Sleep estimation system, sleep estimation method, and program

Abstract

To easily estimate sleep stages.SOLUTION: A sleep estimation device includes: a first acquisition unit for acquiring blood flow data; a generation unit for generating a frequency spectrum of the blood flow data by executing frequency analysis processing on the blood flow data; and a first determination unit for determining the sleep stage of a subject on the basis of the frequency spectrum.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to the estimation of the sleep stage of a subject.

Background Art

[0002] Patent Document 1 describes a technique for detecting a sleep stage.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

[0004] A sleep estimation system according to an aspect of the present disclosure includes a determination unit that determines the sleep stage of a subject based on the frequency spectrum of the blood flow data of the subject, and a notification unit that executes a notification process based on the determination result of the determination unit. The notification unit notifies the subject to wake up when the determination unit determines that at least the frequency spectrum includes a region including a frequency band of 0.2 Hz or more and 0.3 Hz or less and includes a predetermined shape, or after a predetermined time has elapsed since the determination. The predetermined shape included in the region is a convex upward waveform shape including a first intensity that is the maximum intensity in the region.

[0005] Furthermore, a sleep estimation system according to one aspect of the present disclosure includes a determination unit that determines the sleep stage of a subject based on the frequency spectrum of the subject's blood flow data, and a notification unit that performs notification processing based on the determination result of the determination unit, wherein the determination unit determines whether the frequency spectrum of the blood flow data corresponds to the correct answer when it is given to a trained model that has been trained using training data in which the correct answer label based on biological information different from the blood flow data is associated with a frequency spectrum having a predetermined shape in a region including a frequency band of 0.2 Hz or more and 0.3 Hz or less, and the notification unit notifies the subject to wake up when the determination unit determines that it corresponds to the correct answer, or after a predetermined time has elapsed since the determination, wherein the predetermined shape included in the region is an upwardly convex waveform shape including a first intensity which is the maximum intensity in the region, and the biological information is electroencephalogram data detected by an electroencephalograph.

[0006] Furthermore, a sleep estimation method according to one aspect of the present disclosure is a sleep estimation method performed in a sleep estimation system, comprising: a determination step in which a determination unit in the sleep estimation system determines the sleep stage of a subject based on the frequency spectrum of the subject's blood flow data; and a notification step in which a notification unit in the sleep estimation system notifies the subject to wake up when it determines in the determination step that at least the frequency spectrum includes a predetermined shape in a region including a frequency band of 0.2 Hz or more and 0.3 Hz or less, or after a predetermined time has elapsed since the determination, wherein the predetermined shape included in the region is an upwardly convex waveform shape including a first intensity which is the maximum intensity in the region.

[0007] Furthermore, a sleep estimation method according to one aspect of the present disclosure is a sleep estimation method performed in a sleep estimation system, comprising: a determination step in which a determination unit in the sleep estimation system determines the sleep stage of a subject based on the frequency spectrum of the subject's blood flow data; and a notification step in which a notification unit in the sleep estimation system performs a notification process based on the determination result in the determination step, wherein the determination step determines whether the frequency spectrum of the blood flow data corresponds to the correct answer when the frequency spectrum of the blood flow data is given to a trained model trained using training data in which the correct answer label based on biological information different from the blood flow data is associated with a frequency spectrum having a predetermined shape in a region including a frequency band of 0.2 Hz or more and 0.3 Hz or less; and the notification step provides a notification to the subject prompting awakening when it is determined in the determination step that the answer corresponds to the correct answer, or after a predetermined time has elapsed since the determination, wherein the predetermined shape included in the region is an upwardly convex waveform shape including a first intensity which is the maximum intensity in the region. The aforementioned biological information is electroencephalogram (EEG) data detected by an electroencephalograph. [Brief explanation of the drawing]

[0008] [Figure 1] This is a block diagram showing a schematic configuration example of the sleep estimation system of Embodiment 1. [Figure 2] This graph shows an example of blood flow waveform data detected by a blood flow meter. [Figure 3] This graph shows an example of a frequency spectrum generated by performing a Fourier transform on blood flow waveform data. [Figure 4] This graph shows an example of electrocardiogram waveform data detected by an electrocardiograph. [Figure 5] This graph shows an example of a frequency spectrum generated by performing a Fourier transform on electrocardiogram waveform data. [Figure 6] This image shows an example of the results of wavelet transform processing applied to blood flow waveform data. [Figure 7] This flowchart shows an example of the processing flow by the sleep estimation device of Embodiment 1. [Figure 8] This block diagram shows a schematic configuration example of the sleep estimation system of Embodiment 2. [Modes for carrying out the invention]

[0009] The following describes the determination (estimation) of the sleep stage of a subject in this disclosure. First, the principle underlying the determination of the sleep stage of a subject in this disclosure will be explained. Note that in this specification, when "A~B" is written, it means "A or greater, and B or less." In addition, in this specification, blood flow waveform data is given as an example of blood flow data used to determine the sleep stage.

[0010] 〔principle〕 Figure 2 is a graph showing an example of blood flow waveform data detected by a blood flow meter. In Figure 2, the vertical axis represents a value proportional to the blood flow rate per unit time [unit: dimensionless], and the horizontal axis represents the measurement time [unit: min]. The blood flow meter can acquire raw waveform data W1 and processed waveform data W2 as blood flow waveform data, as shown in Figure 2. Processed waveform data W2 is waveform data obtained by processing raw waveform data W1 to make it easier to obtain the peak of the R wave. Processed waveform data W2 is generated, for example, by performing a smoothing process on raw waveform data W1. In processed waveform data W2, the time interval between adjacent peaks (blood flow rates at the positions indicated by the inverted triangles in the figure) represents the heart rate interval (RRI). Similarly, in the raw heart rate waveform data W12 shown in Figure 4, the time interval between adjacent peaks (blood flow rates at the positions indicated by the inverted triangles in the figure) represents the heart rate interval (RRI).

[0011] A blood flow meter that detects blood flow waveform data is a sensor that can detect blood flow waveform data indicating the blood flow of a subject by receiving scattered light generated by irradiating the blood vessels of the subject with light. The blood flow meter may include a light-emitting unit that irradiates the blood vessels of the subject with light, and a light-receiving unit that receives the scattered light.

[0012] Generally, when laser light is shone on a fluid, the shone laser light is scattered by (i) scatterers contained in the fluid that move with the fluid, and (ii) stationary objects such as pipes that carry the fluid, generating scattered light. Generally, scatterers cause inhomogeneity in the complex refractive index of the fluid.

[0013] Scattered light generated by a scattering object moving with the fluid undergoes a wavelength shift due to the Doppler effect, which depends on the fluid velocity of the scattering object. On the other hand, scattered light generated by a stationary object does not undergo a wavelength shift. These scattered lights interfere with each other, resulting in the observation of optical beats.

[0014] A blood flow meter may be a sensor that utilizes this phenomenon. In other words, a blood flow meter may be a laser Doppler blood flow meter that detects the light beats caused by scattered light generated in the blood fluid by irradiating the blood vessels of a subject with laser light, as blood flow waveform data.

[0015] More specifically, the processor in the blood flow meter may analyze the acquired light signal and calculate frequency analysis data showing the signal intensity for each frequency of the light signal. For example, the processor may perform analysis on the acquired light signal using a method such as FFT (Fast Fourier Transform).

[0016] The processor may further generate blood flow waveform data showing the fluctuation pattern of the subject's blood flow based on the frequency analysis data. For example, the processor may calculate the sum of first moments X of the acquired frequency analysis data as blood flow waveform data. More specifically, the processor may calculate the sum of first moments X of the acquired frequency analysis data using the following formula. The processor uses the following formula to calculate the sum of first moments X in a certain frequency band (e.g., 1 to 20 kHz). X = Σfx × P(fx) You can also calculate this. Here, "fx" is the frequency, and "P(fx)" is the signal strength value at frequency fx.

[0017] The primary moment sum X calculated by the processor based on the frequency analysis data can be a value proportional to the blood flow volume of the subject. The processor may generate pattern data indicating the variation pattern of the blood flow volume of the subject over time by calculating the primary moment sum X for each of the plurality of frequency analysis data. Further, the processor may generate blood flow waveform data using the data included in a part of the frequency bands among the data included in the frequency analysis data. The processor can output the generated blood flow waveform data.

[0018] The blood flow waveform data may include data related to at least one of, for example, the cardiac output and the coefficient of variation of vasomotion (vasomotion) in addition to the blood flow volume. The cardiac output is the amount of blood pumped out by one heartbeat of the heart. Vasomotion is a spontaneous and rhythmic contraction and relaxation movement of blood vessels. The coefficient of variation of vasomotion is a value indicating the variation of the blood flow volume generated based on vasomotion as a variation.

[0019] Further, the blood flow waveform data may include a pulse wave.

[0020] FIG. 3 is a graph showing an example of a frequency spectrum generated by performing a Fourier transform process on the blood flow waveform data as shown in FIG. 2. The vertical axis indicates the intensity of the frequency spectrum [unit: dB], and the horizontal axis indicates the frequency [unit: Hz]. The Fourier transform process is an example of a frequency analysis process and is a process for generating a frequency spectrum of waveform data that does not include temporal changes.

[0021] In FIG. 3, frequency spectra FW1 and FW2 corresponding to each sleep stage are shown. The frequency spectrum FW1 is a frequency spectrum generated as a result of performing a Fourier transform process on the raw waveform data W1. The frequency spectrum FW2 is a frequency spectrum generated as a result of performing a Fourier transform process on the inter-beat interval (RRI) of the processed waveform data W2.

[0022] The above sleep stages can be classified into three stages: wakefulness, REM sleep, and non-REM sleep. Non-REM sleep can be further classified into stages 1 (N1), 2 (N2), and 3 (N3), in order from lightest to most intense. REM sleep is sleep accompanied by rapid eye movement (REM). Non-REM sleep is sleep without rapid eye movement.

[0023] This classification is based on electroencephalogram (EEG) data detected by an EEG device attached to the subject. EEGs are divided into four types, in order of wavelength from longest to shortest: beta waves, alpha waves, theta waves, and delta waves. Beta waves are, for example, EEGs with a frequency of approximately 38-14 Hz. Alpha waves are, for example, EEGs with a frequency of approximately 14-8 Hz. Theta waves are, for example, EEGs with a frequency of approximately 8-4 Hz. Delta waves are, for example, EEGs with a frequency of approximately 4-0.5 Hz.

[0024] A person is asleep when theta and delta waves are dominant compared to beta and alpha waves. Here, "dominant" means that a certain wave makes up a large proportion of the measured brainwaves. It is known that the dominant brainwaves change periodically within the range of theta and delta waves during sleep. Furthermore, if the proportion of theta waves in the brainwaves is below a certain level, the person is in REM sleep, and if the proportion of theta waves is above a certain level, and delta waves are dominant, the person is in non-REM sleep. Stage 1 is, for example, a state where alpha waves are 50% or less and various low-amplitude frequencies are mixed. Stage 2 is, for example, a state where low-amplitude theta and delta waves appear irregularly, but there are no high-amplitude slow waves. Stage 3 is, for example, a state where slow waves of 75 μV and below account for 20% or more. A state where slow waves of 75 μV and below account for 50% or more may be called Stage 4.

[0025] In Figure 3, sleep stages are indicated as follows: wakefulness is "WK," REM sleep is "RM," non-REM sleep stage 1 is "N1," non-REM sleep stage 2 is "N2," and non-REM sleep stage 3 is "N3."

[0026] As shown in the frequency spectra FW1 and FW2 in Figure 3, the frequency spectra obtained from subjects in sleep stages 2 and 3 show a significant change in intensity in the 0.2-0.3 Hz frequency band (a predetermined frequency band). In other words, in the said frequency spectrum, the intensity in a portion of the 0.2-0.3 Hz frequency band, specifically in the first range R1, is greater than the intensity in the second range R2 (excluding the first range R1) by a predetermined value or more. Hereinafter, the intensity in the first range R1 will be referred to as the first intensity, and the intensity in the second range R2 will be referred to as the second intensity.

[0027] The first intensity may be, for example, the maximum intensity in the frequency band of 0.2 to 0.3 Hz. The second intensity may be, for example, the maximum intensity in the second range R2 other than the first range R1 that includes this maximum intensity (e.g., approximately ±0.02 Hz around the maximum intensity, but within the frequency band of 0.2 to 0.3 Hz). The predetermined value should be set to a magnitude sufficient to identify the presence of a characteristic waveform Sh in the 0.2 to 0.3 Hz frequency band that is not present in adjacent frequency bands, for example, based on experiments. The characteristic waveform Sh may be, for example, an upwardly convex shape and a somewhat broad waveform (e.g., a waveform with a full width at half maximum of 0.03 Hz or more). Figure 3 shows examples of the first range R1 and the second range R2 in the frequency spectra FW1 and FW2 corresponding to Stage 3.

[0028] On the other hand, in frequency spectra obtained from subjects in wakefulness, REM sleep, and stage 1 sleep stages, no significant intensity changes were observed in the 0.2-0.3 Hz frequency band, nor was the characteristic waveform Sh described above observed.

[0029] After diligent research, the inventors discovered that when a frequency spectrum showing a significant intensity change (characteristic waveform Sh) in the 0.2-0.3 Hz frequency band is obtained, there is a high probability that the subject is in stage 2 or 3 sleep. In other words, the inventors found that when a frequency spectrum with a characteristic waveform Sh is obtained, there is a high probability that the subject is in stage 2 or 3 sleep. Furthermore, the inventors found that a significant intensity change is observed in the 0.2-0.3 Hz frequency band in the frequency spectrum of blood flow waveform data detected by the aforementioned blood flow meter (e.g., laser Doppler blood flow meter). Based on these findings, the inventors have developed a sleep estimation device capable of improving the accuracy of determining the sleep stage of a subject.

[0030] (Comparison with electrocardiogram waveform data) Furthermore, there are differences between the electrocardiogram (ECG) waveform data detected by the electrocardiograph and the blood flow waveform data detected by the blood flow meter, as explained below.

[0031] Figure 4 is a graph showing an example of electrocardiogram waveform data detected by an electrocardiograph. In Figure 4, the vertical axis represents heart rate intensity [unit: dB], and the horizontal axis represents measurement time [unit: min]. Figure 4 shows raw heart rate waveform data W11 and W12 as electrocardiogram waveform data.

[0032] Figure 5 is a graph showing an example of a frequency spectrum generated by performing a Fourier transform on electrocardiogram waveform data. The vertical axis represents the intensity of the frequency spectrum [unit: dB], and the horizontal axis represents the frequency [unit: Hz].

[0033] Figure 5 shows the frequency spectra FW11 and FW12 corresponding to each sleep stage. Frequency spectrum FW11 is the frequency spectrum generated by performing a Fourier transform on the raw waveform data W11. Frequency spectrum FW12 is the frequency spectrum generated by performing a Fourier transform on the heart rate interval (RRI). The sleep stages corresponding to frequency spectra FW11 and FW12 are identified based on electroencephalogram (EEG) data acquired from an EEG device attached to the subject.

[0034] As shown in Figure 5, no significant intensity changes (characteristic waveform Sh) were observed in the 0.2-0.3 Hz frequency band in the frequency spectra FW11 and FW12 obtained by converting electrocardiogram waveform data detected from subjects in sleep stages 2 and 3. The frequency spectrum FW12 corresponding to sleep stages 2 and 3 has an upward convex shape in the 0.2-0.3 Hz frequency band. However, the frequency spectrum FW12 corresponding to sleep stage 1 has a similar shape to the frequency spectrum FW12 corresponding to stages 2 and 3. Therefore, it cannot be concluded that the frequency spectrum FW12 corresponding to stages 2 and 3 has a significant intensity change in the 0.2-0.3 Hz frequency band.

[0035] After diligent research, the inventors discovered that a significant intensity change in the 0.2-0.3 Hz frequency band is a phenomenon specific to the frequency spectrum of blood flow waveform data. Based on this, the inventors found that by using the frequency spectrum converted from blood flow waveform data, rather than electrocardiogram waveform data, to determine the sleep stage of a subject, it is highly likely that it is possible to accurately determine whether the subject is in sleep stage 2 or 3.

[0036] (Regarding frequency bands) Due to differences in the blood flow meter used and individual differences in the subjects, there may be some variability in the frequency band in which the above-mentioned significant intensity changes are observed. Considering this, it is quite possible that significant intensity changes not observed in the electrocardiogram waveform data may be observed in the frequency band of FW1 and FW2, corresponding to stages 2 or 3, for example, in the 0.15 to 0.4 Hz range. In the following explanation, we will assume that the frequency band in which the above-mentioned significant intensity changes are observed is 0.2 to 0.3 Hz.

[0037] [Wavelet transform processing] The principle described above was explained using a frequency spectrum obtained by performing a Fourier transform as a frequency analysis process. However, the sleep stage of the subject may be determined based on a frequency spectrum obtained by performing a wavelet transform as a frequency analysis process. The wavelet transform is an example of a time-frequency analysis process. A time-frequency analysis process is a process that generates a frequency spectrum of waveform data that includes temporal changes. The wavelet transform is a process that generates a frequency spectrum of waveform data using a mother wavelet, which is an arbitrary reference waveform.

[0038] The mother wavelet used in wavelet transform processing is defined as follows: In the following equation, "t" is the time variable, "a" is the scale parameter (a parameter that expands or shrinks the mother wavelet in the time axis direction), and "b" is the translate parameter (a parameter that translates the mother wavelet in the time axis direction).

[0039]

number

[0040] Furthermore, the function for performing wavelet transform processing is defined as follows. In the following equation, "f(t)" represents the waveform data and "*" represents the complex conjugate. By substituting the mother wavelet obtained by adjusting the values ​​of "a" and "b" into the following equation, the frequency spectrum of the waveform data can be generated.

[0041]

number

[0042] By using wavelet transform processing, the intensity in the 0.2-0.3 Hz frequency band (hereinafter referred to as the target intensity) can be relatively emphasized compared to other frequency bands. As explained in the principle described above, when a frequency spectrum with a characteristic waveform Sh in the 0.2-0.3 Hz frequency band is obtained, there is a high probability that the subject is in stage 2 or 3 sleep. Therefore, by emphasizing the target intensity using wavelet transform processing, it is possible to increase the accuracy of determining whether or not the subject is in stage 2 or 3 sleep.

[0043] In the wavelet transform process, a mother wavelet set to emphasize the target intensity as described above may be used. A mother wavelet with increased target intensity can be set by adjusting the values ​​of "a" and "b" above. Alternatively, Morlet may be used as the mother wavelet. In this case, the scale parameter "a" has the relationship "ω = 2π / a" and represents the local angular frequency. Since the angular frequency "ω" can be expressed as "ω = 2πf", the wavelet transform process may be performed with this "f (frequency)" part set to 0.2 to 0.3 Hz. The target intensity may be the intensity of the entire frequency band from 0.2 to 0.3 Hz, or the intensity in a part of the frequency band (e.g., the first range R1).

[0044] Furthermore, the wavelet transform processing of the blood flow waveform data can generate intensity change data showing the temporal change in intensity in each frequency band within a predetermined time period. The predetermined time period should be set to a time that allows for accurate determination of the subject's sleep stage, for example, through an experiment. In this embodiment, the predetermined time period may be set to, for example, 2.5 minutes.

[0045] Unlike the Fourier transform, the wavelet transform can generate a frequency spectrum that includes the temporal variation of intensity, thus allowing for a larger amount of data than the Fourier transform. Generally, in the generation of trained models described below, the more data (the more features to train), the more accurate the trained model that can produce output data. Therefore, using intensity variation data is effective when generating trained models.

[0046] Figure 6 shows an example of the results of wavelet transform processing on blood flow waveform data. Figure 6 shows an image generated by performing wavelet transform processing on the blood flow waveform data (raw waveform data W1) that forms the basis of the stage 3 frequency spectrum FW1 shown in Figure 3. This image is an example of intensity change data with the target intensity emphasized. Hereafter, images showing intensity change data will be referred to as wavelet images.

[0047] In Figure 6, the vertical axis represents frequency [unit: Hz], and the horizontal axis represents time [unit: min]. The grayscale in the wavelet image in Figure 6 represents intensity [unit: dB]. In other words, a wavelet image is data showing the intensity distribution of the frequency spectrum in a plane defined by frequency and time.

[0048] In the wavelet image of this embodiment, frequency bands (intensity) may be represented by a color gradient. In the wavelet image, for example, low frequency bands can be represented by cool colors, and high frequency bands by warm colors. Specifically, the frequency bands may be represented in order from lowest to highest as dark blue, blue, light blue, yellow-green, light yellow-green, yellow, orange, and red. As long as the intensity distribution is visible in the wavelet image, the frequency bands may be represented by other colors or by grayscale. Figure 6 shows an example of a grayscale image of a wavelet image represented by the above colors.

[0049] In the wavelet image of Figure 6, the first region AR1, distributed along the time axis at and around 0.2 Hz, shows a higher intensity than the intensity in the frequency band adjacent to the first region AR1. Specifically, in the frequency band of approximately 0.2 Hz ± approximately 0.05 Hz within the first region AR1, an intensity band shown in red is distributed along the time axis, and around this intensity band, intensity regions shown in orange, yellow, light yellow-green, and yellow-green are distributed. In Figure 6, a portion of the intensity band shown in red is indicated by reference numeral 101. Also, a portion of the intensity regions shown in orange, yellow, light yellow-green, and yellow-green is indicated by reference numeral 102. On the other hand, in the adjacent frequency band, intensity regions shown in light blue, blue, and dark blue are mainly distributed, but intensity regions shown in red, orange, and yellow are not distributed. In Figure 6, a portion of the intensity regions shown in light blue, blue, and dark blue is indicated by reference numeral 103.

[0050] Furthermore, as shown in Figure 6, in the wavelet image, a sawtooth-shaped intensity band is formed along the time axis in frequency bands higher than 0.2–0.3 Hz. In the wavelet image of Figure 6, a sawtooth-shaped intensity band is formed in the second region AR2 (frequency band of approximately 0.7 Hz and above). The second region AR2 has a lower intensity than the first region AR1 in the frequency band of approximately 0.9–1.0 Hz, and the intensity gradually decreases towards frequency bands below 0.9 Hz and above 1.0 Hz.

[0051] The sawtooth-shaped intensity bands shown above represent the intensity distribution corresponding to heart rate. This intensity distribution takes the form of bands along the time axis during sleep, and the more light the sleep, the more the band shape becomes distorted. The intensity distribution corresponding to heart rate is an intensity distribution that cannot be obtained through Fourier transform processing. By using wavelet images in the generation of the pre-trained model described below, it is possible to generate a pre-trained model that also takes heart rate into account.

[0052] Figure 6 shows an example of a wavelet image with enhanced target intensity. However, it should be noted that even in wavelet images without enhanced target intensity, the intensity of the first region AR1 is higher than the intensity in adjacent frequency bands.

[0053] [Generating a pre-trained model] In determining the sleep stage of a subject, a pre-trained model (approximator) for determining the subject's sleep stage can be used. The pre-trained model is a mathematical model (a neural network including input, hidden, and output layers) that mimics the neurons of the human brain's nervous system, trained to determine the user's sleep stage. Any mathematical model that can generate a pre-trained model capable of determining the subject's sleep stage is acceptable. The mathematical model may be, for example, a Convolutional Neural Network (CNN), a Recurrent Neural Network (RNN), or a Long Short-Term Memory (LSTM).

[0054] Learning refers to adjusting the strength of connections between units and the bias of those connections so that the correct calculation results are output from the output layer. In this embodiment, when learning is performed, learning data is input to the input layer. In the hidden layer, calculations based on calculation data are performed on the learning data, and the calculation results in the hidden layer are output as output data from the output layer. The training data and the output data are compared, and the calculation data is adjusted to reduce the error. This process is repeatedly performed for each of the multiple learning data to generate a trained model with adjusted calculation data. In other words, in this embodiment, the trained model may be generated by so-called supervised learning using learning data and training data. The sleep estimation device 51, described later, can determine the sleep stage of a subject by using the trained model generated in this way.

[0055] The training data is example data used to generate a trained model. The training data may be frequency spectra generated from blood flow waveform data. In this embodiment, wavelet images are used. The wavelet images may be those with enhanced target intensity or those without. The training data may exhibit different behaviors during wakefulness and sleep, or change behavior depending on the depth of sleep. The training data may consist of various types of data (e.g., frequency spectra of blood flow waveform data showing different waveforms).

[0056] Training data is data to which correct labels are associated with training data. For example, training data may be used in which the sleep stage of a person from whom blood flow waveform data was obtained is associated as the correct label for a frequency spectrum as training data. As mentioned above, the sleep stage of a subject may be identified based on electroencephalogram (EEG) data detected by an EEG machine. Codes indicating each sleep stage may be used as the correct label. Alternatively, a code indicating the correct answer for a specific sleep stage (e.g., stage 2 or 3) may be used as the correct label, and a code indicating the incorrect answer for other sleep stages may be used. In this embodiment, as an example of training data, data associated with wavelet images known to correspond to stage 2 or 3 (wavelet images identified as stage 2 or 3 by EEG data) as the correct label may be used.

[0057] The computational data includes data related to the calculations used to generate the trained model, such as the calculation formula, the variables in the formula (e.g., bias and weights), and the activation function. The bias and weights define the strength of the connections between each unit. By adjusting the bias and weights, the accuracy of the trained model can be improved. For example, backpropagation and gradient descent may be used as methods for adjusting the computational data.

[0058] [Embodiment 1] Below is an example of a sleep estimation system 1 capable of determining the sleep stage of a subject, constructed based on the above principle. The sleep estimation system 1 of this embodiment may be a system capable of determining the sleep stage of a subject using the above-trained model.

[0059] <Sleep Stage Estimation System> Figure 1 is a block diagram showing a schematic configuration example of the sleep estimation system 1 of Embodiment 1. As shown in Figure 1, the sleep estimation system 1 includes an accelerometer 2, a blood flow meter 3, and a portable terminal 5. In the portable terminal 5, as part of the function of the control unit that comprehensively controls each component of the portable terminal 5, a sleep estimation device 51 is constructed that determines the sleep stage of a subject by executing an application capable of determining the sleep stage of the subject.

[0060] <Accelerometer> The accelerometer 2 is a sensor capable of detecting acceleration generated by the movement of the subject. The accelerometer 2 may transmit the detected acceleration as acceleration data to the sleep estimation device 51 via wireless or wired communication. The accelerometer 2 is attached to a part of the subject's body, such as the head or finger. As the accelerometer 2, known sensors such as frequency-varying, piezoelectric, piezoresistive, or capacitive sensors may be used.

[0061] <Blood flow meter> The blood flow meter 3 may be the blood flow meter described in the principle above. The blood flow meter 3 may be, for example, a laser Doppler blood flow meter. In this embodiment, the blood flow meter 3 may transmit raw waveform data W1 as blood flow waveform data to the sleep estimation device 51. The blood flow meter 3 may transmit processed waveform data W2 to the sleep estimation device 51 instead of raw waveform data W1. The blood flow meter 3 may not generate processed waveform data W2, and the sleep estimation device 51 may generate processed waveform data W2. The blood flow meter 3 is attached to a part of the subject's body, such as the ear, finger, wrist, arm, forehead, nose, or neck.

[0062] <Mobile devices> The mobile terminal 5 is any terminal capable of data communication with at least the accelerometer 2 and the blood flow meter 3. The mobile terminal 5 may be, for example, a smartphone or a tablet. The mobile terminal 5 is equipped with a sleep estimation device 51, as well as a memory unit 52 and a notification unit 53.

[0063] The memory unit 52 can store programs and data used by the control unit (particularly the sleep estimation device 51). For example, the memory unit 52 can store the trained model generated as described above, and thresholds for determining whether the subject is at rest or not.

[0064] The notification unit 53 can notify the surrounding area of ​​the mobile terminal 5 (e.g., the subject) of various information. In this embodiment, the notification unit 53 can notify various information in accordance with the notification instructions from the sleep estimation device 51. The notification unit 53 may be at least one of a sound output device that outputs sound, a vibration device that vibrates the mobile terminal 5, and a display device that displays images.

[0065] (Sleep estimation device) The sleep estimation device 51 can determine the sleep stage of a subject to whom the accelerometer 2 and blood flow meter 3 are attached. The sleep estimation device 51 may include a second acquisition unit 11, a second determination unit 12, a first acquisition unit 13 (acquisition unit), a generation unit 14, a first determination unit 15 (determination unit), and a notification unit 16.

[0066] The second acquisition unit 11 can acquire acceleration data from the accelerometer 2. The second determination unit 12 can determine whether the subject is stationary or not based on the acceleration data acquired by the second acquisition unit 11. For example, the second determination unit 12 may determine that the subject is stationary if the acceleration indicated by the acceleration data is less than a threshold stored in the storage unit 52. The second determination unit 12 may transmit the determination result data to the generation unit 14.

[0067] The first acquisition unit 13 can acquire blood flow waveform data (raw waveform data W1) from the blood flow meter 3. The generation unit 14 can generate a frequency spectrum of the blood flow waveform data by performing frequency analysis processing on the blood flow waveform data acquired by the first acquisition unit 13. The first determination unit 15 can determine the sleep stage of the subject based on the frequency spectrum generated by the generation unit 14. In other words, the first determination unit 15 can determine the transition of the depth of the subject's sleep based on the frequency spectrum.

[0068] In this embodiment, the generation unit 14 can generate a wavelet image by performing a wavelet transform process as a frequency analysis process, in which the target intensity is relatively emphasized compared to other frequency bands. This wavelet image is processed data that shows the result of the time-frequency analysis process of the blood flow waveform data. The first determination unit 15 can determine the sleep stage of the subject based on the wavelet image generated by the generation unit 14.

[0069] In this embodiment, if the first determination unit 15 determines that the characteristic waveform Sh is present in the 0.2 to 0.3 Hz frequency band of the frequency spectrum, it can determine that the subject's sleep stage is stage 2 or 3. In this case, the first determination unit 15 may determine that the subject's sleep stage has transitioned from stage 1 to stage 2 or 3. On the other hand, if the first determination unit 15 determines that the characteristic waveform Sh is not present in the 0.2 to 0.3 Hz frequency band, it can determine that the subject's sleep stage is a sleep stage other than stage 2 or 3. If the first determination unit 15 determines that the subject is in stage 2 or 3 sleep stage, and then determines that the characteristic waveform Sh is not present in the 0.2 to 0.3 Hz frequency band, it may determine that the subject's sleep stage has transitioned from stage 2 or 3 to stage 1. The determination of whether or not the characteristic waveform Sh is present can also be performed, for example, by determining whether or not the first intensity is greater than or equal to a predetermined value than the second intensity.

[0070] In this embodiment, the first determination unit 15 may perform the determination of the subject's sleep stage by using a trained model. In this case, the first determination unit 15 can output the determination result of the subject's sleep stage from the output layer of the trained model by providing the wavelet image generated by the generation unit 14 as input data to the input layer of the trained model.

[0071] As described above, the trained model is generated using data linked as ground truth labels to wavelet images known to correspond to stage 2 or 3 as an example of training data. Therefore, the first determination unit 15 can determine whether the subject's sleep stage is stage 2 or 3, or whether they have transitioned from stage 1 to stage 2 or 3, by providing the wavelet images generated by the generation unit 14 to the trained model. In particular, the first determination unit 15 may be able to accurately determine whether the subject's sleep stage is stage 2 or 3 when the trained model is provided with a frequency spectrum that includes the characteristic waveform Sh in the 0.2~0.3Hz frequency band.

[0072] In this case, when estimating sleep stages using blood flow waveform data, the sleep estimation device may not be able to determine whether the subject is asleep or not based solely on the blood flow waveform data. For example, in the case of a subject whose blood flow is stable even when awake, there may not be a significant difference between the blood flow waveform data when awake and the blood flow waveform data when asleep. In this case, the sleep estimation device may not be able to determine whether the subject is asleep or not. In this embodiment, when the second determination unit 12 determines that the subject is at rest, the generation unit 14 performs frequency analysis processing on the blood flow waveform data, thereby enabling the determination processing by the first determination unit 15 to be performed when there is a high probability that the subject has fallen asleep.

[0073] The notification unit 16 can perform notification processing based on the determination result of the first determination unit 15. The notification unit 16 may also transmit a notification instruction in accordance with the notification processing to the notification unit 53. This allows the notification unit 53 to broadcast information in accordance with the notification processing to those around the mobile terminal 5. The notification unit 16 may include a first notification unit 161 and a second notification unit 162.

[0074] The first notification unit 161 can execute the first notification process as the notification process described above, after a predetermined time has elapsed since the first determination unit 15 detected a transition from stage 1 to stage 2 or 3. The first notification process is a notification process associated with the detection of the transition, and may be, for example, an alarm process to encourage the subject to wake up, or a process to notify that the transition has been detected. The predetermined time may be set appropriately, for example, through experimentation, depending on the purpose of the notification. In this embodiment, the predetermined time may be set to, for example, a time when it is estimated that the subject is likely to wake up easily, when time is measured from the moment the transition from stage 1 to stage 2 or 3 occurs. As a result of the first notification process, for example, the subject can wake up at a good wake-up timing after falling asleep. The first notification unit 161 may also execute the first notification process when the first determination unit 15 detects a transition from stage 1 to stage 2 or 3.

[0075] The second notification unit 162 can execute a second notification process as the notification process described above, after a predetermined time has elapsed since the first determination unit 15 detected a transition to stage 2 or stage 3-1. The second notification process is a notification process associated with the detection of the transition, and may be, for example, an alarm process to encourage the subject to wake up, or a process to notify that the transition has been detected. The predetermined time may be set appropriately, for example, through experimentation, depending on the purpose of the notification. In this embodiment, the predetermined time may be set to a time that is estimated to be when the subject is likely to wake up easily, for example, when time is measured from the moment of transition from stage 2 or 3 to stage 1. Through the second notification process, for example, the subject can wake up at a good wake-up timing after falling asleep. The second notification unit 162 may also execute the second notification process when the first determination unit 15 detects a transition from stage 2 or 3 to stage 1.

[0076] [Processing flow] Figure 7 is a flowchart showing an example of the processing flow (sleep estimation method) by the sleep estimation device 51. When the sleep stage of a subject is determined by the sleep estimation device 51, the blood flow meter 3 may be attached to the subject and then start detecting blood flow waveform data.

[0077] As shown in Figure 7, in the sleep estimation device 51, the first acquisition unit 13 can acquire blood flow waveform data from the blood flow meter 3 (S1: first acquisition step, acquisition step). The generation unit 14 can generate the frequency spectrum of the blood flow waveform data by performing frequency analysis processing on the blood flow waveform data. In this embodiment, the generation unit 14 can generate a wavelet image with the target intensity emphasized by performing wavelet transform processing on the blood flow waveform data (S2: generation step). The first determination unit 15 can determine the sleep stage of the subject based on the frequency spectrum. In this embodiment, the first determination unit 15 can determine the sleep stage of the subject by inputting the wavelet image with the target intensity emphasized into a trained model (S3) (S4: first determination step, determination step).

[0078] The first determination unit 15 can determine whether or not the sleep has transitioned from stage 1 to stage 2 or 3 based on the sleep stage determination result (S5). If the first determination unit 15 determines that the sleep has transitioned from stage 1 to stage 2 or 3 (YES in S5), the first notification unit 161 can determine whether or not a predetermined time has elapsed since that determination (S6). If the first notification unit 161 determines that a predetermined time has elapsed (YES in S6), it can perform an alarm process as a first notification process, for example, by having the notification unit 53 announce an alarm sound (S7). The notification unit 53 can announce an alarm sound upon receiving a notification instruction from the first notification unit 161.

[0079] If the result in S5 is NO, the process may return to S1. If the result in S6 is NO, the process in S6 may be repeated. Also, if the first determination unit 15 determines in S5 that a transition has occurred from stage 2 or 3 to stage 1, the second notification unit 162 may execute a second notification process, for example, after a predetermined time has elapsed.

[0080] [Challenges in the prior art and the effects of the sleep estimation device described herein] Patent Document 1 discloses a method for detecting non-REM sleep, which includes the following steps 1 to 4. Step 1: Generate time-series data of the subject's heartbeat intervals. Step 2: Set a window of a predetermined time length that moves along the time axis of the time series data, and perform spectral analysis on the time series data within the window containing each of the multiple decision points on the time axis. Step 3: Calculate the power concentration of the high-frequency components of heart rate variability from the spectrum of each window. • A step to determine whether or not it is non-REM sleep based on the calculated level of concentration.

[0081] In the above method, a pulse wave meter or electrocardiograph is used as the device for detecting the time-series data in step 1.

[0082] However, handling an electroencephalograph (EEG) and acquiring EEG data requires highly specialized knowledge. Furthermore, attaching the EEG device is complicated. Therefore, it is difficult for subjects to easily acquire EEG data and to easily understand their own sleep stages.

[0083] Furthermore, as mentioned above, the inventors found that if a significant intensity change is observed in the 0.2-0.3 Hz frequency band in the frequency spectrum of blood flow waveform data, there is a high probability that the subject is in stage 2 or 3 of sleep.

[0084] The sleep estimation device 51 of this disclosure can determine a subject's sleep stage using blood flow waveform data. Handling the blood flow meter and acquiring blood flow waveform data does not require the same level of expertise as an electroencephalograph (EEG). Furthermore, attaching the blood flow meter is easier than with an EEG. In short, the sleep estimation device 51 can acquire blood flow waveform data from a subject relatively easily, and therefore can determine the subject's sleep stage relatively easily. Moreover, the sleep estimation device 51 allows the subject to easily understand their own sleep stage.

[0085] Furthermore, the sleep estimation device 51 of this disclosure can determine that the subject's sleep stage is stage 2 or 3 when it obtains a frequency spectrum having the significant intensity change described above that is not observed in electrocardiogram waveform data. Therefore, the sleep estimation device 51 can increase the likelihood of accurately estimating that the subject's sleep stage is stage 2 or 3.

[0086] [Embodiment 2] Other embodiments of this disclosure are described below. For the sake of clarity, components having the same function as those described in the above embodiments are denoted by the same reference numerals, and their descriptions are not repeated. Figure 8 is a block diagram showing a schematic configuration example of the sleep estimation system 1A of Embodiment 2.

[0087] In the sleep estimation system 1 of Embodiment 1, the mobile terminal 5 can acquire blood flow waveform data from a blood flow meter 3 attached to the subject, for example, via wireless communication. Then, the sleep estimation device 51 built on the mobile terminal 5 can determine the subject's sleep stage based on the blood flow waveform data.

[0088] On the other hand, as shown in Figure 8, in the sleep estimation system 1A of Embodiment 2, the wearable device 20 may be equipped with a blood flow meter 3. In addition, the wearable device 20 may have a sleep estimation device 51 as part of the function of a control unit that comprehensively controls each component of the wearable device 20. In other words, in the sleep estimation system 1A, the sleep estimation device 51 may be implemented in the wearable device 20 together with the blood flow meter 3. Therefore, the acquisition of blood flow waveform data and the determination of sleep stages based on the blood flow waveform data can be performed with a single device. Furthermore, various devices or parts required for wireless or wired communication between the two devices become unnecessary. The wearable device 20 only needs to be attached to the subject in the same position as where the blood flow meter 3 is attached.

[0089] [Variation] The inventions described in this disclosure have been explained above based on the drawings and embodiments. However, the inventions described in this disclosure are not limited to the embodiments described above. That is, the inventions described in this disclosure can be modified in various ways within the scope shown in this disclosure, and embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the inventions described in this disclosure. In other words, it should be noted that it is easy for those skilled in the art to make various modifications or alterations based on this disclosure. Furthermore, it should be noted that these modifications or alterations are included in the scope of this disclosure.

[0090] (Variation of frequency analysis processing) For example, the generation unit 14 only needs to be able to generate a wavelet transform image by performing a wavelet transform process on the blood flow waveform data, and it is not necessarily required to perform the wavelet transform process in a way that emphasizes the target intensity.

[0091] Furthermore, the generation unit 14 may perform time-frequency analysis processing other than wavelet transform processing on the blood flow waveform data. The generation unit 14 may also perform, for example, short-time Fourier transform processing on the blood flow waveform data. Short-time Fourier transform processing involves performing Fourier transform processing on each of the multiple waveform data extracted along the time axis using a window function. In this case, the generation unit 14 may generate an image similar to a wavelet image (intensity change data) as processed data by performing short-time Fourier transform processing in which the target intensity is relatively emphasized compared to other frequency bands.

[0092] Furthermore, the generation unit 14 may perform processing other than time-frequency analysis as part of the frequency analysis processing. For example, the generation unit 14 may perform a Fourier transform. When the generation unit 14 performs a Fourier transform, it can generate a frequency spectrum FW1 or FW2 as shown in Figure 3. Alternatively, the generation unit 14 may generate a frequency spectrum (waveform) as shown in Figure 3 by performing a short-time Fourier transform.

[0093] Thus, the generation unit 14 can generate various frequency spectra. Therefore, the first determination unit 15 can determine the sleep stage of the subject by inputting the various frequency spectra into the trained model. However, the trained model is generated using the same type of frequency spectra as those generated by the generation unit 14 as training data and target data.

[0094] (Modified version of the determination process in the first determination unit) The trained model does not need to be stored in the memory unit 52 of the mobile terminal 5 or wearable device 20. In other words, the sleep estimation system 1 or 1A does not need to use the trained model to determine the sleep stage of the subject.

[0095] For example, the first determination unit 15 may determine that the subject's sleep stage is stage 2 or 3 if it determines that the first intensity in the frequency spectrum generated by the generation unit 14 is greater than the second intensity by a predetermined value or more. The memory unit 52 may store predetermined values ​​instead of the learned model.

[0096] Furthermore, when the first determination unit 15 determines the sleep stage of a subject using the frequency spectrum FW1, it may determine that the subject's sleep stage is stage 2 or 3 if it can extract a characteristic waveform Sh in the 0.2 to 0.3 Hz frequency band. In this case, the memory unit 52 may store a reference waveform capable of extracting a characteristic waveform Sh in the frequency spectrum FW1, instead of a trained model. The first determination unit 15 may determine that it has been able to extract a characteristic waveform Sh in the 0.2 to 0.3 Hz frequency band if it determines that a waveform matching the reference waveform exists in the 0.2 to 0.3 Hz frequency band. The first determination unit 15 may determine that a waveform matching the reference waveform exists in the 0.2 to 0.3 Hz frequency band if the degree of agreement between the waveform included in the 0.2 to 0.3 Hz frequency band and the reference waveform is, for example, above a threshold set experimentally. Furthermore, the first determination unit 15 may determine whether or not the waveform Sh characteristic of the 0.2 to 0.3 Hz frequency band is included based on other indicators (e.g., the degree of change in the slope of the waveform). In addition, the same determination may be made for the frequency spectrum FW2 as for the frequency spectrum FW1.

[0097] (A modified example of sleep estimation system 1 or 1A) The accelerometer 2 detects the acceleration generated by the subject's movement, and the sleep estimation device 51 does not need to determine the subject's sleep state based on this acceleration. In this case, the sleep estimation system 1 or 1A does not need to be equipped with the accelerometer 2, and the sleep estimation device 51 does not need to be equipped with the second acquisition unit 11 and the second determination unit 12.

[0098] [Examples of implementation using software] The control block of the sleep estimation device 51 may be implemented by logic circuits (hardware) formed on an integrated circuit (IC chip) or the like, or by software.

[0099] In the latter case, the sleep estimation device 51 includes a computer that executes instructions for a program, which is software that realizes each function. This computer includes, for example, at least one processor (control device) and at least one computer-readable recording medium that stores the program. The object of this disclosure is achieved when the processor reads the program from the recording medium and executes it in the computer. For example, a CPU (Central Processing Unit) can be used as the processor. As the recording medium, a "tangible medium that is not temporary," such as ROM (Read Only Memory), can be used, as well as tape, disk, card, semiconductor memory, programmable logic circuit, etc. It may also further include RAM (Random Access Memory) for deploying the program. Furthermore, the program may be supplied to the computer via any transmission medium capable of transmitting the program (such as a communication network or broadcast wave). One aspect of this disclosure can also be realized in the form of a data signal embedded in a carrier wave, in which the program is embodied by electronic transmission.

[0100] [Examples of wording used in this disclosure] Furthermore, one aspect of this disclosure may be expressed as follows:

[0101] A sleep estimation device according to one aspect of the present disclosure includes a first acquisition unit that acquires blood flow data indicating the blood flow of a subject, a generation unit that generates a frequency spectrum of the blood flow data by performing frequency analysis processing on the blood flow data, and a first determination unit that determines the sleep stage of the subject based on the frequency spectrum.

[0102] Furthermore, a sleep estimation device according to one aspect of the present disclosure includes: an acquisition unit that acquires blood flow data indicating the blood flow of a subject; a generation unit that generates processed data indicating the result of time-frequency analysis of the blood flow data by performing a wavelet transform or short-time Fourier transform on the blood flow data in which the intensity in a predetermined frequency band is relatively emphasized compared to other frequency bands; and a determination unit that determines the sleep stage of the subject based on the processed data.

[0103] Furthermore, a sleep estimation method according to one aspect of the present disclosure includes a first acquisition step of acquiring blood flow data indicating the blood flow of a subject, a generation step of generating a frequency spectrum of the blood flow data by performing frequency analysis processing on the blood flow data, and a first determination step of determining the sleep stage of the subject based on the frequency spectrum.

[0104] Furthermore, a sleep estimation method according to one aspect of the present disclosure includes: an acquisition step of acquiring blood flow data indicating the blood flow of a subject; a generation step of generating processed data indicating the result of a time-frequency analysis of the blood flow data by performing a wavelet transform or short-time Fourier transform on the blood flow data in which the intensity in a predetermined frequency band is relatively emphasized compared to other frequency bands; and a determination step of determining the sleep stage of the subject based on the processed data.

[0105] Furthermore, a sleep estimation system according to one aspect of the present disclosure includes a first acquisition unit that acquires blood flow data indicating the blood flow of a subject, a generation unit that generates a frequency spectrum of the blood flow data by performing frequency analysis processing on the blood flow data, and a first determination unit that determines the sleep stage of the subject based on the frequency spectrum. When the sleep stages in non-REM sleep are designated as stages 1, 2, and 3 in order from the lightest sleep stage, the first determination unit determines that the sleep stage of the subject has transitioned from stage 1 to stage 2 or 3 if the generated frequency spectrum has a first intensity, which is the maximum intensity in a first range that is part of a region including a frequency band of 0.2 Hz or more and 0.3 Hz or less, greater than the second intensity in a second range that is part of the region other than the first range.

[0106] Furthermore, a sleep estimation system according to one aspect of the present disclosure may include a first notification unit that executes a first notification process when the first determination unit determines that a transition has occurred from stage 1 to stage 2 or 3, or after a predetermined time has elapsed since the determination that a transition has occurred from stage 1 to stage 2 or 3.

[0107] Furthermore, in a sleep estimation system according to one aspect of the present disclosure, when the first determination unit determines that the subject's sleep stage has transitioned from stage 1 to stage 2 or 3, the frequency spectrum generated by the generation unit may include an upwardly convex broad waveform centered on a first intensity in the frequency band.

[0108] Furthermore, a sleep estimation system according to one aspect of the present disclosure includes: a first acquisition unit that acquires blood flow data indicating the blood flow of a subject; a generation unit that generates a frequency spectrum of the blood flow data by performing frequency analysis processing on the blood flow data; and a first determination unit that determines the sleep stage of the subject based on the frequency spectrum. When the sleep stages in non-REM sleep are designated as stages 1, 2, and 3 in order from the lightest sleep stage, the generation unit generates intensity change data indicating the temporal change in intensity in each frequency band within a predetermined time period as the frequency spectrum by performing time-frequency analysis processing as the frequency analysis processing. The first determination unit determines that the sleep stage of the subject has transitioned from stage 1 to stage 2 or 3 when it provides the intensity change data, which includes a frequency spectrum having a predetermined shape in a frequency band of 0.2 Hz or higher and 0.3 Hz or lower, generated by the generation unit, to a trained model that has been trained using the intensity change data as training data, which includes a frequency spectrum having a predetermined shape in a frequency band of 0.2 Hz or higher and 0.3 Hz or lower, identified as corresponding to stage 2 or 3 based on biological information different from the blood flow data, to a trained model.

[0109] Furthermore, a sleep estimation system according to one aspect of the present disclosure includes: a first acquisition unit that acquires blood flow data indicating the blood flow of a subject; a generation unit that generates a frequency spectrum of the blood flow data by performing frequency analysis processing on the blood flow data; a first determination unit that determines that the sleep stage of the subject has transitioned from stage 1 to stage 2 or 3 when the sleep stages in non-REM sleep are designated as stages 1, 2, and 3 in order from the lightest sleep stage; a second acquisition unit that acquires acceleration data indicating the acceleration caused by the movement of the subject; and a second determination unit that determines whether the subject is stationary based on the acceleration data, wherein the generation unit executes the frequency analysis processing when the second determination unit determines that the subject is stationary.

[0110] Furthermore, a sleep estimation system according to one aspect of the present disclosure includes: an acquisition unit that acquires blood flow data indicating the blood flow of a subject; a generation unit that generates processed data indicating the results of time-frequency analysis of the blood flow data by performing a process on the blood flow data that relatively emphasizes the intensity in a frequency band of 0.2 Hz or higher and 0.3 Hz or lower compared to other frequency bands; and a determination unit that determines the sleep stage of the subject based on the processed data. The generation unit generates intensity change data indicating the temporal change in intensity in each frequency band within a predetermined time as the processed data, and determines non-REM sleep. When the sleep stages are designated as stages 1, 2, and 3 in order from the lightest sleep stage, the determination unit determines that the subject's sleep stage has transitioned from stage 1 to stage 2 or 3 when it provides the intensity change data having the predetermined shape, generated by the generation unit, to a trained model that has been trained using the intensity change data having the predetermined shape, which is identified as corresponding to stage 2 or 3 based on biological information different from the blood flow data and includes a frequency spectrum having a predetermined shape in a frequency band of 0.2 Hz or higher and 0.3 Hz or lower, as training data.

[0111] Furthermore, in a sleep estimation system according to one aspect of the present disclosure, the generation unit may generate the processed data by performing a wavelet transform or a short-time Fourier transform as the enhancement process.

[0112] Furthermore, in a sleep estimation system according to one aspect of this disclosure, the biological information may be electroencephalogram data detected by an electroencephalograph.

[0113] Furthermore, in a sleep estimation system according to one aspect of the present disclosure, the predetermined shape may be an upwardly convex broad waveform.

[0114] Furthermore, a sleep estimation system according to one aspect of the present disclosure may include a blood flow meter that detects the blood flow data by receiving scattered light generated by irradiating the blood vessels of the subject with light.

[0115] Furthermore, a sleep estimation system according to one aspect of the present disclosure may include a wearable device equipped with a blood flow meter that detects the blood flow data by receiving scattered light generated by irradiating the blood vessels of the subject with light.

[0116] Furthermore, a sleep estimation method according to one aspect of the present disclosure includes a first acquisition step of acquiring blood flow data indicating the blood flow of a subject, a generation step of generating a frequency spectrum of the blood flow data by performing frequency analysis processing on the blood flow data, and a first determination step of determining the sleep stage of the subject based on the frequency spectrum, wherein when the sleep stages in non-REM sleep are designated as stages 1, 2, and 3 in order from the lightest sleep stage, the first determination step determines that the sleep stage of the subject has transitioned from stage 1 to stage 2 or 3 if the generated frequency spectrum is such that the first intensity, which is the maximum intensity in a first range that is part of a region including a frequency band of 0.2 Hz or more and 0.3 Hz or less, is greater than the second intensity in a second range that is part of the region other than the first range.

[0117] Furthermore, a sleep estimation method according to one aspect of the present disclosure includes: an acquisition step of acquiring blood flow data indicating the blood flow of a subject; a generation step of generating processed data indicating the result of time-frequency analysis of the blood flow data by performing a process on the blood flow data to relatively emphasize the intensity in a frequency band of 0.2 Hz or higher and 0.3 Hz or lower compared to other frequency bands; and a determination step of determining the sleep stage of the subject based on the processed data. In the generation step, intensity change data indicating the temporal change in intensity in each frequency band within a predetermined time is generated as the processed data. When the sleep stages in non-REM sleep are designated as stages 1, 2, and 3 in order from the lightest sleep stage, the determination step determines that the sleep stage of the subject has transitioned from stage 1 to stage 2 or 3 when the intensity change data having the predetermined shape, generated in the generation step, is provided to a trained model that has been trained using the intensity change data having the predetermined shape, which is identified as corresponding to stage 2 or 3 based on biological information different from the blood flow data, as training data.

[0118] Furthermore, a sleep estimation system according to one aspect of the present disclosure includes a determination unit that determines the sleep stage of a subject based on the frequency spectrum of the subject's blood flow data, and when the sleep stages in non-REM sleep are designated as stages 1, 2, and 3 in order from the lightest sleep stage, the determination unit determines that the subject's sleep stage has transitioned from stage 1 to stage 2 or 3 if the frequency spectrum includes a predetermined shape in a region that includes a frequency band of 0.2 Hz or higher and 0.3 Hz or lower, and the predetermined shape included in the region is an upwardly convex waveform shape that includes a first intensity which is the maximum intensity in the region.

[0119] Furthermore, in a sleep estimation system according to one aspect of the present disclosure, the determination unit may determine that the predetermined shape is included if, in the frequency spectrum, the first intensity in a first range of a part of the frequency band is greater than the second intensity in a second range other than the first range.

[0120] Furthermore, a sleep estimation system according to one aspect of the present disclosure may include a first notification unit that executes a first notification process when the determination unit determines that a transition has occurred from stage 1 to stage 2 or 3, or after a predetermined time has elapsed since the determination that a transition has occurred from stage 1 to stage 2 or 3.

[0121] Furthermore, a sleep estimation system according to one aspect of the present disclosure may include a blood flow meter that detects the blood flow data by receiving scattered light generated by irradiating the blood vessels of the subject with light.

[0122] Furthermore, a sleep estimation system according to one aspect of the present disclosure may include a wearable device equipped with a blood flow meter that detects the blood flow data by receiving scattered light generated by irradiating the blood vessels of the subject with light.

[0123] Furthermore, a sleep estimation system according to one aspect of the present disclosure includes a determination unit that determines the sleep stage of a subject based on the frequency spectrum of the subject's blood flow data. When the sleep stages in non-REM sleep are designated as stages 1, 2, and 3 in order from the lightest sleep stage, the determination unit determines whether the subject's sleep stage has transitioned from stage 1 to stage 2 or 3 when the frequency spectrum of the blood flow data is provided to a trained model that has been trained using a frequency spectrum having a predetermined shape in a region including a frequency band of 0.2 Hz or higher and 0.3 Hz or lower, which has been identified as corresponding to stage 2 or 3 based on biological information different from the blood flow data. The predetermined shape included in the region is an upwardly convex waveform shape including a first intensity which is the maximum intensity in the region, and the biological information is electroencephalogram data detected by an electroencephalograph.

[0124] Furthermore, a sleep estimation system according to one aspect of the present disclosure may include a first notification unit that executes a first notification process when the determination unit determines that a transition has occurred from stage 1 to stage 2 or 3, or after a predetermined time has elapsed since the determination that a transition has occurred from stage 1 to stage 2 or 3.

[0125] Furthermore, a sleep estimation system according to one aspect of the present disclosure may include a blood flow meter that detects the blood flow data by receiving scattered light generated by irradiating the blood vessels of the subject with light.

[0126] Furthermore, a sleep estimation system according to one aspect of the present disclosure may include a wearable device equipped with a blood flow meter that detects the blood flow data by receiving scattered light generated by irradiating the blood vessels of the subject with light.

[0127] Furthermore, a sleep estimation method according to one aspect of the present disclosure is a sleep estimation method performed in a sleep estimation system, wherein a determination unit in the sleep estimation system includes a determination step of determining the sleep stage of a subject based on the frequency spectrum of the subject's blood flow data, and when the sleep stages in non-REM sleep are designated as stages 1, 2, and 3 in order from the lightest sleep stage, the determination step determines that the sleep stage of the subject has transitioned from stage 1 to stage 2 or 3 if at least the frequency spectrum includes a predetermined shape in a region that includes a frequency band of 0.2 Hz or more and 0.3 Hz or less, and the predetermined shape included in the region is an upwardly convex waveform shape that includes a first intensity which is the maximum intensity in the region.

[0128] Furthermore, a sleep estimation method according to one aspect of the present disclosure is a sleep estimation method performed in a sleep estimation system, wherein a determination unit in the sleep estimation system includes a determination step of determining the sleep stage of a subject based on the frequency spectrum of the subject's blood flow data, and when the sleep stages in non-REM sleep are designated as stages 1, 2, and 3 in order from the lightest sleep stage, the determination step determines whether the subject's sleep stage has transitioned from stage 1 to stage 2 or 3 when the frequency spectrum of the blood flow data is given to a trained model that has been trained using a frequency spectrum having a predetermined shape in a region including a frequency band of 0.2 Hz or more and 0.3 Hz or less, which has been identified as corresponding to stage 2 or 3 based on biological information different from the blood flow data, and the predetermined shape included in the region is an upwardly convex waveform shape including a first intensity which is the maximum intensity in the region, and the biological information may be electroencephalogram data detected by an electroencephalograph. [Explanation of Symbols]

[0129] 1. 1A Sleep Estimation System 3 Blood flow meter 11 Second acquisition part 12 Second Judgment Section 13 1st Acquisition Department (Acquisition Department) 14 Generation part 15 1st judgment section (judgment section) 20 Wearable Devices 51 Sleep estimation device 161 First Notification Department 162 2nd Notification Department

Claims

1. A determination unit that determines the sleep stage of the subject based on the frequency spectrum of the subject's blood flow data, The system includes a notification unit that performs notification processing based on the determination result of the determination unit, When the sleep stages in non-REM sleep are designated as stages 1, 2, and 3, in order from the lightest sleep stage, The determination unit determines that the subject's sleep stage has transitioned from stage 1 to stage 2 or 3 if the frequency spectrum includes a predetermined shape in a region that includes a frequency band of 0.2 Hz or higher and 0.3 Hz or lower. The notification unit, when the determination unit determines that the subject's sleep stage has transitioned from stage 1 to stage 2 or 3, or after a predetermined time has elapsed since the determination, will notify the subject to wake up. The predetermined shape included in the region is an upwardly convex waveform shape that includes a first intensity which is the maximum intensity in the region. The predetermined time is the time when the subject is estimated to be most likely to wake up, measured from the point in time when the subject's sleep stage transitions from stage 1 to stage 2 or 3. Sleep estimation system.

2. The sleep estimation system according to claim 1, wherein the determination unit determines that the predetermined shape is included when the first intensity in a first range of a part of the frequency band in the frequency spectrum is greater than the second intensity in a second range other than the first range.

3. The sleep estimation system according to claim 1 or 2, further comprising a blood flow meter that detects the blood flow data by receiving scattered light generated by irradiating the blood vessels of the subject with light.

4. The sleep estimation system according to claim 1 or 2, further comprising a wearable device equipped with a blood flow meter that detects the blood flow data by receiving scattered light generated by irradiating the blood vessels of the subject with light.

5. The system comprises a determination unit that determines the sleep stage of a subject based on the frequency spectrum of the subject's blood flow data, and a notification unit that performs notification processing based on the determination result of the determination unit. The determination unit determines whether the frequency spectrum of the blood flow data corresponds to the correct answer when it is given to a trained model that has been trained using training data in which correct labels based on biological information different from the blood flow data are associated with frequency spectra having a predetermined shape in a region including a frequency band of 0.2 Hz or higher and 0.3 Hz or lower. The notification unit, when the determination unit determines that the answer is correct, or after a predetermined time has elapsed since the determination, will notify the subject to wake up. The predetermined shape included in the region is an upwardly convex waveform shape that includes a first intensity which is the maximum intensity in the region. The aforementioned biological information is electroencephalogram (EEG) data detected by an electroencephalograph. The correct label is Stage 2 or 3, in which the sleep stages in non-REM sleep are designated as Stage 1, Stage 2, and Stage 3 in order from the lightest sleep stage. The predetermined time is the time when the subject is estimated to be most likely to wake up, measured from the point in time when the subject's sleep stage transitions from stage 1 to stage 2 or 3. Sleep estimation system.

6. The sleep estimation system according to claim 5, further comprising a blood flow meter that detects the blood flow data by receiving scattered light generated by irradiating the blood vessels of the subject with light.

7. The sleep estimation system according to claim 5, comprising a wearable device equipped with a blood flow meter that detects the blood flow data by receiving scattered light generated by irradiating the blood vessels of the subject with light.

8. A sleep estimation method performed in a sleep estimation system, The determination unit of the sleep estimation system includes a determination step of determining the sleep stage of the subject based on the frequency spectrum of the subject's blood flow data, The notification unit of the sleep estimation system includes a notification step of performing a notification process based on the determination result of the determination unit, When the sleep stages in non-REM sleep are designated as stages 1, 2, and 3, in order from the lightest sleep stage, In the determination step, if the frequency spectrum includes a predetermined shape in a region that includes a frequency band of 0.2 Hz or higher and 0.3 Hz or lower, it is determined that the subject's sleep stage has transitioned from stage 1 to stage 2 or 3. In the notification step, when the determination step determines that the subject's sleep stage has transitioned from stage 1 to stage 2 or 3, or after a predetermined time has elapsed since the determination, a notification is given to the subject to encourage awakening. The predetermined shape included in the region is an upwardly convex waveform shape that includes a first intensity which is the maximum intensity in the region. The predetermined time is the time when the subject is estimated to be most likely to wake up, measured from the point in time when the subject's sleep stage transitions from stage 1 to stage 2 or 3. Sleep estimation method.

9. A sleep estimation method performed in a sleep estimation system, The determination unit of the sleep estimation system includes a determination step of determining the sleep stage of the subject based on the frequency spectrum of the subject's blood flow data, The notification unit of the sleep estimation system includes a notification step that performs a notification process based on the determination result in the determination step, In the determination step, when the frequency spectrum of the blood flow data is given to a trained model that has been trained using training data in which correct labels based on biological information different from the blood flow data are associated with frequency spectra having a predetermined shape in a region including a frequency band of 0.2 Hz or higher and 0.3 Hz or lower, it is determined whether it corresponds to the correct answer. In the notification step, when it is determined in the determination step that the answer is correct, or after a predetermined time has elapsed since the determination, a notification is made to the subject to encourage awakening. The predetermined shape included in the region is an upwardly convex waveform shape that includes a first intensity which is the maximum intensity in the region. The aforementioned biological information is electroencephalogram (EEG) data detected by an electroencephalograph. The correct label is Stage 2 or 3, in which the sleep stages in non-REM sleep are designated as Stage 1, Stage 2, and Stage 3 in order from the lightest sleep stage. The predetermined time is the time when the subject is estimated to be most likely to wake up, measured from the point in time when the subject's sleep stage transitions from stage 1 to stage 2 or 3. Sleep estimation method.

10. A program for causing a computer to function as the sleep estimation system described in claim 1, wherein the program causes the computer to function as the determination unit and the notification unit.

11. A program for causing a computer to function as the sleep estimation system described in claim 5, wherein the program causes the computer to function as the determination unit and the notification unit.

Images