtained in such that a first rhythm curve R1 is obtained with the minimum values of the moving average A(t) at every unit time in respect of passing direction of time, a second rhythm curve R2 is obtained with the minimum values of the moving average A(t) at every unit time in respect of receding direction of time and the larger value between the first and second rhythm curves R1, R2 at each time is plotted as a point of the trend line Tr(t).

5. A system according to claim 3, which further comprises means for generating an awakening stimulus to be given to the user, means for setting an awakening time, and means for controlling operation of said awakening stimulus generating means.

6. A system according to claim 5, which further comprises means for judging a state of resleeping on the basis of said biological signal H(t) measured after an actuation of said awakening stimulus generating means and actuating again the awakening stimulus generating means.

7. A system for discriminating a sleep state of and selectively waking a subject, the system comprising means for setting a measuring time, means for measuring as a biological signal of the subject at least one biological signal per unit time selected from the group consisting of pulse rate and respiration rate, a variation index computing means for providing variation indexes C(t) representing variation of the biological signal H(t) in the form of a linear combination of first variation amount I(t) denoting a tendency of increment in time series of said biological signal H(t) from a starting time of the measuring to a time when the biological signal is obtained with second variation amount D(t) denoting a standard variation of the biological signal H(t) during said measuring time, said variation index computing means for performing a regression operation on said biological signal, and a sleep index operating means for providing sleep indexes allowing a NREM sleep state and all other sleep states to be discriminated from one another on the basis of said variation indexes C(t) continuously exceeding a predetermined threshold C(th).