JP7256936B2 - STRESS MEASUREMENT DEVICE, STRESS MEASUREMENT METHOD AND PROGRAM - Google Patents

Description

Detailed description of the invention

The present invention relates to an apparatus or method for measuring animal stress based on phase coherence and respiratory rate calculated from heartbeat waveforms and respiratory patterns extracted from animal biological information.
Furthermore, the present invention relates to an apparatus or method for measuring animal stress by extracting a human heartbeat waveform and respiratory waveform from a sheet-like piezoelectric sensor attached to a chair and calculating phase coherence and respiratory rate.

The number of patients with respiratory and circulatory diseases is increasing due to the stress caused by the sophistication of society and the declining birthrate and aging population. It's getting
Conventionally, stress assessment methods have mainly been conducted at medical institutions, including subjective assessment through interviews, physiological indicators such as blood pressure, heart rate, skin temperature, and perspiration, and measurements of humoral stress markers such as catecholamines and cortisol. It was not an indicator that an individual could measure.

As a technique for solving such problems, Patent Literature 1 discloses that phase coherence calculated from a heartbeat waveform and a breathing pattern can serve as an index of stress.
That is, from a sheet-type piezoelectric sensor placed under a bed sheet, biometric information including at least information about heartbeat and respiration of an animal is obtained, a breathing pattern is extracted from the biometric information, and a breathing pattern is extracted from the biometric information. It is disclosed that the beat-to-beat variability can be calculated and the phase coherence of the instantaneous phase difference between the breathing pattern and the beat-to-beat variability can be calculated.
The phase coherence takes a value between 0 and 1, approaches 1 as the instantaneous phase difference between heartbeat intervals and the respiratory pattern approaches a constant relationship, and approaches 0 as the instantaneous phase difference becomes random. It is disclosed that the phase coherence is close to 1 during quiet relaxation, and the phase coherence decreases when stress is applied, so that stress can be estimated by the phase coherence.

Re-table 2017/141976

Problems to be Solved by the Invention

However, although it has been shown that the phase coherence decreases in the stress state, it is not necessary that the phase coherence close to 0 indicates the stress state and the phase coherence close to 1 indicates the relaxed state.
For example, during a meal, the need to send blood to the gastrointestinal tract increases cardiac sympathetic nerve activity and increases the heart rate, which is a physiological response. As a result, there is a problem that the phase coherence (λ) is lowered even though no stress is applied, and it cannot necessarily be concluded that the device is under stress condition from the decrease in phase coherence (λ).
In addition, even if we try to measure the stress of a subject who works while sitting on a chair with a sheet-type piezoelectric sensor installed, the body movement signal is mixed with the biological information signal related to heartbeat and respiration, resulting in useful heartbeat information and There was a problem that respiration information could not be acquired and phase coherence could not be calculated.
In order to solve such problems, the present inventors have conducted intensive research, and as an index representing stress, the phase coherence (λ), which is an index that reflects autonomic nerve activity, is combined with the respiratory rate affected by the higher central nervous system. In addition, we devised a means to measure stress by combining phase coherence and respiratory rate.
In addition, from a biovibration signal that contains many body motion signals, such as the ballistocardiographic signal (BCG) obtained by installing a sheet-type piezoelectric sensor on a chair, there is a portion where heartbeat information and respiratory information are buried in the body motion signal. developed a means of selecting

Means to solve problems

In order to solve the above-mentioned problems, a stress measuring device according to the present invention comprises: an instantaneous phase of a heartbeat interval variation derived from respiration obtained from an animal; phase coherence calculating means for calculating phase coherence of the instantaneous phase difference based on the instantaneous phase difference between the instantaneous phase of the pattern; respiratory rate calculating means for calculating the respiratory rate from the respiratory pattern; and the phase coherence and the respiratory a stress index generating means for generating a stress index from the number.
The stress index may be two-dimensional data consisting of the phase coherence and the respiration rate, or a numerical value obtained by adding predetermined weights to the phase coherence and the respiration rate.
Further, means for selecting a portion where the heartbeat information or the respiratory information is buried in the body motion signal from the biological vibration signal including the heartbeat information, the respiratory information, and the body motion signal, and interpolating the heart rate and the respiratory rate of the portion. You may prepare.

Effect of the invention

A state in which the phase coherence is close to 0 is a stress state or a depressed state, and a state in which the phase coherence is close to 1 is an active state or a relaxed state. can be measured to
Moreover, even when the subject is working while sitting on a chair during working hours, the stress can be continuously measured without restraining the subject.
According to the present invention, the subject can know his or her own stress state, namely which state of stress, depress, active or relax, just by looking at the display points on the λ-RR plane.
If the phase coherence (λ) is low, it can be used as a basis for determining whether the work is hard and stressful or unmotivated and depressed.
Also, when the phase coherence (λ) is high, it can be determined whether the body and mind are in a relaxed state, or whether they are in an active state after moderate exercise.
Moreover, if expressed numerically, the subject can comprehend the course of his/her own stress state as a stress index graph.
By arranging this with other biometric information in the same time series, mutual relationships can be known.
Furthermore, even when there are many body motion signals such as a sitting BCG signal and information related to heartbeat and respiration is often buried in the body motion signals, effective heartbeat information and respiration information can be extracted to measure stress. becomes possible.
For example, it is possible to continuously grasp the rhythm of an individual's stress during working hours (8 hours) without restricting the individual. can contribute to providing

Schematic block diagram of the stress measurement device An embodiment of a stress measuring device A case of randomly assigning a mental arithmetic task λ-RR distribution in the case of randomly assigned mental arithmetic tasks A case of imposing a 50-question correct mental arithmetic task λ-RR distribution in a case of imposing a 50-question correct mental arithmetic task λ-RR distribution (by state) in case of imposing 50 correct answer mental arithmetic tasks λ-RR distribution in stress measurement during exercise (subject A) λ-RR distribution in stress measurement during exercise (subject B) Weighting when the λ-RR plane is divided into 16 A case in which the ACR of the ballistocardiographic signal is clear A case where the ACR of the ballistocardiographic signal is unclear λ calculated from biological vibration signals in a sitting position, Stress index

A stress measurement device 1 using phase coherence includes at least an information acquisition unit 2 and an information processing unit 3, as shown in FIG. Furthermore, the stress constant device 1 may include an operation section 4 , an output section 5 and a storage section 6 .

The information acquisition unit 2 acquires information necessary for calculating phase coherence, and may include a sensor for measuring an animal and an input unit for inputting sensor information by wire or wirelessly. Alternatively, the configuration may include an input unit capable of inputting information from another recording medium on 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 in some cases may include a sensor for measuring biological information connected to the input unit by wire or wirelessly. When measuring biological information with a sensor, the sampling frequency is preferably 100 Hz or higher.

The information processing section 3 processes input information, and can use, for example, the arithmetic processing function of a CPU (Central Processing Unit) of a computer. Further, some information processing can be realized by analog circuits instead of digital circuits.
For example, as information processing, the frequency filter may be realized by analog filters such as a low-pass filter (LPF) or a high-pass filter (HPF) composed of capacitors, resistors, operational amplifiers, etc., or filtering is performed by the arithmetic processing function of the CPU. It may be realized by a digital filter.
The information processing section 3 may include both a digital circuit and an analog circuit depending on the type of information processing. may
The information processing unit 3 has different functions or processes depending on the input information. It has a calculation function.
Further, the information processing section 3 has a stress index generation function for generating a stress index based on the calculated phase coherence and respiration rate.
The stress index is time-series two-dimensional data of λ-RR and time-series numerical data (stress index) obtained by weighting and quantifying the two-dimensional data.

The operation unit 4 is provided with operation terminals such as switches, a touch panel, buttons, knobs, a keyboard, a mouse, and a voice input microphone for the user to operate the stress measurement device 1 .
Further, 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 or stress index, or may output biological information other than the phase coherence.
For example, a display that displays the results as an image, a printer that outputs the results on paper, a speaker that outputs the results by voice, a wired or wireless output terminal that outputs the results as electronic information, and the like can be used. The display as the output unit 5 may also be used as a touch panel in the operation unit 4 or a display for displaying operation contents.
The storage unit 6 can store the information acquired by the information acquisition unit 2, the phase coherence calculated by the information processing unit 3, the stress index, and the like.

FIG. 2 shows an example of the stress measuring device 1. As shown in FIG. The stress measurement device 1 includes a sensor 21, an analog-digital conversion circuit 31, an effective biological vibration signal selection means 32, a heartbeat extraction means 33, a heartbeat interval calculation means 34, a respiratory waveform extraction means 36, Hilbert transform filters 35 and 37, an instantaneous position Phase difference calculation means 39, phase coherence calculation means 40, respiratory rate calculation means 38, stress index generation means 41, operation buttons 51, touch panel 52, voice input microphone 53, display 54, wireless communication means 55, speaker 56, recording device 57.

The sensor 21 detects biological information including information about heartbeat and biological information including information about respiration.
For example, there are electrocardiogram sensors, phonocardiogram sensors, pulse wave sensors, and sensors that measure vibrations associated with heartbeats that detect biological information including information about heartbeats. Detectors include a sensor for measuring respiratory flow velocity, a sensor for measuring the heat flow of exhaled air, a sensor for measuring vibration of the thorax associated with respiratory motion, and the like.

In addition, the sensor for measuring vibration may be of a contact type or a non-contact type. In the case of a sensor that measures contact type vibration, by placing it in direct or indirect contact with the animal, the ballistocardiogram waveform can be obtained. Alternatively, a biological vibration signal can be detected. Contact-type sensors for detecting ballistocardiogram waveforms or biological vibration signals are placed directly on or in the vicinity of various living organisms that generate vibrations, and detect vibrations from the living organisms and output them as electrical signals.
As a sensor for measuring vibration, a piezo element is preferably used as a piezoelectric sensor, but other sensors such as polymer piezoelectric material (polyolefin material) may be used.
The piezoelectric sensor is preferably film-shaped.
Furthermore, in the case of a piezoelectric sensor, it is possible to obtain a ballistocardiogram waveform or a biological vibration signal without restraining the animal, and stress-free measurement is possible, which is preferable.
However, the piezoelectric sensor can also be attached to a wristband, belt, wristwatch, ring, headband, or the like, and attached to an animal to be used as a wearable sensor.
In addition, as a sensor for measuring other types of vibration, for example, using a high-sensitivity acceleration sensor, the acceleration sensor is placed in contact with the body such as a wristwatch or a mobile terminal, or a part of a bed, chair, etc. A ballistocardiogram waveform or biological vibration signal may be obtained by detecting a change in air pressure or liquid pressure in the tube with a pressure sensor or the like to obtain a ballistocardiogram waveform or biological vibration signal. good too.
Furthermore, as a sensor for measuring vibration, a non-contact sensor that can acquire a ballistocardiogram waveform or a biological vibration signal by receiving and transmitting signals using microwaves or the like may be used.

A signal detected by the sensor 21 is input to the analog-digital conversion circuit 31 of the stress measurement device 1 by wire or wirelessly.
The analog-digital conversion circuit 31 is a circuit that converts an analog signal from the sensor 21 into a digital signal. The analog-to-digital conversion circuit 31 may be provided within the sensor 21, or may be omitted if the sensor 21 can detect digital signals.
Further, the analog signal from the sensor 21 may be converted into a digital signal by the analog-digital conversion circuit 31 after being subjected to processing such as filtering.

The output of the analog-digital conversion circuit 31 is input to the effective biological vibration signal selection means 32 .
The effective biological vibration signal selection means 32 selects, for example, a part from which heartbeat information and breathing information can be extracted as a valid biological vibration signal, and a part where heartbeat information and breathing information are buried in a body motion signal (BM signal). Select the bio-vibration signals that are not Furthermore, a function of complementing heartbeat information and respiration information corresponding to ineffective bio-vibration signals may be provided.
For example, if there is body motion within a time window of 10 seconds (a biovibration signal of 3 V or higher is considered to be body motion) or if there is no signal, the window data is replaced with the complementary signal.
In addition, the ballistocardiographic signal (BCG signal) extracted from the biological vibration signal mixed with the body motion signal is passed through a band-pass filter and then subjected to full-wave rectification integration to calculate the autocorrelation function (ACR). , select those parts of the autocorrelation function (ACR) with clear oscillations as valid ballistocardiographic signals (BCG signals).
A portion where the oscillation of the autocorrelation function (ACR) is clear means a portion where there are at least 5 or more peaks of the autocorrelation function (ACR) in a window of 10 seconds. 5 or more is based on a heartbeat period of no more than 2 seconds.
Alternatively, it refers to a portion where the vibration amplitude of the autocorrelation function (ACR) is equal to or greater than the threshold value (0.2 to 0.5), but these requirements may be appropriately changed according to the state of the biological vibration signal.
That is, a ballistocardiographic signal (BCG signal) is passed through a 5-15 Hz band-pass filter, the absolute value is taken, and an envelope curve of the heartbeat component obtained through a 0.1-3.0 Hz band-pass filter is obtained, Its autocorrelation function (ACR) is calculated, and the portions whose vibrations satisfy the above requirements are selected as valid biological vibration signals. It is also possible to perform processing such as complementation for the ineffective portion of the biological vibration signal.
By this means, even when a large body motion signal is mixed in the bio-vibration signal obtained by the sheet-like piezoelectric sensor installed on the chair, the effective portion of the bio-vibration signal is selected, and the heartbeat interval (BBI) is determined. It is possible to do the calculations and determine the phase coherence (λ).

The heartbeat extracting means 33 is a means for extracting a heartbeat-related signal from the signals detected by the sensor 21, and suitable processing is appropriately selected according to the type of sensor or the input signal.
When an electrocardiogram waveform or a ballistocardiogram waveform is input, it is usually affected by respiration. If there is no problem in calculating the intervals, the heart rate extracting means 33 may not be used.
When a bio-vibration signal is input, the bio-vibration signal usually includes not only the ballistic motion caused by the heartbeat, but also the vibration caused by breathing, body movement, vocalization, vibration based on the external environment, and the like. In some cases, it is preferable to perform processing to remove these noises.
As such processing, for example, after emphasizing the strength of the electrocardiogram waveform or vibration signal to the nth power (n is an integer of 2 or more, and when n is an odd number, the absolute value is taken), a bandpass filter (BPF ) may be passed through.
The BPF of the heartbeat extracting means 33 preferably has a lower limit frequency 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. Below, it is preferably 8 Hz or less, 6 Hz or less, 5 Hz or less, or 3 Hz or less, and preferably has a passband combining any of these lower limit frequencies and any of the upper limit frequencies.

Furthermore, as a method of calculating the heartbeat interval from the ballistocardiogram waveform, the ballistocardiogram waveform is passed through a high-pass filter (HPF) whose lower limit frequency is higher than the frequency of the heartbeat, and the absolute value of the post-HPF signal is By taking the envelope signal of the signal that has passed through the HPF used, each heartbeat peak of the ballistocardiogram waveform can be obtained. can ask.
The normal heartbeat frequency is about 3 Hz at maximum, but the lower limit frequency of this high-pass filter is preferably 5 Hz or higher, and may be 10 Hz, 20 Hz, 30 Hz, or 40 Hz.

The respiratory waveform extracting means 36 is means for extracting a signal relating to a respiratory pattern from the signals detected by the sensor 21, and appropriate processing is selected according to the type of sensor or the input signal.
When a respiratory sensor is used as part of the sensor 21 and the respiratory pattern is actually measured by the respiratory sensor, the respiratory waveform extracting means 36 may not be provided, and the respiratory waveform extracting means 36 may remove signals that become noise. You may perform processing to do.
When extracting a signal related to a respiratory pattern from an electrocardiogram waveform, a ballistocardiogram waveform, or a biological vibration signal, such processing includes, for example, increasing the strength of the electrocardiogram waveform or vibration signal to the power of n (where n is an integer of 2 or more). , the absolute value is taken if n is an odd number), and then passed through a low-pass filter (LPF) having a passband in the frequency range of 0.5 Hz or less.
The cutoff frequency of the LPF of the respiratory waveform extraction means 36 may be 0.3 Hz, 0.4 Hz, 0.6 Hz, 0.7 Hz and 0.8 Hz.
Also, instead of the LPF, the BPF may be passed through. In this case, the lower limit frequency of the BPF may be set to a sufficiently low frequency, for example, 0.1 Hz.

The heartbeat interval calculation means 34 receives the signal from the heartbeat extraction means 33 and calculates the heartbeat interval. The heartbeat interval is preferably measured, for example, between P, R, T, or U waves in an electrocardiogram, especially the interval from one R wave to the next R wave, since the R wave has a sharp peak.
Also in the case of a heartbeat-related signal extracted from a ballistocardiogram waveform or a biological vibration signal, it is preferable to measure the interval of the waveform corresponding to the sharp peak of the R wave.

The Hilbert transform filter 35 outputs the instantaneous phase and the instantaneous amplitude of the heartbeat interval variation, and the Hilbert transform filter 37 outputs the instantaneous phase and the instantaneous amplitude of the breathing pattern.
For the Hilbert transform, a 90-degree phase difference wave generator may be realized by an analog circuit, or a finite impulse response type digital filter may be used.
An analytic signal is obtained by adding the Hilbert-transformed signal and the real signal, and the instantaneous phase can be obtained from the ratio between the real part and the imaginary part of the analytic signal.

The instantaneous phase difference calculating means 39 calculates the instantaneous phase difference between the instantaneous phase of the heartbeat interval variation and the instantaneous phase of the breathing pattern, and outputs the result to the phase coherence calculating means 40 .
The phase coherence calculating means 40 calculates the phase coherence using the instantaneous phase difference between the instantaneous phase of the heartbeat interval variation and the instantaneous phase of the breathing pattern.
The data for obtaining the phase coherence is calculated with a window length of at least one respiratory cycle.

Respiration rate calculation means 38 receives the respiration pattern output from respiration waveform extraction means 36, obtains its power spectrum, and calculates the respiration rate as the reciprocal of the peak frequency of the power spectrum.

計算課題の遂行は認識努力と集中力を要し、切迫感などにより高位中枢が刺激される。高位中枢からの下行神経回路により呼吸中枢にある呼吸リズム発生器が影響を受け、呼吸数が増加するものと推察される。また、鬱病患者では安静時の位相コヒーレンス（λ）は健常者に比べて低く、呼吸数も低下する傾向があることが報告されている。位相コヒーレンス（λ）と呼吸数を用いて二次元で表現すると、例えば、図６に示すように、位相コヒーレンス（λ）が低い領域でも呼吸数が高いｓｔｒｅｓｓ状態と、呼吸数が低いｄｅｐｒｅｓｓ状態があることが見えてくる。The stress index generating means 41 will be described.
The phase coherence (λ) ranges from 0 to 1. The closer the value is to 0, the more stressful the cardiac sympathetic nerve activity is, and the closer to 1 is, the more relaxed the parasympathetic nerve activity is. However, it cannot be said that the closer the phase coherence (λ) is to 0, the higher the degree of stress. This is because the enhancement of cardiac sympathetic nerve activity also occurs during exercise and eating behavior, and the phase coherence (λ) does not specifically reflect the degree of stress. However, we found that the respiration rate increased with a decrease in the phase coherence (λ) in most cases when psychological stress was imposed on computational tasks with a limited answer time.
Therefore, in addition to the phase coherence (λ), we added the respiration rate and considered stress evaluation in a two-dimensional display.
[Adoption of respiratory rate]
Table 1 shows changes in cardiopulmonary variables when 19 subjects were subjected to a computational task. Increases in blood pressure, heart rate, and respiratory rate were observed during the computational task. Among these variables, the respiratory rate has a small standard error, and it can be judged that there are few individual differences as an index. Therefore, respiratory rate was adopted as a second index for measuring stress.

Performing computational tasks requires cognitive effort and concentration, and a sense of urgency stimulates the higher brain. It is presumed that the respiratory rhythm generator in the respiratory center is affected by the descending neural circuit from the higher center, and the respiratory rate increases. It has also been reported that depressed patients have a lower resting phase coherence (λ) than healthy subjects, and their respiratory rate tends to decrease. When expressed in two dimensions using phase coherence (λ) and respiratory rate, for example, as shown in FIG. I can see something.

[λ-RR data generation]
From the phase coherence (λ) calculated by the phase coherence calculation means 40 and the respiration rate calculated by the respiration rate calculation means 38 at the same time as the phase coherence (λ), the two-dimensional stress of λ-RR Generate indicators as time series data.
Then, this stress index is displayed on a two-dimensional plane (λ-RR plane) in which the horizontal axis is the phase coherence and the vertical axis is the respiration rate.
For example, in the λ-RR plane shown in FIG. 4, the λ axis is 0 to 1, the RR axis is 10 to 30 (times/minute), and the λ axis is incremented by 0.2 and the RR axis is incremented by 5.
In FIG. 4, the stress is displayed on the two-dimensional λ-RR plane, which can be interpreted in various ways.
The left half with a low λ indicates a state in which the autonomic nerve (sympathetic nerve/parasympathetic nerve) activity is dominant, and the right half with a high λ indicates a state in which the parasympathetic nerve activity is dominant. It is assumed that the lower right is the relaxed state, the upper right half is the active state, the upper left is the stress state, and the lower left is the depress state.
For example, the active state is a state in which the respiration rate is high during light exercise, but autonomic nerve activity is dominated by the parasympathetic nerves. not).
By expressing the stress on the λ-RR plane in this way, it is possible to easily know where the subject's stress is located on the λ-RR plane. For example, if the subject's stress is in the stress state of the λ-RR plane, he/she will be able to take a rest, and if he/she is in the depress state, he/she will be able to take measures such as switching to being motivated.

[Time series digitization of λ-RR data]
For example, as shown in FIG. 10, the λ-RR plane is divided into 16, the λ-axis direction is weighted in 2-point increments, and the RR-axis direction is weighted in 1-point increments. Numericalize each region of the plane. This quantified stress index is called a stress index, but the numerical value itself does not have a physiological basis. At rest, the phase coherence (λ) is high and the RR is small, and under stress, the phase coherence (λ) is low and the RR is large, which is weighted so that the numerical value increases, and the degree of stress is quantified. is.
For example, like the stress index at the bottom of FIG. 3, it is possible to express how the subject's stress changes over time as a time-series numerical graph. Stress is evaluated on a 10-grade scale from 0 to 9. For example, 0 to 3: relaxed state, 4 to 6: mild stress state, 7 to 9: stress state.
In this way, when stress is expressed as time-series data, by arranging it with other biological information in the same time series, it is possible to know the mutual relationship and to know the stress in detail.
In addition, by graphing the stress index for working hours (e.g. 8 hours), it is possible to grasp the rhythm of the subject's stress state. It becomes possible to contribute to the mental health of people.

The operation button 51, the touch panel 52, and the voice input microphone 53 are input means for the user to operate the stress measurement device 1, and can operate the stress measurement device 1 and output necessary information. can.
The display 54 and the speaker 56 can be used as output means for outputting heartbeat, respiratory pattern, respiratory arrhythmia, phase coherence (λ), stress index generated from phase coherence (λ) and respiration rate, and the like. .
The wireless communication means 55 may be used as an output means such as the calculated phase coherence (λ) or a stress index generated from the phase coherence (λ) and the respiration rate, or as an input means for inputting a signal from the sensor 21. may be used as Alternatively, the stress index or the like may be output by voice.
The recording device 57 records input information, programs of various means, measurement results of phase coherence, stress index, and the like.

[Measurement of stress when imposing a mental arithmetic test]
FIG. 3 shows an example of a calculation test in which subjects were randomly given mental arithmetic tasks.
A sheet-shaped piezoelectric sensor installed on the seat surface of a chair acquires a biological vibration signal and processes it to obtain a heart rate (HR), a respiration rate (RR), a phase coherence (λ), and a stress index. .
The top row in FIG. 3 is the heart rate (HR), the second row is the respiratory rate (RR), and the third row is the phase coherence (λ). The convex portions of the rectangular signals of heart rate (HR), respiration rate (RR), and phase coherence (λ) are time periods in which mental arithmetic tasks were randomly imposed, and are periods of psychological stress.
The bottom row is the stress index, and it can be seen that the phase coherence (λ) approaches 0 and the stress index increases during the mental arithmetic task.
FIG. 4 shows the results obtained by collecting the results of nine subjects and plotting them on the λ-RR plane. Even in areas where λ is low, there are areas where the respiration rate is high and areas where the respiration rate is low. Areas with a low respiratory rate are assumed to be in a depressed state, and areas with a high respiratory rate are assumed to be in a stressed state due to random mental arithmetic tasks.

[Measurement of stress when imposing a 50-question correct mental arithmetic test]
FIG. 5 shows an example of a calculation test in which subjects are continuously given mental arithmetic tasks until they answer 50 questions correctly. Acquisition of the biological vibration signal and its processing method are the same as in the case of FIG. 3 in which the mental arithmetic task is randomly imposed.
The convex portions of the rectangular signals appearing in rows 1 to 3 of FIG. It can be seen that the phase coherence (λ) is low and the stress index is high at .
FIG. 6 shows the results of summarizing the results of seven subjects and plotting them on the λ-RR plane. Even in areas where λ is low, there are areas where the respiration rate is high and areas where it is low. Compared with FIG. 4, the variation in respiration rate is large when λ is small.
Furthermore, when the data of the 50-question correct mental arithmetic test is displayed according to the status, it becomes as shown in FIG. ○ (Rest) is the state of resting, × (Arithmetic) is the state of doing a mental arithmetic test, and □ (Recovery) is the state of finishing the mental arithmetic test and going to rest.
Many areas with low respiratory rate were seen before the computational task was performed, and it is presumed that this reflected the subject's depressive feelings about the computational task.
In addition, it can be read that there are people who shift from (λ high, RR low) → (λ low, RR high) and people who shift from (λ low, RR low) → (λ low, RR high). .

[Stress measurement during exercise]
Breathing has both autonomic and voluntary characteristics, and during exercise, breathing is performed in accordance with the movement rhythm, and the breathing rate may increase. The increased respiratory rate is not an involuntary reflex response, but the result of voluntary breathing control. Therefore, even if the phase coherence (λ) is close to 1, it cannot necessarily be concluded that the state is in a resting relaxed state.
FIG. 8 shows the λ-RR distribution based on the stress measurement results during exercise for subject A, and FIG. 9 shows the λ-RR distribution based on the stress measurement results for subject B during exercise. In both cases, heartbeat intervals were measured from an electrocardiogram obtained by chest bipolar leads, and respiratory waveforms were measured by the subject wearing a face mask connected to a pneumotachograph, and processed in the same manner as in the mental arithmetic test. The exercise protocol consisted of a resting state in which the subject sat on a stationary bicycle for 3 minutes, and then gradually increased the load by 20 W every minute, and pedal rotation was 50-60. Rotation/minute.
FIG. 8 shows the λ-RR distribution of subject A. FIG. ◯ is the rest state, ▪ is the state of applying a load of 20-140 W (light to moderate), and × is the state of applying a load of 140-240 W (strong exercise). It can be seen that λ does not change from rest to light to moderate exercise load and only RR increases, while λ decreases and RR increases under strong exercise load.
FIG. 9 is the λ-RR distribution of Subject B. FIG. It can be seen that both λ and RR do not change much under light to moderate exercise load from the resting state, and that λ decreases and RR increases under strong exercise load.
In general, the physical stress caused by exercise load depends on the physical ability of the subject, that is, the aerobic metabolic capacity, and under a strong exercise load that exceeds the aerobic metabolic capacity, λ decreases and RR increases in both subjects A and B, resulting in a stress state. It can be seen that
Also, during exercise, oxygen uptake in the lungs must be increased to meet the oxygen demand of active muscles, increasing minute ventilation. Minute ventilation is expressed as the product of minute respiratory rate (RR) and tidal volume. It can be seen that the minute ventilation is increased by the increase in volume. Whether the RR is increased or the tidal volume is increased during exercise depends on the adaptation of respiration to the exercise of the individual subject. It is inferred that the parasympathetic nerve activity is still dominant in the exercise load of .
At rest, both the λ-RR distribution in FIG. 8 and the λ-RR distribution in FIG. It can be said that it is in a relaxed state.

[Example of selection of effective part of biological vibration signal]
FIG. 11 shows an example in which the heartbeat interval (BBI) can be calculated correctly because the oscillation amplitude of the autocorrelation function (ACR) is large. The lower side of the first row is the ballistocardiographic signal (BCG signal) extracted from the biological vibration signal acquired by the piezoelectric sensor, and the upper side of the first row is its 5 to 15 Hz component. The second row shows the signal obtained by full-wave rectifying and integrating the 5 to 15 Hz component in the first row (the signal with the peak circled) and its instantaneous phase. The autocorrelation function (ACR) of the full wave rectified and integrated signal is well defined as shown in the third row. In such cases, the beat-to-beat interval (BBI) is calculated without problems.
On the other hand, FIG. 12 shows an example in which the heartbeat interval (BBI) cannot be calculated correctly because the vibration amplitude of the autocorrelation function (ACR) is small because the body motion signal is mixed in the biological vibration signal. The lower side of the first row is the ballistocardiographic signal (BCG signal) extracted from the biological vibration signal acquired by the piezoelectric sensor, and the upper side is its 5 to 15 Hz component, but the periodicity is not clear. In such a case, the periodicity of the signal obtained by full-wave rectification and integration (signal with a circle at the peak) and its instantaneous phase shown in the second row is also disturbed, and the autocorrelation function (ACR) shown in the third row is The vibration amplitude also becomes smaller.
The vibration amplitude of the autocorrelation function (ACR) becomes smaller not only when the body motion signal is mixed with the biovibration signal, but also when the heartbeat component is unclear. It can be attributed to both obvious things.

[Measurement example based on sitting bio-vibration signals]
Even if there are portions where body motion signals are mixed and heartbeat components are unclear, the effective biological vibration signal selection means 32 selects the effective portions, obtains the phase coherence (λ), and obtains the stress index. An example is shown in FIG.
A subject sat on a chair with a sheet-like piezoelectric sensor installed on the seat surface, and acquired a biovibration signal while performing normal office work.
When doing office work while sitting on a chair, body motion signals are mixed in with bio-vibration signals, even if the person turns around a little, and information about heartbeat and respiration is buried in the body motion signals. Therefore, the autocorrelation function (ACR) of the ballistocardiographic signal extracted from the biological vibration signal is calculated with a time window of 10 seconds, and the average value of the vibration amplitude of the ACR is 0.4 or more while shifting the time window by 5 seconds. part was selected. A little before 10:30 in FIG. 13, there is a part where the data is complemented (a part where the signal changes little), but this part is a part where the vibration amplitude of the autocorrelation function (ACR) is small, and the measurement data is This is the part that was not selected.

The first row in FIG. 13 shows the heart rate (HR), which is calculated by the heartbeat interval calculator 34 based on the biological vibration signal acquired by the sensor 21 .
The second row shows the respiration rate (RR), which is calculated by the respiration rate calculator 38 based on the biological vibration signal acquired by the sensor 21 .
The third level is the phase coherence (λ). Calculating an instantaneous phase difference based on the instantaneous phase of the heartbeat interval fluctuation calculated by the heartbeat interval calculating means 34 and the instantaneous phase of the breathing pattern in the same time series as the heartbeat interval fluctuation derived from breathing extracted by the respiratory waveform extracting means 36 A means 39 calculates the instantaneous phase difference. The phase coherence calculator 40 calculates the phase coherence based on the instantaneous phase difference.
The fourth row shows the stress index, which is calculated by the stress index generating means 41 based on the phase coherence and the respiration rate.
As described above, according to the present invention, the stress index can be continuously measured without disturbing the subject's usual activities even when the subject is working while sitting on a chair. .

1 stress measurement device 2 information acquisition unit 3 information processing unit 4 operation unit 5 output unit 6 storage unit 21 sensor 31 analog-digital conversion circuit 32 effective biological vibration signal selection means 33 heartbeat extraction means 34 heartbeat interval calculation means 35, 37 Hilbert transform Filter 36 Respiratory waveform extracting means 38 Respiratory rate calculating means 39 Instantaneous phase difference calculating means 40 Phase coherence calculating means 41 Stress index generating means 51 Operation button 52 Touch panel 53 Voice input microphone 54 Display display 55 Wireless communication means 56 Speaker 57 Recording device

Claims (6)

An instantaneous phase difference based on an instantaneous phase difference between an instantaneous phase of a respiration-derived beat-to-beat interval variation obtained from an animal and an instantaneous phase of the animal's breathing pattern in the same time series as the respiration-derived beat-to-beat variation. a phase coherence calculating means for calculating the phase coherence of the respiratory rate, a respiratory rate calculating means for calculating a respiratory rate from the respiratory pattern, and a stress index generating means for generating a stress index from the phase coherence and the respiratory rate. An animal stress measuring device characterized by: 2. The animal stress measuring apparatus according to claim 1, wherein said stress index is two-dimensional data consisting of said phase coherence and said respiration rate. 2. The animal stress measuring apparatus according to claim 1, wherein the stress index is a numerical value obtained by giving a predetermined weight to each region obtained by dividing a two-dimensional plane composed of the phase coherence and the respiration rate. A means for selecting a portion in which the heartbeat information or the respiratory information is buried in the body motion signal from the biological vibration signal including the heartbeat information, the respiratory information, and the body motion signal, and complementing the heartbeat information or the respiratory information in that portion. 4. The animal stress measuring device according to any one of claims 1 to 3, comprising: An instantaneous phase difference based on an instantaneous phase difference between an instantaneous phase of a respiration-derived beat-to-beat interval variation obtained from an animal and an instantaneous phase of the animal's breathing pattern in the same time series as the respiration-derived beat-to-beat variation. calculating a phase coherence of said breathing pattern; calculating a breathing rate from said breathing pattern; and generating a stress index from said phase coherence and said breathing rate. An instantaneous phase difference based on an instantaneous phase difference between an instantaneous phase of a respiration-derived beat-to-beat interval variation obtained from an animal and an instantaneous phase of the animal's breathing pattern in the same time series as the respiration-derived beat-to-beat variation. calculating a phase coherence of , calculating a respiratory rate from the respiratory pattern, and generating a stress index from the phase coherence and the respiratory rate.

Images