JP2019134953A - Biological information acquisition device and signal processing method - Google Patents

Abstract

To calculate a heartbeat interval from biological information including a ballistocardiogram or a biovibration signal.SOLUTION: A biological information acquisition device comprises: biological information acquiring means for acquiring biological information including at least a ballistocardiogram or a biovibration signal of an animal; and heartbeat interval calculating means for calculating variation in heartbeat interval from the biological information. The heartbeat interval calculating means subjects a signal derived from a ballistocardiogram waveform or a biovibration signal to a process of filtering with a frequency higher than a heartbeat frequency as a minimum frequency and taking an absolute value with respect to a signal having passed through the filter.SELECTED DRAWING: None

Description

  The present invention relates to an apparatus or method for acquiring information serving as an indicator of psychological stress or sleep state based on biological information acquired from an animal, or an animal sleep state measurement apparatus or method. Furthermore, the present invention relates to a method for extracting a heartbeat waveform or a respiratory waveform from biological information acquired from an animal.

  As the stress brought on by sophistication of society and the aging of the birthrate continue to increase, the number of patients with respiratory and circulatory diseases is increasing, and understanding and prevention of respiratory circulatory functions is becoming increasingly important. Conventionally, psychological stress assessment methods have mainly been measured at medical institutions such as subjective assessments through interviews, physiological indices such as blood pressure, heart rate, skin temperature, and sweating, and measurement of liquid stress markers such as catecholamine and cortisol. It was not an indicator that individuals could measure. A method of measuring an electrocardiogram to extract heart rate rhythm fluctuation, indirectly estimating autonomic nerve activity from its frequency analysis and using it as a stress index has also been proposed, but affected by the respiratory frequency and tidal volume, Since the index changes even in the absence of stress load, it is not a generalized evaluation method. In addition, there is a problem that individual differences such as age dependency and fitness influence cannot be quantitatively evaluated (Non-patent Document 1).

Non-Patent Document 2 discloses that the phase coherence of heartbeat interval fluctuation and the instantaneous phase difference of the respiratory pattern correlates with psychological stress, and the phase coherence may be evaluated as an index of psychological stress. . In Non-Patent Document 2, an electrocardiogram is obtained with electrodes attached to the chest, the respiratory pattern is measured with a breathing effort belt worn around the chest and abdomen, and heart rate interval fluctuations (respiratory sinus arrhythmia) are calculated from the electrocardiogram. Further, the instantaneous phase of the fluctuation of the heartbeat interval is calculated, the instantaneous phase of the breathing pattern is calculated, and the instantaneous phase difference Ψ (t) _j between the heartbeat and the breathing rhythm at time t is obtained from these, and the obtained instantaneous It is described that the phase coherence is calculated by the equation (1) using the phase difference. N in Expression (1) is the number of sampled data, and is obtained by averaging N data.

  The phase coherence takes a value of 0 to 1, approaching 1 as the fluctuation of the heartbeat interval and the instantaneous phase difference of the breathing pattern are closer to a fixed relationship, and approaching 0 as the instantaneous phase difference becomes random. When resting and relaxing, the phase coherence is close to 1, and when mental stress is applied, the phase coherence decreases. Therefore, it is possible to estimate psychological stress with a normalized index.

By the way, it is said that human sleep is roughly classified into non-REM sleep and REM sleep, and is repeated at a cycle of about 90 minutes during sleep. In order to objectively evaluate the quality and quantity of sleep (sleep cycle and sleep depth), polysomnography is required, and measurements are taken at medical institutions and borrowed from medical institutions. A plurality of sensors such as an electroencephalogram, an electrocardiogram, an electromyogram, and an electrooculogram are necessary, and handling is complicated. In addition, it is necessary to fix a plurality of sensors to various parts of the body, and movement is restricted by restraint by the sensors. There is a possibility that the sensor may come off due to unconscious movement during sleep, and there is a problem in that the signal cannot be detected and the state cannot be grasped if the sensor is removed. Furthermore, since the apparatus is large, it is not an index that can be evaluated by an individual, but has not yet reached a situation where an individual can grasp the quality of sleep. For this reason, there is a need for a biometric technique and apparatus that allows an individual to easily grasp the stress state and sleep quality.

  Also, when the elderly die lonely or die unfortunately at a nursing home, it is often impossible to judge whether the death is due to an accident or illness. In order to prevent such a situation, the appearance of a biological information measuring device that can be easily and non-invasively measured without restricting daily physical activity is awaited. Patent Document 1 discloses that a piezoelectric sensor that detects vibration generated from a human body is installed on the upper or lower portion of a bed pad or mattress, and a body vibration signal is detected using an unconstrained type piezoelectric sensor. Has been. Furthermore, after the body vibration signal detected by the piezoelectric sensor is amplified by the differential signal amplification amplifier, the heart beat (heart rate conversion: 30 to 240 times / minute, frequency band) that is the body vibration caused by the target person Conversion: 0.5 to 4 Hz) vibration, lung respiratory activity (respiration rate conversion: 60 times / min or less, frequency band conversion: 1 Hz or less), and sputum vibration caused by sputum are extracted using a separation filter function It is disclosed. In Patent Document 1, a person is present at a predetermined place when the state of pulsation vibration caused by heart beat, pulmonary respiratory vibration caused by pulmonary respiratory activity or sputum vibration exceeds a predetermined existence duration or longer. It is determined that the person is absent at the predetermined location by the fact that the absence of pulsation vibration, pulmonary respiratory vibration or vaginal vibration exceeds the predetermined absence continuation time.

  In the human absence detection device of Patent Document 1, a body vibration signal is obtained using one piezoelectric sensor, and the body vibration signal is separated by a frequency filter to obtain four signals of breathing, heartbeat, sputum, and body movement. It was. The frequency filter is a low-pass filter (LPF) or high-pass filter (HPF) analog filter composed of capacitors, resistors, operational amplifiers, etc., or a body vibration signal converted into a digital signal by an A / D converter and digitized data. It is disclosed that one or both of digital filters that perform filtering by arithmetic processing of a CPU (central processing unit) are used.

JP 2013-210367 A

Gary G Berntson, J. Thomas Bigger Jr. et al., “Heart rate variability: Origin, methods, and interpretive caveats”. Psychophysiology (USA) 34, p.623-648, 1997 Niizeki K and Saitoh T., “Incoherent oscillations of respiratory sinus arrhythmia during acute mental stress in humans”. American Journal of Physiology, Heart and Circulatory Physiology (USA) 302, p.359-367, 2012

  As described above, conventionally, in order to measure a sleep state, measurement in a medical institution or a large dedicated device is required, and it has been difficult to measure in a general normal life. In recent years, sleep disorders have frequently occurred, and a means for easily measuring, observing or evaluating a sleep state has been desired as part of self-management. In addition, since the conventional apparatus needs to fix a plurality of sensors to various parts of the body, and the movement is restricted by restraint by the sensors, there has been a demand for reducing the number of sensors as much as possible. In addition, since there is a possibility that the sensor may come off due to unconscious movement during sleep, it has also been desired that measurement can be performed with a sensor that can be easily attached. Furthermore, it has been required to alleviate the complexity of fixing a plurality of sensors to various parts of the body.

  As another problem, as described above, the stress evaluation method described in Non-Patent Document 1 has quantitatively large individual differences such as changes in indices even in the absence of a stress load, age dependency and fitness effects. There was a problem that could not be evaluated. Further, the method described in Non-Patent Document 2 requires an electrocardiogram sensor for detecting a heartbeat and a sensor for detecting a breathing pattern. In particular, as a sensor for detecting a breathing pattern, a sensor that measures a respiration rate by wearing a mask covering the mouth and nose is difficult to measure in daily life, during exercise, or during sleep. There was a risk that stress would be caused by wearing the mask. Therefore, it has been desired to develop a method that can evaluate stress in a stress-free state with less burden.

  As another problem, as described in Patent Document 1, when acquiring heart rate information and respiration information from a body vibration signal acquired from one piezoelectric sensor using a frequency filter, As heart rate information, it has been desired to obtain information as accurate as possible. In particular, there is a need for a method of indirectly acquiring heartbeat interval fluctuations and breathing patterns from a signal in which a plurality of pieces of information of heartbeat information or breathing information are mixed with accuracy that can also be used for stress evaluation as described above.

  An object of the present invention is to solve at least a part of the problems described above.

  In order to solve the above-mentioned problems, the sleep state measuring apparatus of the present invention provides an instantaneous phase of fluctuation of a heartbeat interval acquired during sleep of an animal, and an instantaneous respiratory pattern of the animal in the same time series as the fluctuation of the heartbeat interval. Phase coherence calculating means for calculating phase coherence based on the instantaneous phase difference from the phase is included. Furthermore, the sleep state measurement apparatus may include a sensor that measures an electrocardiogram, and a respiratory waveform extraction unit that extracts a signal related to a respiratory pattern from the electrocardiogram of the animal measured by the sensor. In addition, the sleep state measurement device includes a sensor that measures vibration, a heartbeat interval calculation unit that calculates a fluctuation of a heartbeat interval from a signal measured by the sensor, and a breath that extracts a signal related to a respiratory pattern from the signal measured by the sensor Waveform extracting means may be included. In the sleep state measuring device, the sampling frequency of the sensor is preferably 100 Hz or more. Furthermore, the sleep state measurement apparatus may have a determination function for determining a sleep state or the like based on the phase coherence calculated by the phase coherence calculation unit. The determination function may further determine a respiratory state during sleep from the respiratory pattern.

  Moreover, the method for measuring a sleep state of the present invention measures the sleep state based on the phase coherence of the instantaneous phase difference between the fluctuation of the heartbeat interval and the respiratory frequency in the same time series acquired during the sleep of the animal. Furthermore, in the method for measuring the sleep state, an electrocardiogram of the animal may be measured, and a signal related to a respiratory pattern may be acquired from the measured electrocardiogram. Further, in the method for measuring a sleep state, a cardiac motion diagram waveform or a biological vibration signal of the animal is measured, and a signal relating to a heartbeat interval variation and a respiratory pattern is obtained from the measured cardiac motion diagram waveform or the biological vibration signal. Further, the transfer characteristic of the heart and ball motion is estimated from the animal's heart and ball movement diagram waveform or the biological vibration signal, and the inverse transfer function of the transfer characteristic is determined as the animal's heart and ball movement diagram waveform or the biological vibration signal. It is preferable to use for acquiring a heartbeat waveform.

Further, the phase coherence calculation device of the present invention includes biological information acquisition means for acquiring biological information including at least information related to the heartbeat and respiration information of an animal, and respiratory waveform extraction means for extracting a respiratory pattern from the biological information. , Heart rate interval calculation means for calculating heartbeat interval fluctuations from the biological information, and phase coherence calculation means for calculating phase coherence of the instantaneous phase difference between the breathing pattern and heartbeat interval fluctuations. In the phase coherence calculating apparatus, the biological information may be an electrocardiogram, and the biological information may be a cardiogram waveform or a biological vibration signal. In the phase coherence calculation apparatus, the biological information is preferably acquired at a sampling frequency of 100 Hz or more.

  A stress measurement apparatus according to the present invention includes the above-described phase coherence calculation apparatus. Moreover, the sleep state measuring apparatus of this invention is equipped with the said phase coherence calculation apparatus.

  Further, the heartbeat waveform extracting method of the present invention is a method for extracting a heartbeat waveform from an animal's cardiac motion waveform or biological vibration signal, and the transfer characteristics of the cardiac motion from the cardiac motion diagram waveform or biological vibration signal. And the heartbeat waveform is acquired using the inverse transfer function of the transfer characteristic as the cardiogram waveform or biological vibration signal. In the above heartbeat waveform extraction method, the cardiac motion diagram waveform or the biological vibration signal is cut out so as to include one heartbeat, and a plurality of the waveform fragments are overlapped and averaged to transfer the cardiac motion. Is preferably estimated.

  Further, the heartbeat interval calculation method of the present invention is a method for calculating a heartbeat interval from an animal cardiogram waveform or biological vibration signal, and for a signal derived from the cardiogram waveform or biological vibration signal, A process of obtaining an absolute value of a signal after passing through a high-pass filter having a frequency lower than the heartbeat frequency as a lower limit frequency and passing through the high-pass filter is performed.

  The signal processing method of the present invention is a biological information signal processing method including biological information including information related to the heartbeat of an animal or information related to respiration, and an upper limit frequency from a power spectrum of a signal derived from the biological information. Alternatively, it includes a process of obtaining a lower limit frequency and passing through a filter having the upper limit frequency or the lower limit frequency as a cutoff frequency.

  The sleep state measuring apparatus and method of the present invention uses the fact that the phase coherence of the instantaneous phase difference between the instantaneous phase of the heartbeat interval and the instantaneous phase of the breathing pattern correlates with the amplitude of the δ wave. Can be measured. Phase coherence is less affected by the respiratory frequency and can more accurately measure the sleep state. In addition, it is possible to realize a sleep state measuring device with a single sensor, and it is possible to provide a device that is less burdensome for the user and less likely to feel stress. In particular, in the case of a sensor that measures vibration, measurement can be performed without restraining the animal, and a simpler device can be provided. The phase coherence can be measured in real time, the occurrence point of non-REM sleep time and the rhythm cycle of non-REM-REM sleep can be measured, and the sleep state can be accurately grasped. Moreover, since the respiratory pattern is measured or extracted at the same time, it is possible to detect an apnea state during sleep.

  The phase coherence calculation apparatus according to the present invention obtains the fluctuation of the heartbeat interval and the respiration pattern from the biological information including at least information related to the heartbeat and respiration of the animal, and calculates the phase coherence. An apparatus that is realized and less burdensome to the user can be provided. In particular, in the case of a sensor that measures vibration, measurement can be performed without restraining the animal, and a simpler device can be provided. By measuring the phase coherence, it is possible to confirm the psychological stress state and sleep state of the measurement subject in real time without depending on the respiratory frequency.

The figure which shows the temporal change of the amplitude of (delta) wave in the brain wave during sleep, and the phase coherence during sleep. The schematic block diagram of a sleep state measuring apparatus. One embodiment of a sleep state measuring device. (A) is an electrocardiogram waveform (upper), an extracted respiratory pattern (lower solid line), and an actual measured value (lower dotted line) of the respiratory pattern, (B) is a graph showing fluctuations in heartbeat intervals, (C) The solid line is the calculated phase coherence λ (lower) and the instantaneous phase (upper solid line) of the fluctuation of the heartbeat interval and the instantaneous phase of the breathing pattern (upper dotted line). The figure which shows the correlation of the estimated respiration frequency extracted from the electrocardiogram when changing the respiration rate by 8, 10, 12, 15, 18, 20, 24 times per minute, and the actual measurement respiration frequency. Phase coherence λ _ecg calculated using the estimated respiratory frequency when the respiration rate is changed at 8, 10, 12, 15, 18, 20, 24 times per minute and phase coherence λ calculated using the measured respiratory frequency FIG. The schematic block diagram of a phase coherence calculation apparatus. (A) is an electrocardiogram waveform (upper) and a simulated electrocardiogram waveform (lower) extracted from the electrocardiogram waveform, and (B) is an average transfer characteristic obtained from the electrocardiogram waveform. The figure which compared the phase coherence at the time of changing with 1kHz, 500Hz, 200Hz, 100Hz, 50Hz as a sampling frequency at the time of acquiring an electrocardiogram waveform. (A) is the instantaneous phase of the heartbeat (solid line) and breathing (dotted line) at rest, and (B) is a Lissajous diagram of the heartbeat and breathing instantaneous phase at rest. (C) is the instantaneous phase of heartbeat (solid line) and breathing (dotted line) during the mental arithmetic task (stress state), and (D) is a Lissajous diagram of the instantaneous phase of heartbeat and breathing during the mental arithmetic task ( E) and (F) are diagrams showing changes in respiratory arrhythmia (RSA) and phase coherence during a resting state and mental arithmetic task. The heart rate interval (RRI), respiratory arrhythmia amplitude (A _RSA ), phase coherence (λ), and respiratory frequency (f _R ) when the respiratory frequency is spontaneously changed from 15 times / minute to 25 times / minute are shown. Figure. The figure which shows the structure of a sheet type piezoelectric sensor. The figure which shows the signal waveform detected and calculated in the Example. The figure which shows the power spectral density (PSD) of the signal derived from an electrocardiogram and a biological vibration signal.

[Sleep state measuring device and method]
Sleep states are classified according to brain wave and eye movement patterns. The electroencephalogram is measured by an electroencephalogram (Electroencephalogram, “EEG”) that records temporal changes in the current generated by the potential difference associated with the neural activity of the brain. In the awake state, α waves (frequency: 8 to 13 Hz, amplitude: about 50 μV) are the main components when resting, but α waves are suppressed during mental activity, and β waves (frequency: 14 to Hz, amplitude) with a small amplitude. : In many cases, about 30 μV or less) appears. In a shallow sleep stage, α waves gradually decrease, θ waves (frequency: 4 to 8 Hz) appear, and δ waves (frequency: 1 to 4 Hz) appear during deep sleep. A β wave having a faster frequency than the α wave may be called a fast wave. In addition, an electroencephalogram having a frequency lower than that of an α wave is sometimes referred to as a slow wave, and a θ wave or a δ wave is a kind of slow wave.

Sleep is broadly classified into REM sleep and non-REM sleep, but non-REM sleep is further classified into four stages from stage I to IV. Stage I (slumbering period, falling asleep period) is a state in which the rhythm of the α wave is lost and it gradually flattens. Stage II is a state where sleep is deeper than stage I, and a sleep spindle and a K complex wave appear. Stage III is quite deep sleep, and δ waves account for 20% or more and less than 50%. Stage IV is a stage in which δ waves occupy 50% or more, and is the deepest sleep state. Non-REM sleep stages III and IV are also called slow wave sleep because the brain waves show a slow wave. On the other hand, REM sleep is a deep sleep state that requires a stronger stimulus than slow wave sleep in order to wake up, although it shows a brain wave with a low-amplitude fast wave pattern that resembles awakening, and is accompanied by rapid eye movement. The sleep state usually fluctuates in a cycle of about 90 minutes (sleep cycle), and sleep gradually deepens from stage I, and then sleep becomes shallow and then shifts to REM sleep. However, sleep time is also affected, and non-REM sleep is predominant in the first half of sleep, and REM sleep is predominant in the second half. As described above, a δ wave having a frequency of 1 to 4 Hz in an electroencephalogram is observed in slow wave sleep, and thus the δ wave can be used as an indicator of sleep depth.

  The present inventors have found that the phase coherence of the heartbeat interval variation during sleep and the instantaneous phase difference of the respiratory pattern (hereinafter referred to as “phase coherence”) correlates with the δ wave in the brain wave during sleep. It was invented that the sleep state can be measured by measuring the phase coherence. FIG. 1 shows temporal changes in the amplitude (δ) and phase coherence (λ) of a δ wave in an electroencephalogram of a subject during sleep. From FIG. 1, the phase coherence (λ) tends to approach 1 when the amplitude of the δ wave increases, and the phase coherence (λ) tends to approach 0 when the amplitude of the δ wave decreases, confirming the correlation between the two. In the measurement by 11 subjects, the average correlation coefficient between λ and δ waves was 0.53, and the standard deviation was 0.10. Moreover, the change of λ preceded the change of δ wave on average by 11.6 minutes.

  The phase coherence indicates the degree of variation in the phase difference between two signals. In the present invention, it indicates the degree of variation between the phase difference of the heart rate interval derived from respiration and the phase difference of the respiration pattern. The heart rate interval is shortened when inhaling (during inspiration) and longer when exhaling (during exhalation) due to respiratory arrhythmia (RSA). It fluctuates with a similar period. And according to the above findings, when the depth of sleep is deep (when δ waves are dominant), the fluctuation of the heartbeat interval and the variation in the phase difference of the breathing pattern are small (phase coherence is close to 1), As the depth decreases, the variation increases (phase coherence is close to 0).

  The phase coherence can calculate the instantaneous phase difference between the instantaneous phase of the fluctuation of the heartbeat interval and the instantaneous phase of the breathing pattern in the same time series from the information of each heartbeat interval and the breathing pattern. Information about each heartbeat interval can be acquired from biological information including heartbeat information. Biological information including heartbeat information includes, for example, an electrocardiogram obtained by measuring temporal changes in action potentials associated with heartbeats, a cardiocardiogram waveform obtained by measuring temporal changes in vibrations associated with heartbeats, and animal vibrations (cardiac bullets). A biological vibration signal or the like obtained by measuring a temporal change in the movement (including movement) can be used. The respiratory pattern can also be acquired from biological information including information on the respiratory pattern. Biological information including respiratory pattern information includes, for example, measured values that measure the flow of air due to breathing, measured values that measure changes in thoracic impedance associated with breathing, measured values that measure temperature changes due to breathing, and respiratory motion. Measured values, electrocardiograms, electrocardiogram waveforms, biological vibration signals, and the like obtained by measuring the accompanying abdominal movement can be used. An electrocardiogram measures temporal changes in myocardial action potential associated with a heartbeat from the body surface, but also includes temporal changes in body surface potential associated with respiration, both heartbeat information and respiratory pattern information. Is biometric information. In addition, the cardiocardiogram waveform measures temporal changes in vibration associated with the heartbeat, but also includes temporal changes in vibration associated with breathing, and includes both heartbeat information and respiratory pattern information. It is biometric information including. The biological vibration signal is for measuring the vibration of an animal. Such vibration includes vibrations associated with heartbeats and vibrations associated with respiration, and includes both information on heartbeats and respiratory pattern information. Information. By combining one or more of ECG, ECG waveform or biological vibration signal, heart rate information or respiratory pattern information is extracted at a level necessary for calculating phase coherence, and extracted heart rate information or respiratory pattern information Based on the information, it can be used to calculate phase coherence.

The phase coherence was calculated by calculating the instantaneous phase ψ _h (t) of the fluctuation of the heartbeat interval accompanying breathing and the instantaneous phase ψ _r (t) of the breathing pattern, and calculating the difference between these (instantaneous phase difference). Calculated using the instantaneous phase difference. For the instantaneous phase ψ _h (t) of the heartbeat interval fluctuation, the temporal change (S (t)) of the heartbeat interval fluctuation accompanying breathing is calculated from the heartbeat interval data, and the temporal change of the heartbeat interval fluctuation (S (t) )) Can be calculated by using the Hilbert transform represented by the following formula (2) as an analysis signal. Note that H [...] in the equations (2) and (3) is a Hilbert variable.

The instantaneous phase ψ _r (t) of the respiration pattern is calculated from the respiration pattern information by calculating the respiration pattern temporal change (R (t)), and the respiration pattern temporal change (R (t)) is expressed by the following equation. It can be calculated by Hilbert transform represented by (3).

The instantaneous phase difference ψ (t) is calculated using the instantaneous phase ψ _h (t) of the heartbeat interval fluctuation and the instantaneous phase ψ _r (t) of the breathing pattern calculated by the equations (2) and (3). (4).
Ψ (t) = ψ _h (t) −ψ _r (t) + 2nπ (4)
Here, n is an appropriate integer satisfying −π ≦ Ψ ≦ π.

From these, the phase coherence at time t _k can be calculated from the following equation (1). N in Expression (1) is the number of sampled data, and is obtained by averaging N data.

  As shown in FIG. 2, the sleep state measurement apparatus 1 using phase coherence includes at least an information acquisition unit 2 and an information processing unit 3. Furthermore, the sleep state measuring apparatus 1 may include an operation unit 4, an output unit 5, and a storage unit 6.

  The information acquisition unit 2 acquires information necessary for calculation of phase coherence, and may include a sensor for measuring animals and an input unit for inputting sensor information by wire or wirelessly. The configuration may include an input unit capable of inputting information from another recording medium in which already measured information is recorded, by wire or wirelessly. That is, the information acquisition unit 2 includes at least an input unit for inputting information, and may include a sensor for measuring biological information connected to the input unit by wire or wirelessly depending on circumstances. When measuring biological information with a sensor, the sampling frequency is preferably 100 Hz or more.

The information processing unit 3 processes input information and can use, for example, an arithmetic processing function of a CPU (central processing unit) of a computer. Also, during information processing,
It can also be realized by an analog circuit instead of a digital circuit. For example, when performing a frequency filter as information processing, it may be realized by an analog filter such as a low pass filter (LPF) or a high pass filter (HPF) constituted by a capacitor, a resistor, an operational amplifier, etc. You may implement | achieve with the digital filter which filters. The information processing unit 3 may include both a digital circuit and an analog circuit depending on the type of information processing. If the input information is analog, the information processing unit 3 converts the information into a digital signal by an analog-digital conversion circuit. May be. The information processing unit 3 has different functions or processes depending on the input information, but at least the phase coherence that calculates the phase coherence using the instantaneous phase difference between the instantaneous phase of the heartbeat interval fluctuation and the instantaneous phase of the breathing pattern. Has a calculation function. A device having a phase coherence calculation function is called a “phase coherence calculation device”, and the sleep state measurement device 1 of the present invention is also a kind of phase coherence calculation device.

  In order to calculate the phase coherence, for example, the following information can be input and calculated. A) The instantaneous phase difference between the instantaneous phase of the heartbeat interval variation and the instantaneous phase of the breathing pattern in the same time series is input to the information acquisition unit 2, and the instantaneous phase difference input by the information processing unit 3 using the phase coherence calculation function Is used to calculate the phase coherence. B) The information acquisition unit 2 has an instantaneous phase difference calculation function for inputting the instantaneous phase of the heartbeat interval variation and the instantaneous phase of the breathing pattern in the same time series, and the information processing unit 3 calculates the instantaneous phase difference between them. Then, the instantaneous phase difference is calculated by the instantaneous phase difference calculation function, and the phase coherence is calculated by the phase coherence calculation function using the calculated instantaneous phase difference. C) The heart rate interval variation and breathing pattern in the same time series are input to the information acquisition unit 2, and the information processing unit 3 has an instantaneous phase calculation function, and the instantaneous phase calculation function The instantaneous phase of the respiratory pattern is calculated, and the phase coherence is calculated by the instantaneous phase difference calculating function and the phase coherence calculating function using the calculated instantaneous phase. D) Biological information including information related to heartbeats and biological information including information related to breathing are input to the information acquisition unit 2, and the information processing unit 3 has a heartbeat interval calculation function and a breathing pattern calculation function, The calculation function calculates the fluctuation of the heartbeat interval from the biological information including the heartbeat information, and the respiratory pattern calculation function calculates the respiratory pattern from the biological information including the respiratory pattern information. Thereafter, the same processing as the above C) I do. E) A function for inputting biological information including both information relating to heartbeat and information relating to breathing to the information acquisition unit 2 and for the information processing unit 3 to detect or extract heartbeat information or respiratory pattern information from the biological information. The subsequent processing may be performed using the detected or extracted heartbeat information or respiratory pattern information.

  Furthermore, the information processing unit 3 may have a determination function for determining a sleep state or the like based on the calculated phase coherence. As a determination function, for example, the calculated phase coherence is compared with a threshold, and if it is larger than the threshold, it may be determined that the sleep is deep, and if it is smaller, it may be determined that the sleep is shallow, or the phase coherence is less than the threshold. Alternatively, the sleep quality may be evaluated based on the time when the value is greater than the value, or the sleep quality may be evaluated based on the phase coherence fluctuation period. The threshold value may be a predetermined numerical value, may be specified from the numerical value of the phase coherence calculated in the past of the measurement target, or a plurality of numerical values may be set and the sleep quality may be evaluated step by step. Also good. Furthermore, the respiratory function during sleep can be determined by the determination function of the information processing unit 3. For example, although depending on the method of acquiring the breathing pattern, in any of the acquisition methods, the central sleep apnea (the apnea that occurs when the respiratory motion stops due to abnormalities in the respiratory center of the brain, etc.) Since the movement stops, the breathing pattern cannot be detected, so it can be determined that the patient is in an apnea state. In addition, if the breathing pattern is obtained by measuring the airflow due to breathing, the breathing airflow stops and the breathing pattern is no longer detected, so not only central sleep apnea but also obstructive sleep apnea (Although there is respiratory motion, apnea due to airway obstruction, etc.) can also be determined.

  The operation unit 4 is provided with operation terminals such as a switch, a touch panel, a button, a knob, a keyboard, a mouse, and a voice input microphone for the user to operate the sleep state measuring apparatus 1. In addition, the operation unit 4 may be provided with a display for displaying operation contents and the like. The output unit 5 may output the calculated phase coherence, may output biological information other than the phase coherence, or may output the sleep state determined by the determination function. As the output unit 5, a display that displays the result as an image, a printer that outputs the result as paper, a speaker that outputs the result as sound, a wired or wireless output terminal that outputs the result as electronic information, and the like can be used. . In addition, the structure as which the display as the output part 5 is combined with the display which displays the touchscreen in the operation part 4, an operation content, etc. may be sufficient. The storage unit 6 can store information acquired by the information acquisition unit 2, results calculated by the information processing unit 3, results determined by the determination function, and the like.

  FIG. 3 is an example of the sleep state measuring apparatus 1. The sleep state measuring apparatus 1 includes a sensor 21, an analog-digital conversion circuit 31, a heartbeat extraction means 32, a respiratory waveform extraction means 33, a heartbeat interval calculation means 34, a Hilbert transform filters 35 and 36, an instantaneous phase difference calculation means 37, a phase coherence. A calculation unit 38, operation buttons 41, a touch panel 42, a voice input microphone 43, a display display 44, a wireless communication unit 45, a speaker 46, and a recording device 47 are provided.

  The sensor 21 detects biological information including information related to heartbeat and biological information including information related to respiration. For example, there is an electrocardiogram measurement sensor or a vibration measurement sensor that detects biological information including information related to the heartbeat, and an electrocardiogram measurement sensor or vibration that detects biological information including information related to respiration. Sensor or breath sensor. Although the number is one in FIG. 3, a plurality of types of sensors may be included. An electrocardiogram measurement sensor or a vibration measurement sensor can detect a signal including both biological information related to heartbeats and biological information related to respiration, and can be detected by one sensor. May be detected by another sensor. In the case of an electrocardiogram measurement sensor, it is preferable to measure by attaching a dedicated electronic circuit to the chest of a living body using a disposable electrode, and an electrocardiogram waveform is measured by the electrode. The derivation method may be unipolar induction or bipolar induction.

The sensor for measuring the vibration may be a contact type or a non-contact type. In the case of a sensor for measuring the contact type vibration, it is arranged in contact with the animal directly or indirectly, so that the cardiocardiogram waveform. Alternatively, a biological vibration signal can be detected. Sensors that measure contact-type vibrations for detecting cardiopulsatile waveforms or biological vibration signals are placed directly or in the vicinity of various organisms that generate vibrations, and detect vibrations from living organisms and output them as electrical signals It is enough if possible. As a sensor for measuring vibration, a piezoelectric element is preferably used as a piezoelectric sensor, but other sensors such as a polymer piezoelectric material (polyolefin-based material) may be used. As a material of the piezo element, for example, a porous polypropylene electret film (ElectroMechanical Film (EMFI)), PVDF (polyvinylidene fluoride film), or polyvinylidene fluoride and ethylene trifluoride copolymer (P (VDF-TrFE)). Or polyvinylidene fluoride and a tetrafluoroethylene copolymer (P (VDF-TFE)) may be used. The piezoelectric sensor is preferably in the form of a film. Furthermore, the piezoelectric sensor is preferable because it can acquire a cardiogram waveform or a biological vibration signal without restraining the animal, and can measure more stress-free. However, the piezoelectric sensor can also be used by attaching it to an animal by attaching it to a wristband, belt or wristwatch. In addition, as a sensor for measuring other types of vibrations, for example, a high-sensitivity acceleration sensor is used, such as a wristwatch or a portable terminal, in contact with the body, or an acceleration sensor is installed on a part of a bed, chair, etc. Then, the cardiocardiogram waveform or the biological vibration signal may be acquired, or the change of the air pressure or the liquid pressure in the tube is detected by a pressure sensor or the like to acquire the cardiocardiogram waveform or the biological vibration signal. Also good. Furthermore, as a sensor for measuring vibration, a non-contact type sensor that can acquire a cardiodynamic chart waveform or a biological vibration signal in a non-contact manner along with signal transmission / reception using a microwave or the like may be used. Microwave Doppler sensor, UWB (ultra wide band) impulse reflection delay time is measured, and the distance from the object is measured. Cardiac motion chart waveform or biological vibration signal obtained by using the LED, cardiac motion chart waveform or biological vibration signal obtained from the reflected or transmitted light using the LED light, and heart obtained from the reflected wave of the ultrasonic wave A ballistic waveform or a biological vibration signal may be used. These sensors using microwaves and the like can be miniaturized, can acquire signals without contact and without constraint, and can acquire signals from a remote location. The acceleration sensor can also be reduced in size.

  In the case of a respiration sensor, for example, an actual value obtained by measuring the air flow due to respiration, an actual value obtained by measuring a change in thorax impedance due to respiration, an actual value obtained by measuring a temperature change due to respiration, and an abdominal movement accompanying respiration exercise The time change of the breathing pattern is measured by the actually measured value or the like. The signal detected by the sensor 21 is input to the analog-digital conversion circuit 31 of the sleep state measurement device 1 by wire or wireless. The upper graph in FIG. 4A is an electrocardiogram waveform measured by an electrocardiogram measurement sensor.

  The analog-digital conversion circuit 31 is a circuit that converts an analog signal from the sensor 21 into a digital signal. An analog-digital conversion circuit 31 may be provided in the sensor 21, or may not be provided when the sensor 21 can detect a digital signal. Further, the analog signal from the sensor 21 may be converted into a digital signal by the analog-digital conversion circuit 31 after performing processing such as filtering.

  The heartbeat extracting means 32 is a means for extracting a signal relating to heartbeat from the signal detected by the sensor 21, and an appropriate process is appropriately selected according to the type of sensor or the input signal. When an ECG waveform or ECG waveform is input, since the ECG waveform or ECG waveform is usually affected by respiration, it is preferable to perform processing to remove the respiratory component, If there is no problem in the calculation of the heartbeat interval, the heartbeat extraction means may not be used. In addition, when a biological vibration signal is input, the biological vibration signal usually includes not only the cardiac movement due to the heartbeat but also vibration based on breathing, vibration based on body movement, vocalization, external environment, and the like. In some cases, it is preferable to perform a process for removing these noises. As such processing, for example, the intensity of the electrocardiogram waveform or vibration signal is enhanced to the nth power (where n is an integer equal to or larger than 2 and when n is an odd number), and then the bandpass filter (BPF) ) May be allowed to pass. The BPF of the heart rate extraction means 32 preferably has a lower limit frequency of the pass band of 0.5 Hz or more, 0.6 Hz or more, 0.7 Hz or more, 0.8 Hz or more, 0.9 Hz or 1 Hz or more, and an upper limit frequency of 10 Hz. Hereinafter, it is preferably 8 Hz or less, 6 Hz or less, 5 Hz or less, or 3 Hz or less, and preferably has a pass band that combines any one of these lower limit frequencies and any one of the upper limit frequencies. The lower limit frequency of the heartbeat extracting means 32 may be the same as the upper limit frequency of the breathing waveform extracting means 33, or lower than the upper limit frequency of the breathing waveform extracting means 33, and a part of the passband of the heartbeat extracting means 32 is breathing. It may be superimposed on the passband of the waveform extraction means 33. Further, as the heartbeat interval extraction method, it is preferable to obtain the upper limit frequency or the lower limit frequency of the filter from the acquired cardiogram waveform or biological vibration signal, and more preferably, the acquired cardiogram is periodically or irregularly. It is preferable to obtain the upper limit frequency or the lower limit frequency of the filter from the waveform or the biological vibration signal. For example, the power spectrum of a cardiodynamic waveform or a biological vibration signal or those signals pre-processed (for example, noise removal, enhancement processing, etc.) (a signal derived from a cardiac movement diagram waveform or a biological vibration signal) Find the density from 0.5Hz or more, identify the first peak, and pass through the frequency band of the low frequency side and / or high frequency side where the peak falls to a predetermined threshold (for example, half width of the peak) It is also good. The power spectrum can be obtained by, for example, Fourier transform. In this way, by performing signal processing using a filter having an upper limit frequency or a lower limit frequency obtained from the acquired cardiocardiogram waveform or biological vibration signal, the biological information unique to the acquired biological body and the posture and physical condition at the time of acquisition are obtained. In addition, conditions such as environment were reflected, filters corresponding to individual differences and acquisition conditions could be set, and phase coherence could be calculated in real time. When a respiration sensor is used as a part of the sensor 21, it is not necessary to input a signal from the respiration sensor to the heartbeat extracting means 32.

  Further, as a method of calculating the heartbeat interval from the heart ballistic waveform, the heart ballistic waveform is passed through a high pass filter (HPF) having a frequency lower than the heartbeat frequency as a lower limit frequency, and the absolute value of the signal after HPF From the envelope signal of the signal that has passed through the used HPF, it is possible to obtain each heartbeat peak of the electrocardiogram waveform, and calculate the heartbeat interval from the peak value or the start point of the heartbeat peak. Can be sought. The frequency of a normal heartbeat is about 3 Hz at the maximum, but the lower limit frequency of this high-pass filter is preferably 5 Hz or more, and may be 10 Hz, 20 Hz, 30 Hz, or 40 Hz. The value of the phase coherence (λ) obtained by calculating the fluctuation of the heartbeat interval by this signal processing method was very close to the value of the phase coherence (λ) obtained from the electrocardiogram waveform. It is more preferable to use the BPF (or LPF) of the pass frequency obtained from the above power spectrum for the signal obtained by taking the absolute value of the signal after HPF. Furthermore, preprocessing for noise removal and the like may be performed before passing the HPF. A biological vibration signal (including a cardiocardiogram waveform) includes a vibration waveform associated with a heartbeat, but when a vibration component associated with respiratory motion is superimposed on a heartbeat component, the waveform is not stable, and corresponds to RRI from the original waveform. Conventionally, it has been difficult to obtain a beat interval for each beat. In this method, the respiratory frequency component superimposed on the fundamental frequency component of the heartbeat is removed in advance, and then the beat interval is obtained from the high-frequency vibration component derived from the heartbeat, so that a more accurate beat interval can be obtained. As a result, the detection of respiratory arrhythmia, which is a fluctuation of the heartbeat interval, is also accurate, and the calculated phase coherence almost coincides with that obtained from the electrocardiogram and the measured respiration.

  The respiration waveform extraction unit 33 is a unit that extracts a signal related to a respiration pattern from the signal detected by the sensor 21, and an appropriate process is appropriately selected according to the type of sensor or the input signal. When a respiration sensor is used as a part of the sensor 21 and the respiration pattern is actually measured by the respiration sensor, the respiration waveform extraction means 33 may not be provided, and the respiration waveform extraction means 33 removes a signal that becomes noise. You may perform the process to do. In the case of extracting a signal related to a respiratory pattern from an electrocardiogram waveform, a cardiocardiogram waveform, or a biological vibration signal, as such processing, for example, the intensity of the electrocardiogram waveform or the vibration signal is set to the nth power (n is an integer of 2 or more If n is an odd number, an absolute value is taken), and after emphasizing processing, a low pass filter (LPF) having a pass band in a frequency range of 0.5 Hz or less may be passed. The cutoff frequency of the LPF of the respiratory waveform extraction means 33 may be 0.3, 0.4, 0.6, 0.7 Hz, and 0.8 Hz. Further, the cutoff frequency of the respiration waveform extraction unit 33 may be the same as the lower limit frequency of the heartbeat extraction unit 32, or may be higher than the lower limit frequency and a part of the passband may be superimposed. Further, as a respiratory waveform extraction method, it is preferable to obtain an upper limit frequency or a lower limit frequency of the filter from the acquired electrocardiogram waveform, cardiogram waveform or biological vibration signal, and more preferably, the acquired heart is periodically or irregularly acquired. It is preferable to obtain the upper limit frequency or the lower limit frequency of the filter from the ballistic waveform or the biological vibration signal. For ECG waveforms, ECG waveforms, biological vibration signals, or pre-processed signals (for example, noise removal filters or enhancement processes), find the power spectrum and search for the power spectrum density from the low frequency side. The first peak is identified, and the frequency on the high frequency side where the peak falls to a predetermined threshold (for example, the half-value width of the peak) may be set as the cutoff frequency. In this way, by performing signal processing using the filter of the upper limit frequency or the lower limit frequency obtained from the acquired electrocardiogram waveform, cardiocardiogram waveform or biological vibration signal, the biological information specific to the acquired biological information or Conditions such as posture, physical condition, environment, etc. were reflected, filters corresponding to individual differences and acquisition conditions could be set, and phase coherence could be calculated in real time. Further, the BPF may be passed instead of the LPF. In this case, it is sufficient that the lower limit frequency of the BPF is a sufficiently low frequency, for example, it may be set to 0.1 Hz. The lower graph in FIG. 4A shows a respiration pattern, the solid line is a respiration pattern extracted from an electrocardiogram waveform, and the dotted line is an actual measurement value obtained by measuring the air flow due to respiration. From the lower graph of FIG. 4 (A), it can be confirmed that the cycle matches the measured value even in the respiration pattern extracted from the electrocardiogram waveform.

FIG. 5 shows the correlation between the estimated respiratory frequency extracted from the electrocardiogram and the measured respiratory frequency when the respiratory rate is changed by 8, 10, 12, 15, 18, 20, 24 times per minute in 8 subjects. FIG. The dotted line indicates the 95% confidence interval of the respiratory frequency. If the respiration frequency exceeds about 0.4 Hz, the 95% confidence interval falls out of the identity line and is underestimated, but if the respiration frequency is less than 0.4 Hz, the respiration frequency can be estimated with an accuracy within 10%. FIG. 6 is a diagram showing the correlation between the phase coherence λ _ecg calculated using the estimated respiration frequency of FIG. 5 and the phase coherence λ calculated using the actually measured respiration frequency. Until the respiratory frequency was 0.33 Hz (20 times / min), phase coherence was obtained with an accuracy of 90% or more.

  The heartbeat interval calculation means 34 receives the signal from the heartbeat extraction means 32 and calculates the interval between heartbeats. As the interval between heartbeats, for example, the interval between the P wave, R wave, T wave or U wave of the electrocardiogram, particularly the R wave has a sharp peak, and therefore it is preferable to measure the interval from the R wave to the next R wave. In the case of a heartbeat diagram waveform or a signal related to a heartbeat extracted from a biological vibration signal, it is preferable to measure an interval between waveforms corresponding to a sharp peak R wave. FIG. 4B is a graph showing fluctuations in the heart rate interval calculated from the electrocardiogram in FIG. 4A. The vertical axis is the heart rate interval (ms) and the horizontal axis is time (s). From FIG. 4B, it can be confirmed that the heartbeat interval fluctuates at a constant period. Note that the amplitude of respiratory arrhythmia (RSA) can also be used as one of indices for evaluating psychological stress and the like, but as will be described later, the amplitude of respiratory arrhythmia (RSA) also changes depending on the respiratory frequency. It is preferable to make an auxiliary or additional evaluation in combination with the evaluation by phase coherence. In addition, the heartbeat interval calculating unit 34 may directly calculate the interval between heartbeats from biological information including information related to the heartbeat. In this case, the heart rate interval calculating unit 34 including the function of the heart rate extracting unit 32 may be used. Depending on the signal processing method, the heart rate extracting unit 32 may not be required, and the heart rate can be directly calculated from the biological information including information on the heart rate. The interval can be calculated.

  The Hilbert transform filter 35 outputs an instantaneous phase and an instantaneous amplitude for fluctuations in the heartbeat interval, and the Hilbert transform filter 36 outputs an instantaneous phase and an instantaneous amplitude for a breathing pattern. In the Hilbert transform, a 90-degree phase difference wave detector may be realized by an analog circuit, or may be constituted by a finite impulse response type digital filter. The analytic signal is obtained by adding the Hilbert-transformed signal and the real signal, and the instantaneous phase can be obtained from the ratio of the real part and the imaginary part of the analytic signal. In the upper graph of FIG. 4C, the solid line is the instantaneous phase of heartbeat interval fluctuation, the dotted line is the instantaneous phase of the breathing pattern, the vertical axis is the phase (radian), and the horizontal axis is time (s). It is.

  The instantaneous phase difference calculating unit 37 calculates the phase difference (instantaneous phase difference) between the instantaneous phase of heartbeat interval fluctuation and the instantaneous phase of the breathing pattern, and outputs the result to the phase coherence calculating unit 38. As described above, the phase coherence calculating unit 38 calculates the phase coherence using the instantaneous phase difference between the instantaneous phase of the heartbeat interval fluctuation and the instantaneous phase of the breathing pattern. The data for obtaining the phase coherence is calculated with the window length of one breath cycle at the minimum. The lower graph in FIG. 4C is the calculated phase coherence λ. In FIG. 4C, it can be confirmed that the phase coherence is always close to 1 and there is relatively little variation. The sleep state measurement apparatus 1 may further have a function of determining a sleep state from the calculated phase coherence λ.

The operation button 41, the touch panel 42, and the voice input microphone 43 are input means for the user to operate the sleep state measurement device 1, and operate the sleep state measurement device 1 or output necessary information. be able to. The display 44 and the speaker 46 can be used as output means for outputting a heartbeat, a respiratory pattern, a respiratory arrhythmia, a phase coherence λ, a sleep state estimated from the phase coherence λ, and the like. The wireless communication means 45
The calculated phase coherence λ or sleep state may be used as an output means, or may be used as an input means for inputting a signal from the sensor 20. Or you may output a sleep state etc. with an audio | voice. The recording device 47 records input information, programs of various means, measurement results, and the like.

  Sleep state measuring device 1 can also be realized by a portable terminal (for example, a cellular phone, a smart phone, etc.) and a sensor. An A / D conversion circuit and a wireless communication function are provided in a sensor, for example, an electrocardiogram sensor, a signal detected by the sensor is converted into a digital signal by the A / D conversion circuit, and the digital signal is transmitted to the portable terminal by the wireless communication function. As the wireless communication function, for example, Bluetooth (registered trademark), Wi-fi (registered trademark), or the like is preferably used.

  The sleep state measurement apparatus 1 of the present invention uses the fact that the phase coherence of the instantaneous phase difference between the instantaneous phase of the heartbeat interval and the instantaneous phase of the breathing pattern correlates with the amplitude of the δ wave. Can be measured. As will be described later, the phase coherence is less affected by the respiratory frequency and can measure the sleep state more accurately. Moreover, it is possible to realize the sleep state measuring apparatus 1 with a single sensor, and it is possible to provide an apparatus that is less burdensome to the user and less likely to feel stress. In particular, in the case of a sensor that measures vibration, measurement can be performed without restraining the animal, and a simpler device can be provided. The phase coherence can be measured in real time, the occurrence point of non-REM sleep time and the rhythm cycle of non-REM-REM sleep can be measured, and the sleep state can be accurately grasped. Moreover, since the respiratory pattern is measured or extracted at the same time, it is possible to detect an apnea state during sleep. Furthermore, since the magnitude of the respiratory arrhythmia of the measurement subject can also be measured, it can be used as an auxiliary determination index for the sleep state.

[Phase Coherence Calculation Apparatus and Method]
Phase coherence can be used not only for sleep states but also for assessment of psychological stress. In particular, the phase coherence calculating apparatus described in the present embodiment calculates the phase coherence by acquiring the fluctuation of the heartbeat interval and the respiration pattern from the biological information including at least both information related to the heartbeat and respiration of the animal.

  FIG. 7 is a schematic block diagram of the phase coherence calculation device 11. The phase coherence calculation device 11 includes at least a biological information acquisition unit 12, a respiratory waveform extraction unit 13, a heartbeat interval calculation unit 14, and a phase coherence calculation unit 15. Furthermore, the phase coherence calculation device 11 may include other means described in FIG. In the phase coherence calculation apparatus 11 in the present embodiment, the biological information acquisition unit 12 acquires biological information including at least both information related to the heartbeat and respiration of the animal. As such biological information, for example, an electrocardiogram, Examples include a cardiogram waveform or a biological vibration signal. The biological information acquisition means 12 may include a sensor for measuring an animal and an input unit for inputting sensor information by wire or wireless, or another recording medium in which already measured information is recorded. It may be configured to include an input unit that can input information from That is, the information acquisition unit 2 includes at least an input unit for inputting information, and may include a sensor for measuring biological information connected to the input unit by wire or wirelessly depending on circumstances. When measuring biological information with a sensor, the sampling frequency is preferably 100 Hz or more.

As described in detail in the description of the sleep state measuring apparatus 1, the breathing pattern can be extracted from the electrocardiogram and the fluctuation of the heartbeat interval and the breathing pattern can be acquired. In the case of an electrocardiographic waveform or a biological vibration signal, a sensor that measures vibration may be in direct contact with the living body, but a member (floor, bed, chair, desk, clothes, shoes, carpet, etc.) that propagates vibration from the living body. , A sheet, a cover, etc.), and in the case of a non-contact type, they may be arranged away from the living body. The member on which the sensor for measuring contact-type vibration is installed may be a member in which a plurality of members are interposed from a living body. For example, sensors for measuring vibration may be installed or embedded on the floor, on or under the bed mat, on the front or back surface of the seat plate or back of the chair, or on the front or back surface of the desk top. Further, in the leg of the bed, a sensor for measuring vibration may be provided in one of the protective members provided in contact with the floor, or may be provided on a bed frame, a headboard, a side rail, or the like. It may be provided on a chair leg, armrest, frame or the like, or may be provided on a desk leg, penetrating board, curtain, or the like. Moreover, you may provide the sensor which measures a vibration in the toilet seat or toilet bowl of a toilet, and the sensor which measures a vibration can be arrange | positioned on the surface, back side, or inside of a toilet seat or a toilet bowl. For example, you may arrange | position the sensor part which measures a vibration in the buffer part provided in the contact location with the toilet bowl in the back surface of a toilet seat, or the contact location with the toilet seat in the surface of a toilet bowl. In addition, as a sensor which measures a vibration, a piezoelectric sensor, an acceleration sensor, a pressure sensor, a non-contact type sensor, etc. can be used as above-mentioned.

  In the case of a cardiogram waveform or a biological vibration signal, it is necessary to extract information on heartbeats and information on breathing. By using a low-pass filter (LPF), a band-pass filter (BPF), and a high-pass filter (HPF), it is possible to extract information related to heartbeat and information related to respiration by frequency. In order to more accurately determine the heartbeat from the ECG waveform or biological vibration signal, the transfer characteristic of the ECP is calculated and the inverse transfer function is estimated to correspond to the ECG from the ECG waveform or biological vibration signal. Waveform to be obtained. The inverse transfer function may measure the electrocardiogram and cardiac motion of the person to be measured in advance and investigate the transfer characteristics, but it is also possible to estimate the transfer characteristics from only the electrocardiographic waveform or biological vibration signal. is there. After removing low frequency components from the ECG waveform, a Wiener filter or the like is applied, and a reverse EC function and the original ECG waveform are superimposed and integrated to obtain a simulated ECG waveform. Further, as another method for calculating the heartbeat interval from the cardiocardiogram waveform, the signal after HPF is passed through the high-pass filter (HPF) whose lower limit frequency is higher than the heartbeat frequency. By taking the absolute value, it is possible to obtain each heartbeat peak of the cardiogram waveform from the envelope signal of the signal that has passed through the used HPF, and the heartbeat from the peak value or the start point of the heartbeat peak. An interval may be obtained.

  In FIG. 8A, the upper graph is a cardiac electrocardiogram waveform, and the lower graph is a simulated electrocardiogram waveform extracted from the cardiac ballistic waveform. First, since sharp peaks occur periodically in the ECG waveform, the peak time (T) is obtained, and the time is estimated as the R wave time obtained from the ECG. Starting from a sharp peak of the waveform, the accompanying cardio-acoustic diagram waveform is cut out and collected by waveform fragments of a predetermined window length (up to the next peak at the longest). It is preferable to eliminate waveform fragments having large variations when they are put together, assuming that the time (T) does not coincide with the R-wave time. The solid line in FIG. 8B is the average when 100 waveform fragments are superimposed, and the dotted line is the standard deviation. This solid line is the average transfer characteristic between the electrocardiogram and the electrocardiogram, and a simulated electrocardiogram waveform can be obtained from the electrocardiogram waveform using the inverse transfer function of the transfer characteristic. In this way, it is possible to more accurately extract the waveform corresponding to the electrocardiogram from the cardiogram diagram or the biological vibration signal by calculating the transfer characteristic of the cardiogram and estimating the inverse transfer function. In particular, the phase coherence calculation apparatus and sleep state measurement apparatus of the present invention extract heartbeat intervals and breathing patterns, and it is important to obtain a waveform corresponding to an electrocardiogram more accurately.

  Further, a signal related to respiration can be extracted from the electrocardiogram waveform by passing it through a low-pass filter (LPF), or by extracting amplitude modulation derived from respiration from the simulated electrocardiogram waveform.

  When measuring biological information with a sensor, the sampling frequency is preferably 100 Hz or more. FIG. 9 is a diagram comparing phase coherence when the sampling frequency when acquiring an electrocardiogram waveform is changed to 1 kHz, 500 Hz, 200 Hz, 100 Hz, and 50 Hz. The root mean square error (RMSE) compared with the phase coherence of the electrocardiogram sampled at 1 kHz is 0.028 at 500 Hz, 0.039 at 200 Hz, 0.045 at 100 Hz, and 0.109 at 50 Hz. It was. Thus, if the sampling frequency is 100 Hz or more, sufficiently accurate phase coherence can be obtained.

  FIG. 10 shows the relationship between the fluctuation of the heart rate interval and the instantaneous phase of the breathing pattern at rest and during mental arithmetic task (stress state). FIG. 10A is an instantaneous phase of a heartbeat (solid line) and breathing (dotted line) at rest, and FIG. 10B is a Lissajous diagram of an instantaneous phase of heartbeat and breathing at rest. FIG. 10C is an instantaneous phase of the heartbeat (solid line) and breathing (dotted line) during the mental arithmetic task (stress state), and FIG. 10D is a Lissajous diagram of the instantaneous heartbeat and respiratory phase during the mental arithmetic task. . FIG. 10E shows changes in respiratory arrhythmia (RSA) when a mental arithmetic task is imposed from a resting state and after the mental arithmetic task is completed. FIG. 10F shows the phase coherence change when the mental arithmetic task is imposed from a resting state and after the mental arithmetic task is finished, and the dotted line is the phase coherence calculated using the actually measured respiratory pattern based on the respiratory flow velocity. The solid line is the phase coherence calculated using the respiration pattern calculated from the electrocardiogram. The phase coherence at rest is 0.69 ± 0.12 (95% confidence interval: 0.63 to 0.75), and the phase coherence during mental arithmetic task is 0.45 ± 0.17 (95% confidence). The interval was 0.41 to 0.49), and the phase coherence was significantly reduced during the mental arithmetic task as compared to the rest. From FIG. 10, the instantaneous phase difference is very stable at rest. When mental stress such as a mental arithmetic task is imposed, not only the magnitude of respiratory arrhythmia (RSA) is attenuated, but also the phase difference is disturbed. I can confirm. This means that the cooperative relationship between the respiratory oscillator generated by the respiratory center and the heart beat oscillator under autonomic innervation is disturbed by stress.

FIG. 11 is a diagram showing the heartbeat interval (RRI), respiratory arrhythmia amplitude (ARSA), phase coherence (λ), and respiratory frequency (f _R ) when the respiratory frequency is spontaneously changed.
Initially, the respiration rate was 15 times per minute, but was consciously increased from 300 s to 25 times per minute. As the respiratory rate increases, the respiratory arrhythmia amplitude (A _RSA ) decreases, but the phase coherence hardly changes. Respiratory arrhythmia is a physiological phenomenon derived from autonomic parasympathetic nerve activity, and its amplitude is said to reflect the degree of parasympathetic nerve activity. As shown in FIG. 11, the degree of parasympathetic nerve activity changes. However, autonomic nerve activity could not always be estimated from respiratory arrhythmia because the amplitude of respiratory arrhythmia is affected only by changing the respiratory rate. In this respect, since phase coherence does not change even if the respiratory rate is changed, the autonomic nerve activity can be estimated more accurately by estimating the autonomic nerve activity from the phase coherence. Thus, by measuring the phase coherence, it is possible to confirm the psychological stress state and sleep state of the measurement subject in real time without depending on the respiratory frequency. Furthermore, since the magnitude of the respiratory arrhythmia of the measurement subject can also be measured, it can be used as an auxiliary determination index for the psychological stress state and the sleep state. Furthermore, since fluctuations in the heartbeat interval can also be acquired as information, a slow fluctuation component (LF) and a fast fluctuation component (HF) can be obtained by performing frequency analysis (Fourier transform) on the time change of the fluctuation in the heartbeat interval. It is also possible to detect LH / HF as a sympathetic nerve activity index and detect HF as a parasympathetic nerve activity index. For example, the component of 0.04 to 0.15 Hz in the power spectrum obtained by Fourier transforming the time change of the heartbeat interval may be calculated as LF, and the component of 0.15 to 0.4 Hz may be calculated as HF. Conventionally, these indexes have been used as an analysis of fluctuations in the R-wave heartbeat interval obtained from an electrocardiogram. It can also be used as an analysis method for fluctuations in heart rate intervals, and it has become possible to investigate dysregulation of sympathetic / parasympathetic nerve activity. In addition, when an electrocardiogram is used in the sleep state measurement device of the present invention, these indices may be detected simultaneously.

As a signal processing method of biological information including information related to the heartbeat of an animal or biological information including information related to respiration, an upper limit frequency or a lower limit frequency is obtained from a power spectrum of a signal derived from biological information, and the upper limit frequency or the lower limit frequency is determined. It is preferable to include a process of passing through a filter having a cutoff frequency. By obtaining the upper limit frequency or the lower limit frequency from the power spectrum of the signal derived from the biological information, the biological information unique to the acquired biological information and the conditions such as the posture, physical condition, environment, etc. when acquiring the biological information are reflected. It is possible to set a filter corresponding to a difference or a condition at the time of acquisition. Furthermore, since conditions such as posture, physical condition, environment, etc. always change, it is preferable to update the upper limit frequency or lower limit frequency of the filter regularly or irregularly.

  The biological information including information about the heartbeat can be, for example, an electrocardiogram, a cardiocardiogram waveform, a biological vibration signal obtained by measuring temporal changes in animal vibration (including cardiopulsation), and the like. The biological information including information includes, for example, an actual measurement value obtained by measuring the air flow due to respiration, an actual measurement value obtained by measuring a change in thorax impedance accompanying respiration, an actual measurement value obtained by measuring a temperature change caused by respiration, Actual values obtained by measuring movement, electrocardiograms, electrocardiogram waveforms, biological vibration signals, and the like can be used. The signal derived from biological information is not only biological information including information related to the heartbeat or biological information including information related to respiration itself, but also pre-processed (for example, noise removal and enhancement processing) on the biological information. Including things. Such a signal processing method, for example, in the sleep state measurement device, calculates heart rate interval fluctuation from biological information including heart rate information by the heart rate interval calculation function, or includes respiration pattern information by the respiration pattern calculation function. It may be used when calculating a respiratory pattern from biological information, or in a phase coherence calculating device, calculating fluctuations in heartbeat intervals from biological information including both information related to the heartbeat of an animal and information related to respiration. It may be used when calculating a breathing pattern.

  [Embodiment] In this embodiment, biological information is obtained using a sheet-type piezoelectric sensor (piezo element) as a sensor for measuring vibration, and phase coherence is obtained from the biological information. At the same time, the subject's respiratory pattern was measured with the subject's electrocardiogram and a hot-wire respiratory velocimeter, and the phase coherence was also obtained from these biological information.

  FIG. 12 shows the structure of the sheet-type piezoelectric sensor 210 used in this example. The sheet-type piezoelectric sensor 210 has a positive electrode layer 212 and a negative electrode layer 213 on the upper and lower sides with a sheet-like vibration sensor material 211 interposed therebetween, and further covers these to be protected by outer covers 214 and 215. As the vibration sensor material 211, polyvinylidene fluoride (PVDF), which is a fluorine-based organic thin film ferroelectric material, was used. Although signals may be extracted from the upper and lower electrodes, the negative electrode layer 213 can be configured to have a constant potential and a signal generated by the displacement of the vibration sensor material 211 can be extracted from the positive electrode layer 212. The outer covers 214 and 215 may be kept at a constant potential and may be used as a shielding layer in order to eliminate various external noises, particularly electromagnetic noise. Here, the shielding layer can be set to the same potential as the negative electrode layer 213, and in this case, the negative electrode layer and the protective cover can be combined with one member. Further, an insulating layer (insulating sheet) may be disposed between the positive electrode layer 212 or the negative electrode layer 213 and the outer covers 214 and 215 to insulate them. The positive electrode layer 212 and the negative electrode layer 213 are each assembled with an extraction terminal (not shown) together with a filter circuit and the like, and can apply a voltage to the electrode or output a signal from the electrode, respectively. The piezoelectric sensor 210 is a piezo element. When a mechanical force (a slight force) is applied, an electromotive force is generated in the vibration sensor material 211, and the electric charge accumulated in the vibration sensor material 211 is converted into a current voltage. It can be taken out as an electrical signal.

  In the present embodiment, the sheet-type piezoelectric sensor 210 was laid under the bed sheet, and the biological signal was separated and extracted in real time without any burden on the subject. In the case of measurement in a bed, a sheet-type piezoelectric sensor 210 was installed on a bed mattress, a sheet was placed on it, and a human was rested in a supine position. When a human sleeps on the bed, the motion of the human heart and breathing is transmitted to the sensor 210 as a vibration wave through the body and the body surface, and the sensor 210 generates an electromotive force on the order of μV. In addition to the desired biological information such as heartbeat and respiration, this signal also includes a noise signal that becomes an obstacle, so that subsequent biological information such as heartbeat and respiration can be obtained using a signal processing algorithm (electronic circuit and software). Separated and extracted. The test subject also attached an electrocardiogram electrode to the chest and simultaneously measured an electrocardiogram (ECG) by unipolar derivation. Furthermore, the subject wore a face mask, and simultaneously measured the respiration pattern with a hot-wire respirometer. In addition, the biological signal could be detected in a state where the seat type piezoelectric sensor 210 was laid under a thin cushion on the sitting position (under the buttocks) of the chair and the subject was sitting.

  The biological vibration signal, electrocardiogram, and respiratory flow rate waveform were sampled and stored at 100 Hz. The biological vibration signal was subjected to signal processing using specific digital filters for heartbeat detection and respiration detection. After applying a high-frequency bandpass filter (or a high-pass filter) to the biological vibration signal, full-wave rectification integration is performed to extract only the vibration component derived from the heartbeat, and the peak of this waveform is obtained. The derived vibration component waveform was differentiated, a threshold value was set, a peak (heartbeat pulse) was detected, and the pulsation interval was obtained from the peak interval. Here, the heartbeat interval obtained from the electrocardiogram is referred to as RRI, and the heartbeat interval obtained from the biological vibration signal is referred to as BBI. Moreover, a vibration component derived from respiration was extracted by applying a low-frequency band-pass filter (or a low-pass filter) to the biological vibration signal, and a respiration pattern was estimated. RRI and BBI were resampled at 10 Hz by spline interpolation. The respiration pattern actually measured by the hot-wire respiration velocimeter and the respiration pattern estimated from the biological vibration signal were also resampled at 10 Hz. Respiratory arrhythmia and respiratory pattern were subjected to Hilbert transform, the instantaneous phase was obtained from the analysis signal, and the phase coherence λ was calculated from the phase difference Ψ. The phase coherence was obtained while shifting by 5 seconds with a 10 second calculation window.

  FIG. 13A is an original signal of the biological vibration signal obtained from the sheet type piezoelectric sensor 210. FIG. 13B shows a waveform after signal processing from the original signal (after differential processing of the heartbeat-derived vibration component waveform after full-wave rectification integration). FIG. 13C shows the heartbeat pulse of the electrocardiogram (dotted line) on the upper side and the heartbeat pulse (solid line) obtained from the waveform after signal processing on the lower side. FIG. 13D shows an RRI (dotted line) obtained from an electrocardiogram and a BBI (solid line) obtained from a biological vibration signal. FIG. 13E shows an actually measured respiration pattern (dotted line) and a respiration pattern (solid line) estimated from a biological vibration signal. FIG. 13F shows the phase coherence (dotted line) calculated from the electrocardiogram and the actually measured respiration pattern, and the phase coherence (solid line) estimated from the biological vibration signal. As shown in FIG. 13D, the RRI obtained by simultaneous measurement from the electrocardiogram and the BBI obtained from the biological vibration signal are very similar, and the fluctuation of the heartbeat interval can be calculated from the biological vibration signal by signal processing. It was. Further, as shown in FIG. 13E, the actually measured breathing pattern and the breathing pattern estimated from the biological vibration signal almost coincided. Furthermore, as shown in FIG. 13 (F), the phase coherence calculated from the electrocardiogram and the actually measured respiration pattern almost coincided with the phase coherence calculated only from the biological vibration signal. From these facts, it was possible to calculate heartbeat interval fluctuation (respiratory arrhythmia), respiratory pattern and phase coherence only by measuring the biological vibration signal.

  FIG. 14 shows the power spectrum density (dotted line) of the electrocardiogram and the power spectrum density (solid line) of the waveform after the signal processing of the biological vibration signal shown in FIG. 13 (B). The heart rate frequency (about 1 Hz), which is the basic frequency of the electrocardiogram, matches the peak of the power spectrum of the waveform after signal processing, and the waveform after signal processing has been removed by signal processing except for the frequency component of the basic frequency of the electrocardiogram. I understand. Thus, in this method, after completely removing the respiratory frequency component in advance, the BBI corresponding to the RRI is obtained by obtaining the vibration component derived from the heartbeat, and as a result, the BBI becomes accurate. Some respiratory arrhythmias could also be detected, and the calculated phase coherence almost coincided with that obtained from the ECG and measured respiration.

The phase coherence calculation apparatus of the present invention can be used by being incorporated in various furniture, electronic devices, and the like. For example, a sensor for measuring vibration may be incorporated in a chair or bedding to measure the stress state of the user. In this case, it applies to seats on trains, airplanes, workplace seats, driver's seats on cars, trains, airplanes, etc., and is used for stress management and bedtime prevention, and it is applied to beds in hospitals and nursing homes. It can also be used to manage the health status of patients. In addition, sensors that measure vibration can be built into the floor of toilets, bathrooms, dressing rooms, etc., and can be used to monitor accidents such as fainting and stroke. Furthermore, a phase coherence calculation device can be incorporated in a portable terminal, a computer, etc., and psychological stress can be evaluated in various scenes of daily life.

DESCRIPTION OF SYMBOLS 1 Sleep state measuring apparatus 2 Information acquisition part 3 Information processing part 4 Operation part 5 Output part 6 Storage part

Claims (3)

Biological information acquisition means for acquiring biological information including at least an animal cardiogram or biological vibration signal;
Heart rate interval calculating means for calculating heart rate interval variation from the biological information;
The heartbeat interval calculating means passes a signal having a lower frequency than a heartbeat frequency with respect to the signal derived from the cardiogram waveform or biological vibration signal, and a signal after passing through the filter A biological information acquisition apparatus characterized in that processing for taking an absolute value is performed.   The biometric information acquisition apparatus according to claim 1, wherein a heartbeat interval is calculated by acquiring a peak of each heartbeat of a cardiocardiogram waveform from an envelope signal of the signal after performing the process of obtaining the absolute value. A signal processing method for calculating a heartbeat interval from an animal cardiogram waveform or a biological vibration signal,
A process of obtaining an absolute value of the signal after passing through the filter having a lower limit frequency that is higher than the heartbeat frequency with respect to the signal derived from the cardiogram waveform or the biological vibration signal. A signal processing method characterized by comprising: