JP2016140729A - Sleep stage estimation device, method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately estimate a sleep stage even when a heart rate is strongly affected by an impact other than brain activity.SOLUTION: In a learning phase, a sleep state estimation device calculates a heart rate fluctuation feature amount for each sleep stage from electrocardiographic data of a subject and stores them in a heart rate fluctuation table 211, calculates appearance frequency and transition frequency of the sleeve stage according to time lapse, and stores them in an appearance frequency table 212 and a transition frequency table 213 respectively. In an estimation phase, the sleep state estimation device estimates the sleep stage of an estimation object person on the basis of the heart rate fluctuation feature amount calculated from the electrocardiographic data of the estimation object person and sleep stage model data stored in the respective tables 211, 212, and 213.SELECTED DRAWING: Figure 3

Description

  The present invention relates to a sleep stage estimation device, method, and program for estimating a person's sleep stage.

  The sleep stage is an index representing the depth of sleep, and is defined in 2 to 6 stages according to international standards. The sleep progress diagram showing the sleep stages in time series is important information in grasping the sleep state in the medical and healthcare fields. The sleep stage is determined every time interval (about 30 seconds) based on the R & K method, which is an inspection determination rule, by a professional engineer using electroencephalogram data or electromyogram data obtained by a polygraph examination. However, since this polygraph test has a large physical load on the user, in recent years, research on a sleep stage estimation technique using heart rate data that can be acquired by a measurement device with a small physical load has been actively conducted.

  The conventional sleep stage estimation method using the heart rate data, for example, generates a plurality of feature quantities related to the sleep stage from the heart rate data, and estimates using machine learning (support vector machine, random forest, etc.). (See Non-Patent Document 1, for example).

Xiao, M et al., “Sleep stages classification based on heart rate variability and random forest” Biomedical Signal Processing and Control, p624-633, (2013).

  However, the autonomic nervous activity reflected in the heartbeat changes due to influences other than brain activity during sleep, such as blood pressure adjustment, body temperature adjustment, body movement, and breathing. For this reason, with the technique described in Non-Patent Document 1, it is difficult to estimate the sleep stage obtained from brain activity in perfect agreement from the heartbeat.

  The present invention has been made paying attention to the above circumstances. The purpose of the present invention is to provide a sleep stage estimation device that can estimate the sleep stage more accurately even when the heartbeat is strongly influenced by other than brain activity. It is to provide a method and a program.

  In order to achieve the above object, a first aspect of the present invention is a heart rate variability memory that stores heart rate variability information obtained by aggregating heart rate variability feature quantities calculated for each first unit time from a set of sleep electrocardiograms for each sleep stage. Frequency of appearance of sleep stages for each first unit time for each time zone obtained by dividing the time axis of the sleep electrocardiogram by a second unit time longer than the first unit time A model database, an appearance frequency storage unit that stores information, and a transition frequency storage unit that stores transition frequency information in which the frequency of sleep stage transitions for each first time is tabulated for each of the divided time zones Prepare as. Then, from the sleep electrocardiogram acquired from the subject to be estimated, for each time zone divided by the second unit time, the feature value of the heart rate variation for each first time is calculated, and the calculated heart rate variation The probability of the sleep stage for each first unit time of the subject based on the temporal change in the feature amount of the subject and the information stored in the heart rate variability storage unit, the appearance frequency storage unit, and the transition frequency storage unit The distribution is estimated.

The following aspects can be considered in the first aspect of the present invention.
The first aspect obtains a set of sleep electrocardiograms including time information at the time of sleep given the correct label of the sleep stage for each first unit time given as the sleep electrocardiogram set of the trainee, Means for calculating a heart rate variability feature amount for each of the first unit times from the acquired set of sleep electrocardiograms, totaling the calculated heart rate variability feature amount for each sleep stage, and storing the sum in a heart rate variability storage unit; Means for totalizing the appearance frequency of the sleep stage for each first unit time and storing it in the appearance frequency storage unit for each time zone obtained by dividing the time axis of the sleep electrocardiogram set by the second unit time; For each time zone divided by the second unit time, a means for counting the frequency of sleep stage transitions for each first time and storing it in a transition frequency storage unit is further provided. .

  In the second aspect, when calculating the feature value of heart rate variability, the mean value of heart rate variability RRI (RR Interval), the standard deviation of RRI, and the mean square value of the difference between the preceding and following RRIs are calculated as the heart rate variability feature quantity. Square root (RMSSD), occurrence ratio (pNN50) of the number of times the temporal difference between the preceding and following RRIs is greater than or equal to the preset time value, the length L of the long side of the shape drawn by the Lorentz plot of the preceding and following RRI, RRI Sympathetic nerve activity index (CSI: Cardiac Sympathetic Index) calculated from short side lengths T, L and T of the shape drawn by the Lorentz plot, Parasympathetic nerve activity index (CVI: Cardiac Vagal Index) calculated from L and T, At least one of the high-frequency component (HF) and the low-frequency component (LF) of the power spectrum waveform of RRI, or other feature amount calculated from these feature amounts is calculated.

  According to a second aspect of the present invention, there is provided means for acquiring a set of sleep electrocardiograms including time information at the time of sleep, which is given as a sleep electrocardiogram set of a trainee and provided with a correct label of a sleep stage every first unit time And calculating a heart rate variability feature amount for each first unit time from the acquired set of sleep electrocardiograms, and summing the calculated heart rate variability feature amount for each sleep stage and storing it in the heart rate variability storage unit. And the frequency of appearance of the sleep stage for each first unit time for each time zone obtained by dividing the time axis of the sleep electrocardiogram set by a second unit time longer than the first unit time. The means for storing the totaled appearance frequency information in the appearance frequency storage unit, and the frequency of sleep stage transitions for each of the first time periods are totaled for each of the divided time zones, and the totalized transitions Transition information is stored in the frequency information It is obtained so as to include a means for storing.

  According to the first aspect of the present invention, based on the heart rate variability feature amount calculated from the electrocardiogram data of the estimation subject and the heart rate variability information, the appearance frequency information, and the transition frequency information stored in advance in the model database, The sleep stage of the estimation target person is estimated. For this reason, even if the heartbeat changes due to influences other than brain activity during sleep, such as blood pressure adjustment, body temperature adjustment, body movement, and breathing, it is possible to accurately estimate the sleep stage.

  According to the first aspect of the first aspect of the present invention, sleep stage estimation is provided by including means for generating heart rate variability information, appearance frequency information, and transition frequency information in the learning phase and storing them in the model database. It is possible to generate and store optimal heartbeat variability information, appearance frequency information, and transition frequency information autonomously in the apparatus.

  According to the 2nd mode concerning the 1st viewpoint of this invention, a suitable thing can be chosen from many kinds of feature-values as a feature-value of heart rate variability.

  On the other hand, according to the second aspect of the present invention, the calculated heart rate variability feature amount is aggregated for each sleep stage, stored in the heart rate variability storage unit, and the sleep stage for each first unit time aggregated for each time zone The appearance frequency information is stored in the appearance frequency storage unit, and information indicating the frequency of sleep stage transitions for each first time, which is aggregated for each time period, is stored in the transition frequency storage unit. That is, the estimation model can be created by the sleep stage estimation device.

  That is, according to the present invention, it is possible to provide a sleep stage estimation device, method, and program capable of estimating the sleep stage more accurately even when the heartbeat is strongly influenced by other than brain activity.

The figure which showed an example of the appearance frequency of a sleep stage according to time slot | zone. The figure which shows an example of the appearance probability of a sleep stage transition pattern. The block diagram which shows the function structure of the sleep stage estimation apparatus which concerns on one Embodiment of this invention. The flowchart which shows the process sequence and process content in the learning phase of the sleep stage estimation apparatus shown in FIG. The flowchart which shows the process sequence and process content in the estimation phase of the sleep stage estimation apparatus shown in FIG. The figure which shows an example of the heart rate fluctuation table classified by sleep stage with which the sleep stage estimation apparatus shown in FIG. 3 is provided. The figure which shows an example of the sleep stage appearance frequency table classified by time with which the sleep stage estimation apparatus shown in FIG. 3 is provided. The figure which shows an example of the sleep stage transition frequency table classified by time with which the sleep stage estimation apparatus shown in FIG. 3 is provided. The figure which shows an example of the observation method of time series heart rate data, and the output method of a sleep stage estimated value. The figure which shows an example of a Lorenz plot. The figure which displayed the list of the kind of heart rate variability feature-value.

Embodiments according to the present invention will be described below with reference to the drawings.
(principle)
The present invention focuses on the fact that the appearance frequency and transition probability of the sleep stage change according to the elapsed time after falling asleep, so that even if the heartbeat is strongly influenced by other than brain activity, a more likely sleep stage can be estimated. It is a thing. For example, although there is a low possibility of sudden awakening after falling asleep and falling into a deep sleep again, the state may be estimated with a probability of 0.52 when estimated only with the heartbeat. In the present invention, the estimation accuracy is improved by lowering the estimation probability for such a rare transition or when the estimation confidence level is low.

  FIG. 1 shows an example of the result of calculating the number of appearances of each sleep stage every 30 minutes using the sleep stage data for 45 persons. As is clear from this example, the number of appearances of REM (REM sleep) in 30 minutes from the start of sleep (FIG. 1 (a)) is 180 minutes and 360 minutes from sleep shown in FIGS. 1 (b) and 1 (c), respectively. Less than the number of appearances of REM in 30 minutes after elapse.

  FIG. 2 also shows an example of a result of calculating the appearance probability of each sleep stage transition pattern every 30 minutes using the sleep stage data for 45 people. As shown in the figure, the transition probability from REM to NREM (non-REM sleep) in 30 minutes after the onset of sleep (FIG. 2 (a)) is 150 minutes from the sleep shown in FIGS. 2 (b) and 2 (c), respectively. It turns out that it becomes higher than the transition probability from REM to NREM in 30 minutes after and 300 minutes later.

  Therefore, in the present invention, the appearance frequency and transition probability of the sleep stage according to the elapsed time are incorporated into the sleep stage generation model by the learning phase, and the feature quantity of the user's heartbeat data is obtained in the estimation phase, and the feature quantity of the heartbeat data is obtained. The sleep stage is estimated based on the sleep stage generation model. This makes it possible to estimate the sleep stage more accurately even when the heartbeat is strongly influenced by other than brain activity.

[First Embodiment]
(Constitution)
FIG. 3 is a block diagram showing a functional configuration of the sleep stage estimation apparatus according to the embodiment of the present invention.
The sleep stage estimation apparatus 1 according to the present embodiment includes, for example, a server computer or a personal computer, and includes a control unit 10, a storage unit 20, and an input / output interface unit 30. Among these, the input / output interface unit 30 takes in the electrocardiogram data output from the electrocardiograph 2 and operates data and display data between the input unit 3 such as a keyboard or a touch panel and the display unit 4 such as a liquid crystal device, respectively. I / O is performed.

  The storage unit 20 uses a nonvolatile memory capable of writing and reading, such as an HDD (Hard Disk Drive) and an SSD (Solid State Drive), as a storage medium, and a storage unit necessary for implementing this embodiment. As shown, a model database 21 and an estimated data storage unit 22 are provided.

  The model database 20 includes a heart rate variability table 211, an appearance frequency table 212, and a transition frequency table 213. The heart rate variability table 211 is used for storing the heart rate feature quantity level and the appearance frequency in association with the heart rate feature vector identification information (heart rate feature vector ID) for each sleep stage. An example is shown in FIG. The appearance frequency table 212 is used for storing the sleep stages appearing in the time zone and the appearance frequency for each time zone. An example is shown in FIG. The transition frequency table 213 is used to store sleep stage transition patterns that occur in each time zone and their appearance frequencies. FIG. 8 shows an example. The estimated data storage unit 22 is used to store data representing the sleep stage estimation result.

  The control unit 10 includes a central processing unit (CPU) and a memory. As a control processing function necessary for carrying out the present embodiment, the heart rate variability feature quantity calculation unit 11 and the sleep stage It has a heart rate variability analysis unit 12, an hourly sleep stage appearance frequency analysis unit 13, an hourly sleep stage transition frequency analysis unit 14, a sleep stage probability estimation unit 15, and an estimated data output control unit 16. Each of these control processing functions is realized by causing the CPU to execute a program stored in a program memory (not shown).

  The heart rate variability feature quantity calculation unit 11 calculates the feature quantity of the heart rate variability based on the electrocardiogram data acquired from the electrocardiograph 2 via the input / output interface unit 30. Specifically, RRI (R-RInterval) representing the peak interval is detected from the electrocardiogram waveform, and the detected RRI is divided every n seconds to obtain a frame length N [sec] and a slide width n [sec]. Then, the heart rate variability feature vector ht = {ht1, ht2,..., Htk} is calculated. Here, t indicates an elapsed time, and is counted every n seconds such as t = {n, 2 * n, 3 * n,.

  In the learning phase, the heart rate variability analysis unit 12 by sleep stage calculates the ht calculated by the heart rate variability feature quantity calculation unit 11 based on the time information t, and the sleep stage every n seconds input by the input unit 3. And the distribution of each heart rate feature amount at each sleep stage is calculated. Then, a process of storing the distribution of each heartbeat feature amount calculated for each sleep stage in the heartbeat fluctuation table 211 in the model database 21 is performed.

  In the learning phase, the hourly sleep stage appearance frequency analysis unit 13 divides the time series data of the sleep stage correct answer labels input every n seconds in the input unit 3 every q minutes (for example, 30 minutes). The number of appearances of each sleep stage i in each time slot m is calculated. Then, the calculated number of appearances of each sleep stage i is stored in the appearance frequency table 212 in the model database 21.

  In the learning phase, the hourly sleep stage transition frequency analysis unit 14 performs each sleep stage transition every n seconds in each time zone m divided every q minutes, in the same manner as the hourly sleep stage appearance frequency analysis unit 13. The number of occurrences of is calculated. Then, the calculated number of appearances of each sleep stage transition is stored in the transition frequency table 213 in the model database 21.

  In the estimation phase, the sleep stage probability estimation unit 15 calculates the heart rate variability feature vector calculated by the heart rate variability feature amount calculation unit 11 based on the electrocardiogram data input from the electrocardiograph 2, and the model database 21. Using the information stored in the tables 211, 212, and 213, the sleep stage St every n seconds is estimated. And the process which stores the data showing the estimation result in the estimation data storage part 22 is performed.

  The estimated data output control unit 16 reads data representing the sleep stage estimation result in the corresponding time zone of the corresponding user from the estimated data storage unit 22 in accordance with the output instruction input in the input unit 3. Then, display data for displaying the read estimated data is generated, and the generated display data is output to the display unit 4 via the input / output interface unit 30 and displayed.

(Operation)
Next, the sleep stage estimation operation by the apparatus configured as described above will be described.
(1) Learning phase In the learning phase, while measuring the electrocardiogram waveform of the subject, a measurer such as a medical worker visually observes the sleep stage of the subject every n seconds, and the observation result is expressed as “the correct label of the sleep stage” Is input by the input unit 3. Then, the sleep stage estimation device 1 calculates a feature value of heart rate variability from the electrocardiogram waveform and stores it in the heart rate variability table 21, and also sleep stages according to time based on the correct answer label of the input sleep stage. This is done by calculating the appearance frequency and the sleep stage transition frequency and storing the calculated values in the appearance frequency table 212 and the transition frequency table 213, respectively. FIG. 4 is a flowchart showing a processing procedure and processing contents of the sleep stage estimation device 1 in this learning phase.

  When measurement in the learning phase is started, first, the electrocardiogram data of the subject output from the electrocardiograph 2 in step S11 is taken in via the input / output interface unit 30 under the control of the heart rate variability feature quantity calculation unit 11. Then, based on the electrocardiogram data taken in step S12, the feature amount of heart rate variability is calculated as follows.

  That is, first, RRI (R-RInterval) representing the peak interval of the electrocardiogram waveform is detected from the electrocardiogram data, and the detected RRI is divided every n seconds. Then, a heartbeat variability feature vector ht = {ht1, ht2,..., Tk} is calculated as a frame length N [sec] and a slide width n [sec]. Here, t indicates an elapsed time, and is counted every n seconds such as t = {n, 2 * n, 3 * n,. A specific example is shown in FIG.

  As the heart rate variability feature amount, for example, the average value of heart rate variability RRI (RRInterval), the standard deviation of RRI, the square root of the mean square value of the difference between preceding and following RRI (RMSSD), and the temporal difference between preceding and following RRI is 50 mm. Occurrence ratio (pNN50) of the number of times that became more than second, the length L of the long side of the shape drawn by the preceding and following RRI Lorentz plots, and the length T, L and T of the short side of the shape drawn by the RRI Lorentz plot Sympathetic nerve activity index (CSI: Cardiac Sympathetic Index), parasympathetic nerve activity index (CVI: Cardiac Vagal Index) obtained from L and T, high frequency component (HF) of RRI power spectrum waveform, or low frequency component (LF) ) Or other feature quantities calculated from these feature quantities. An example of the RRI Lorentz blot is shown in FIG. 10, and the definition of each heart rate variability feature amount is shown in FIG. Note that the detailed calculation method of the heart rate variability feature amount is described in Non-Patent Document 1, for example.

  The value of each feature value of the heart rate variability feature vector ht is preferably normalized to a range of, for example, 0 to 1 and then made discrete at an L level. For example, the first element of the feature vector when t = n (first n-second interval) is represented as hn1 = {0, 1,..., L}.

  Subsequently, the heart rate variability analysis unit 12 for each sleep stage analyzes the heart rate variability for each sleep stage based on the heart rate variability feature vector ht calculated by the heart rate variability feature quantity calculation unit 11 as follows. That is, the calculated heart rate variability feature vector ht is associated with the correct answer label of the sleep stage every n seconds input by the input unit 3 based on the time information t. Then, the distribution of each heartbeat feature amount in each sleep stage is calculated.

  Next, in step S13, the appearance frequency of each sleep stage for each time zone (m minutes) is totalized by the hourly sleep stage appearance frequency analysis unit 13 as follows. That is, first, the time-series data of the sleep stage correct answer label every n seconds input in the input unit 3 is divided every q minutes (for example, 30 minutes), and the appearance of each sleep stage i in each divided time zone m. The number of times is calculated. At this time, m = 1 is represented as a section from 0 minute to (q × 1), and m = 2 is represented as a section from (q × 1) to (q × 2). Then, the calculated number of appearances of each sleep stage i is stored in the appearance frequency table 212 together with information indicating the sleep stage in association with the time zone, for example, as shown in FIG.

  Subsequently, in step S14, the appearance frequency of each sleep stage transition pattern for each time zone (m minutes) is totalized by the hourly sleep stage transition frequency analysis unit 14 as follows. That is, first, in each time slot m divided every q minutes, the number of appearances of each sleep stage transition every n seconds is calculated.

  For example, in a certain time zone m, when the sleep stage of the first 0 to n seconds interval is “Wake” and the sleep stage of the n to 2 n seconds interval is “REM sleep (REM)”, the n to 2 n seconds interval The sleep stage transition is counted as a transition pattern of “wake (WAKE) → REM sleep (REM)”. In addition, the transition pattern of the sleep stage in the first 0-n seconds section does not need to be counted. Then, the number of appearances of each sleep stage transition calculated as described above is stored in the transition frequency table 213 together with information indicating the sleep stage transition pattern in association with the time zone, for example, as shown in FIG.

  Finally, in step S15, the heart rate variability feature amount for each sleep stage is tabulated by the heart rate variability analysis unit 12 for each sleep stage, and the appearance frequency that is the calculation result is, for example, as shown in FIG. Along with the feature amount level, it is associated with each sleep stage and stored in the sleep stage heart rate variability table 211. For example, if the total time when the correct answer label of the sleep stage is “Wake” is t_wake = {n, 5 * n, 8 * n}, the count becomes ht_wake, 1 = 1 for all times. Is memorized. At this time, since the discrete value of the heart rate variability feature quantity is 0 ≦ l ≦ L, each value is counted.

(2) Estimation Phase In the estimation phase, only the electrocardiogram data of the estimation target person is measured, the heart rate variability feature amount calculated from the measured electrocardiogram data, the heart rate variability table 211, the appearance frequency table 212, and the transition frequency table 213. The sleep stage St of the estimation subject is estimated every n seconds based on the model data stored in. FIG. 5 is a flowchart showing the processing procedure and processing contents.

  First, in step S21, the heart rate variability feature quantity calculation unit 11 takes in the electrocardiogram data in the sleep of the estimation target person from the electrocardiograph 2, and the input unit 3 measures the sleep stage heart rate variability analysis unit 12. The correct answer label of the sleep stage every n seconds input by the person is captured. In step S22, the heart rate variability feature quantity calculation unit 11 calculates a heart rate variability feature quantity every n seconds from the captured electrocardiogram data. The calculation method is the same as in the learning phase.

  Next, in step S23, under the control of the sleep stage probability estimation unit 15, the sleep stage of the estimation target person is estimated as follows. That is, first, in step S23, sleep stage variables every n seconds are initialized at random. For example, the Gibbs sampling counter g is initialized with 1.

  Next, in step S24, the sleep stage variable every n seconds is updated by Gibbs sampling, and the following processing is repeatedly performed until it is determined in step S25 that the Gibbs sampling counter g has reached the Gibbs sampling repetition number.

(A) As initial values, time t = 0 and sleep stage S0 = “Wake” are set.
(B) While the time t is smaller than the sleep time T, the following processing is executed.
(1) The sleep stage S_ {t, g} is changed to P (St = i | ht, St-1 using the sleep stage St-1 at time t-1, the time zone Ct, and the heart rate variability feature quantity ht at time t. = J, Ct = m).
(2) Time t = t + 1.
(C) The Gibbs sampling counter g is incremented (g = g + 1).

  Subsequently, in step S26, for each time t, the appearance distribution of each sleep stage is totaled from g sleep stage sampling history sets {S_ {t, g}} and output as a sleep stage probability distribution at time t. When outputting the sleep stage, the sleep stage that is the mode value in the sampling history set {S_ {t, g}} is stored in the estimation data storage unit 22 as an estimation result.

P (St = i | ht, St-1 = j, Ct = m) is obtained by the following equation.

Here, θ _mi represents the probability that the sleep stage becomes i in the time zone m, and is calculated by the following equation. M _mi represents the number of times the sleep stage i appears in the time zone m.

Λ _mji represents the probability that the sleep stage transitions from j to i in the time zone m, and is calculated by the following equation. M _mji represents the number of times that a state in which the sleep stage transitions from j to i appears in the time zone m.

Furthermore, φ _ikhtk represents the appearance probability of each heartbeat characteristic amount in the sleep stage i, and is calculated as follows.

When the heart rate variability feature vector kth element h _tk newly obtained at time t at sleep stage i is l, M _ikl refers to the model database and h _tk satisfies the value l Represents the number of occurrence frequencies.

  Note that data representing the estimation result of the sleep stage of the estimation target person stored in the estimation data storage unit 22 is read from the estimation data storage unit 22 under the control of the estimation data output control unit 16. Display data is generated based on the read estimated data, and the display data is supplied to the display unit 4 via the input / output interface unit 30 and displayed. An example of the display result is shown in the lower part of FIG.

(Effect of embodiment)
As described above in detail, in one embodiment, in the learning phase, the heart rate variability feature amount for each sleep stage is calculated from the electrocardiogram data of the subject and stored in the heart rate variability table 211, and the appearance of the sleep stage according to the passage of time The frequency and the transition frequency are obtained and stored in the appearance frequency table 212 and the transition frequency table 213, respectively. In the estimation phase, based on the heart rate variability feature amount calculated from the electrocardiogram data of the estimation target person and the sleep stage model data stored in the tables 211, 212, and 213, the sleep stage of the estimation target person is determined. I try to estimate.

  Therefore, even when the heartbeat changes due to influences other than brain activity during sleep, such as blood pressure adjustment, body temperature adjustment, body movement, breathing, and the like, the sleep stage can be accurately estimated. In addition, by providing the learning phase, it is possible to construct an optimal heart rate fluctuation table 211, appearance frequency table 212, and transition frequency table 213.

[Other Embodiments]
The present invention is not limited to the above embodiment. For example, in the embodiment, the case where the storage unit 20 is provided in the sleep stage estimation apparatus 1 has been described as an example. However, the storage unit 20 is provided in a database server provided in the cloud, and the sleep stage estimation apparatus 1 and the database are provided. Data may be written and read by communicating with the server.

  In the embodiment, the heart rate variability feature amount is calculated based on the electrocardiogram data obtained by the electrocardiograph. However, the pulse wave is measured and the heart rate variability feature amount is calculated based on the measurement data. You may do it.

  Further, the data representing the estimation result may be transmitted to another terminal or server via the communication line. Other control functions provided in the control unit, processing procedure and processing contents by the control function, display of the estimation results The format and the like can be variously modified and implemented without departing from the gist of the present invention.

  In short, the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. Further, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, you may combine suitably the component covering different embodiment.

  DESCRIPTION OF SYMBOLS 1 ... Sleep stage estimation apparatus, 2 ... Electrocardiograph, 3 ... Input part, 4 ... Display part, 10 ... Control unit, 11 ... Heart rate fluctuation feature-value calculation part, 12 ... Heart rate fluctuation analysis part according to sleep stage, 13 ... Time Separate sleep stage appearance frequency analysis unit, 14 ... Sleep stage transition frequency analysis unit according to time, 15 ... Sleep stage probability estimation unit, 16 ... Estimation data output control unit, 20 ... Storage unit, 21 ... Model database, 22 ... Estimation data storage 211, heart rate variability table, 212, appearance frequency table, 213, transition frequency table, 30, input / output interface unit.

Claims (7)

A heart rate variability storage unit that stores heart rate variability information obtained by summing heart rate variability features calculated for each first unit time from a set of sleep electrocardiograms for each sleep stage;
Appearance frequency information is stored for each time zone obtained by dividing the time axis of the sleep electrocardiogram by a second unit time longer than the first unit time. An appearance frequency storage unit,
For each of the divided time zones, a transition frequency storage unit that stores transition frequency information obtained by tabulating the frequency of sleep stage transitions for each first time;
A heart rate variability feature amount calculating means for calculating a feature amount of heart rate variability for each first time for each time zone divided by the second unit time from a sleep electrocardiogram acquired from a subject to be estimated;
The first unit time of the subject based on the temporal change in the calculated feature value of the heart rate variability and the information stored in the heart rate variability storage unit, the appearance frequency storage unit, and the transition frequency storage unit A sleep stage estimation apparatus comprising: estimation means for estimating a probability distribution of each sleep stage. Means for obtaining a set of sleep electrocardiograms including time information at the time of sleep given a correct answer label of the sleep stage for each first unit time given as a sleep electrocardiogram set of the trainer;
Means for calculating a heart rate variability feature amount for each first unit time from the acquired set of sleep electrocardiograms, and summing the calculated heart rate variability feature amount for each sleep stage and storing it in the heart rate variability storage unit When,
For each time zone obtained by dividing the time axis of the sleep electrocardiogram set by the second unit time, the appearance frequency information of the sleep stage for each first unit time is totaled, and the total appearance frequency information is displayed. Means for storing in the frequency storage unit;
And further comprising means for counting the frequency of transition of the sleep stage for each of the divided times and storing the tabulated transition frequency information in the transition frequency storage unit for each of the divided time zones. The sleep stage estimation apparatus according to claim 1, wherein:   The heart rate variability feature quantity calculating means includes, as a heart rate variability feature quantity, an average value of heart rate variability RRI (RR Interval), a standard deviation of RRI, a root mean square of a difference between preceding and following RRIs (RMSSD), and an RRI preceding and following Occurrence rate (pNN50) of the number of times that the time difference between them is equal to or greater than the preset time value, the length L of the long side of the shape drawn by the preceding and following RRI Lorentz plots, the short side of the shape drawn by the RRI Lorentz plot Sympathetic nerve activity index (CSI: Cardiac Sympathetic Index) calculated from lengths T, L, and T, parasympathetic nerve activity index (CVI: Cardiac Vagal Index) calculated from L and T, high frequency components of RRI power spectrum waveform The sleep stage estimation device according to claim 1 or 2, wherein at least one of (HF) or low frequency component (LF) or other feature amount calculated from these feature amounts is calculated. . Means for obtaining a set of sleep electrocardiograms including time information at the time of sleep given a correct answer label of the sleep stage for each first unit time given as a trainer's sleep electrocardiogram set;
Means for calculating a heart rate variability feature amount for each first unit time from the acquired set of sleep electrocardiograms, and summing the calculated heart rate variability feature amount for each sleep stage and storing the sum in a heart rate variability storage unit; ,
For each time zone obtained by dividing the time axis of the sleep electrocardiogram set by a second unit time longer than the first unit time, the appearance frequency of the sleep stage for each first unit time is totaled, and the total Means for storing the generated appearance frequency information in the appearance frequency storage unit;
And a means for counting the frequency of sleep stage transitions for each of the divided time periods and storing the tabulated transition frequency information in a transition frequency storage unit. Sleep stage estimation device. A heart rate variability storage unit storing heart rate variability information obtained by aggregating heart rate variability feature quantities calculated for each first unit time from a set of sleep electrocardiograms for each sleep stage, and a time axis of the sleep electrocardiogram as the first unit time For each time zone divided by a longer second unit time, an appearance frequency storage unit that stores appearance frequency information obtained by totaling the appearance frequencies of sleep stages for each first unit time, and each divided time zone A sleep method estimation method executed by a sleep stage estimation device including a model database having a transition frequency storage unit that stores transition frequency information obtained by tabulating the frequency of sleep stage transitions for each first time. And
From the sleep electrocardiogram obtained from the subject to be estimated, for each time zone divided by the second unit time, a process of calculating a feature value of heart rate variability for each first time;
The first unit time of the subject based on the temporal change in the calculated feature value of the heart rate variability and the information stored in the heart rate variability storage unit, the appearance frequency storage unit, and the transition frequency storage unit A sleep stage estimation method comprising: estimating a probability distribution of each sleep stage. The process of obtaining a set of sleep electrocardiograms including time information at the time of sleep given the correct answer label of the sleep stage for each first unit time given as the sleep electrocardiogram set of the trainer;
From the acquired set of sleep electrocardiograms, a heart rate variability feature amount is calculated for each first unit time, the calculated heart rate variability feature amount is aggregated for each sleep stage, and the aggregated heart rate variability feature amount is calculated. Storing in the heart rate variability storage unit;
For each time zone obtained by dividing the time axis of the sleep electrocardiogram set by the second unit time, the appearance frequency of the sleep stage for each first unit time is totaled, and the totaled appearance frequency information is Storing in the appearance frequency storage unit;
For each divided time zone, totaling the frequency of transition of the sleep stage for each first time, and storing the totalized transition frequency information in the transition frequency storage unit A characteristic sleep stage estimation method.   The program which makes the computer with which the said sleep stage estimation apparatus performs the process which each means with which the sleep stage estimation apparatus in any one of Claim 1 thru | or 4 comprises comprises.