JP2001204714A - Mental stress judging device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a mental stress judging device capable of reducing errors based on the sudden change of a heart rate. SOLUTION: This device is provided with a means for obtaining the heartbeat equivalent signal and the respiratory vibration equivalent signal of an examinee, a first conversion means for performing time/frequency conversion to time-axial heartbeat interval data of the heartbeat equivalent signal, a data specifying means for specifying only data in a frequency band equal to or lower than a prescribed frequency among the pieces of data converted by this first conversion means, a second conversion means for converting frequency/time to data specified by this data specifying means, a movement averaging means for performing movement averaging processing to the time-axial heartbeat interval data converted by this second conversion means based on the respiratory vibration equivalent signal, and a stress analytic means for analyzing stress based on the heartbeat interval data movement-averaged by this movement averaging means.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a mental stress determination device.

[0002]

2. Description of the Related Art In recent years, people have been subjected to various stresses even in a workplace or a place of daily life.

[0003] With the rapid development of office automation and factory automation in particular,
VDT (Vid)
eoDisplayTerminal (video display terminal) work, monitoring work, and mental stress caused by driving a vehicle cause mental overwork and human error. Therefore, to prevent this,
It is important to evaluate mental stress.

As a conventional mental stress determination device,
In some cases, mental stress is determined by monitoring the rise and fall of pulse data calculated from pulse wave data.

[0005]

However, in the conventional mental stress determination apparatus, the mental stress is determined by monitoring the rise and fall of the pulse data calculated from the electroencephalogram data. The error was caused by sudden fluctuations in heart rate caused by fluctuations in the noise of the subject, deep breathing such as lack of extension of the subject, respiratory vibrations such as re-sitting, and body vibrations, and the accuracy of the determination results was reduced. .

In order to avoid this, it is conceivable to secure an experimental environment in which no noise or the like is generated and to concentrate on the experiment for the subject. However, securing the former experimental environment requires a great deal of labor, and the latter requires a great deal of labor. Concentration on the experiment has a problem that the evaluation of the mental stress itself may not be performed accurately.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned conventional problems, and has as its object to provide a mental stress determination apparatus capable of reducing an error based on a sudden change in heart rate.

[0008]

In order to achieve the above object,
The present invention obtains a subject's heartbeat equivalent signal and respiratory vibration equivalent signal when a load is applied to the subject,
First conversion means for performing time-frequency conversion on beat interval data on the time axis of the heartbeat-equivalent signal; and data of a frequency band equal to or lower than a predetermined frequency among data converted by the first conversion means. Data specifying means for specifying only the data, second converting means for performing frequency / time conversion on the data specified by the data specifying means, and a beat interval on the time axis converted by the second converting means. Data, moving average means for performing moving average processing based on the respiratory vibration equivalent signal, and stress analysis means for performing stress analysis based on beat interval data moving averaged by the moving average means, It is characterized by having.

According to a second aspect of the present invention, there is provided the mental stress determining apparatus according to the first aspect, further comprising a judging means for judging a magnitude of the signal corresponding to the respiratory vibration, wherein the moving average means is provided by the judging means. A moving average section of the moving average process is set based on the determined size.

[0010]

According to the present invention, it is possible to provide an apparatus for determining mental stress which can reduce an error based on a sudden change in heart rate.

[0011]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram showing a schematic configuration when a mental stress determination device is applied to a vehicle, FIGS. 2 to 3 are diagrams for explaining a method for acquiring a heartbeat-equivalent signal, and FIG. 4 is a diagram for explaining a method for acquiring a respiratory vibration equivalent signal. FIG. 5 is a diagram illustrating the configuration of the mental stress determination device, FIG. 6 is a diagram illustrating the relationship between the distribution and the stress, FIG. 7 is a diagram illustrating the effect of disturbance, and FIGS. FIG. 4 is an explanatory diagram of an analysis and determination method.

In FIG. 1, reference numeral 1 denotes a vehicle to which a mental stress determination device according to the present invention is applied. Reference numeral 5 denotes a seat provided on the vehicle, and only the driver's seat is shown in the figure. Reference numeral 10 denotes a steering wheel for steering the vehicle, and reference numeral 20 denotes a seat belt for restraining an occupant.
Reference numeral 40 denotes a vehicle speed sensor for detecting the traveling speed of the vehicle 1. 30 is an analysis / determination of mental stress / analysis /
It is a determination device.

As shown in FIG. 2 (a), the steering 10 is provided with a plurality of electrodes 15. The driver contacts the electrode 15 when the driver performs a steering operation during traveling or the like. Thus, a heartbeat-equivalent signal can be obtained.

FIG. 2B is an example of a heartbeat-equivalent signal obtained by the electrode 15 provided on the steering wheel 10.

In this embodiment, the steering 10
Although the heartbeat-equivalent signal is obtained by the electrode 15 provided in the wristband, as shown in FIG. 3A, the driver wraps the wristband 50 provided with the electrode 55 around the wrist, and A heartbeat-equivalent signal may be obtained from a potential signal.

The seat belt 20 is provided with a plurality of piezoelectric sensors 25 as shown in FIG. The piezoelectric sensor 25 is provided on the back surface of the seat belt 20, and is provided so as to directly contact a driver. With the piezoelectric sensor 25, a signal corresponding to respiratory vibration can be acquired from the magnitude and timing of respiration. FIG. 4B is an example of a signal equivalent to respiratory vibration acquired by the electrode 25 provided on the seat belt 20.

Next, the configuration of the mental stress determination device of the present invention will be described with reference to FIG. 5. The vehicle feed signal of the vehicle 1 obtained by the vehicle speed sensor 40, the potential signal of the electrode 15 provided on the steering 10, the seat belt A signal from the piezoelectric sensor 25 provided in the device 20 is input to the analysis / determination device 30, and the mental stress of the driver is analyzed / determined by performing processing described below.

Next, how to give a task and how to give a rest will be described. Since the driving driver concentrates on driving, the task is given. On the other hand, during a stop or the like, a state in which no task is given, that is, a relaxed state, can be said to be a rest. In the embodiment of the present invention, based on this concept, it is determined from the vehicle transmission signal from the vehicle speed sensor 40 whether it is a task or a rest at present.

Next, the relationship between distribution and stress will be described with reference to FIG. FIG. 6A shows the relationship between the variance RRV, the average heart rate BEAT, and mental stress. The variance value RRV is expressed by the following formula 1 using RRl (RlintervaD) indicating the interval on the time axis of a peak equal to or higher than a predetermined level of the cardiac potential signal (R wave) shown in FIG. Is calculated by

The average heart rate BEAT is given by the following equation (2). Is calculated by Here, N is the number of samples of the RRI data. When the calculation is performed while shifting the analysis section in chronological order, the analysis result becomes R as shown in FIG.
It can be represented as a distribution area on the RV-BEAT two-dimensional plane. Specifically, a task state and a rest state are repeatedly imposed on the subject, and from the distribution area, `` a case where the user cannot concentrate on the task at the time of the task '' and a `` case where the person cannot relax at the time of the rest '' are considered as cautionary cases that are not in the healthy state, and The direction of increase / decrease of the stress is determined from the change in the distribution.

FIG. 7 shows the influence of disturbance during the experiment. The disturbance includes a disturbance caused by the subject and a disturbance not caused by the subject. Disturbances caused by the subject include, for example, deep breathing such as lack of extension, respiratory vibrations and body vibrations caused by sitting back, and the like, which fluctuates the heart rate. In addition, the disturbance that is not caused by the subject is the environment, and a quiet and calm place is desirable as the environment.However, it is not realistic to detect a perfect place while driving, and the disturbance due to this environment affects the heart rate .

As shown in FIG. 7A, a disturbance often causes a rapid rise in heart rate. The rapid increase in heart rate causes the beat interval to be extremely low, and the average heart rate BEAT in the rest state in the distribution area on the RRV-BEAT two-dimensional plane.
Although it does not change much, the RRV value has a large value (see FIG. 7C). Since the mental stress is determined at the position of the distribution area, the influence of the disturbance results in a large error mixed in the determination result. Therefore, since a steep change due to disturbance appears in a high frequency band, the detection accuracy is prevented from lowering by removing a high frequency region.

Next, the cardiac potential signal and the signal equivalent to respiratory vibration acquired as described above will be analyzed / determined according to the flowcharts shown in FIGS. 8 and 10 and the signal waveforms shown in FIG. 9. This will be described.

In step 201, the analysis / judgment device 30
The heartbeat-equivalent signal (see FIG. 9A) acquired by
Sampling is performed at 00 Hz or more, and A / D conversion is performed.

In step 202, untargeted signals such as noise due to human body movement and high frequency noise are removed by filtering the sampled cardiac potential signal. As an example of filtering, for example, 5 to 3
A 0 Hz band-pass filter, a matched filter that prepares a signal in a reference form, and correlates with the acquired signal, or the like is used (see FIG. 9B).

In step 203, the signal level after filtering is compared with a predetermined value, and only a signal having a predetermined value or more is left (see FIG. 9C).

In step 204, RRI (R-interval) data, which is the time interval between beats, is detected, and data with the vertical axis representing RRI and the horizontal axis representing time is created (FIG. 9).
(D)).

In step 205, the new RRI (n)
At the time of data detection, the RRI (n-1) data immediately before the data is referenced, and 0.5 RRI (n-1) to 1.5 RRI
Only within the range of (n-1) is set as the detection range of the next R wave. If the R-wave is missing due to arrhythmia, the RRl data is about twice as large as the original and can be detected by this method. This RRI data is regarded as a value containing an error and is inconvenient (step 207, FIG. 9E). 0.
The reason for setting the value to 5 to 1.5 is that it is considered that there is no instantaneous change of 50% or more in the mental task. The arrhythmia in a person living a normal life returns to normal pulsation after the phenomenon occurs, so that subsequent pulsations can be used again.

In step 206, the sampling is converted to regular sampling in a time series by an interpolation process. Originally, since the heartbeat is irregular, the RRl data fluctuates because the heartbeat is irregular. Therefore, the RRI data detected in step 204 becomes irregular sampling data in a time series.

Therefore, in this step, pseudo signals are inserted at regular intervals on the signal shown in FIG. 9D to obtain regular sampling data as shown in FIG. 9F. In order to simplify the processing, the level of the pseudo signal is set to the same level as the previous value on the time axis. It has been experimentally confirmed that the degree of the variation in the level of the RRI data has a close correlation with the sympathetic parasympathetic nervous activity, and thus variance is used as an index representing this variation. However, there is a possibility that the influence of disturbance is mixed in the obtained RRI data.

Therefore, as described above, the steep fluctuation due to the disturbance appears in a high frequency band.
A time / frequency analysis is performed on the RRI time series data interpolated in the above, and a low frequency band of about 0.01 Hz to 0.5 Hz, which is a frequency band caused by mental workload, is specified (see FIG. 9 (g)). .

Next, in step 209, data outside the frequency band due to the mental work load is removed, and frequency / time conversion is performed to reorganize the data on the time axis (see FIG. 9 (h)).

Next, using the signal equivalent to respiratory vibration acquired based on the piezoelectric sensor 25, steps 210 to 211 are performed.
Eliminates the heart rate variability caused by the driver. More specifically, in step 210, the magnitude of the signal equivalent to respiratory vibration is compared with a predetermined threshold value, and is identified as "H" when large, "L" when small, and "M" in the meantime.

Next, at step 211, step 210
A moving average process is performed based on the size identified by.
In the case of “L”, it is determined that the level is a normal level, and the moving average processing is not performed. In the case of "M", it is assumed that the level is a little deep breathing or body motion level, and three points of moving average processing including one point before and after is performed. In the case of "H", it is determined that the level is a large deep breathing or body movement level, and a moving average process of five points including two points before and after is performed (see FIG. 9 (i)).

Next, at step 212, RRV-BEAT is calculated based on the above-mentioned calculation formula.

Next, at step 213, the current vehicle speed data is fetched based on the vehicle feed signal from the vehicle speed sensor 40.
Next, in step 214, it is determined whether the vehicle is in the rest state or the task state based on the vehicle speed data acquired in step 213. Here, one vehicle speed data is 5 km /
If it is less than h, it is determined that the vehicle is stopped and the rest state
If / h or more, it is determined that the vehicle is running and the task state is set.

Next, in steps 215 and 216, based on the rest / task state determined in step 214, the section average state value Gt in the task section in step 215 and the section average state value Gr in the rest section in step 216 are calculated. calculate. (See FIG. 9 (j)).

The section average state value G is specified as a two-dimensional distribution position of the variance value RRV and the average heart rate BEAT by calculating the variance value RRV and the average heart rate BEAT.

Next, referring to FIG. 10, in step 801, the difference X1 between the previous task section average state value Gt1 and the rest section average state value Gr1 calculated in steps 215 and 216 of FIG. State value G
The difference X2 between t2 and the rest section average state value Gr2 is calculated.

At step 802, the difference X2 and the predetermined value Xth
If the difference X2 is larger than the predetermined value Xth, the process ends as a healthy person. If the difference X2 is smaller than the predetermined value Xth, it is determined that the case requires attention, and the next step 803 is executed.
To compare the previous value X1 with the current value X2. If the previous value X1 is equal to the current value X2, it is determined in step 805 that there is no change in mental stress. If the current value X2 is smaller than the previous value X1, it is determined that the current value X2 is smaller than the previous value X1. Basically, healthy individuals have a clear difference in response between when they are temporarily stressed and when they are not.

Looking at the distribution area shown in FIG.
Under non-mental stress, the distribution area of the task state and the distribution state of the rest state are distributed at positions separated from each other, and if the task is correctly performed, the distribution area of the task state has a high burden on the distribution area of the rest state The result is located at the upper left. This is a normal response given the sympathetic parasympathetic activity. However, in a state under mental stress, which is a constant stress lurking in the background, a phenomenon of non-concentration in the task state and non-relaxation in the rest state occur, and the mutual distribution areas come close to each other. Therefore,
High-precision determination is possible by making a determination based on the temporal change of the section average state distance of each distribution area as a scale.

In this method, in order to obtain a more accurate result, the average heart rate BEAT and the variance value RRV are used.
We have focused on the difference in distribution area on the two-dimensional plane.
Even when only the variance RRV of the RI is calculated, it is determined whether or not the subject is a healthy person by comparing the difference Y2 between the variance value of the current task period and the variance value of the current rest period with a predetermined value. By comparing the difference Y2 between the variance value of the period and the variance value of the previous rest period with the difference Y1, it is determined whether the tendency is increasing or decreasing or not. Can be expected.

As described above, according to the present embodiment, the subject's heartbeat-equivalent signal and respiratory vibration-equivalent signal when a load is applied to the subject driver are acquired, and the heartbeat-equivalent signal is displayed on the time axis. Time-frequency conversion is performed on beat interval data, and only data in a frequency band equal to or lower than a predetermined frequency is specified from the converted data, and frequency-time conversion is performed on the specified data. Based on the beat interval data on the time axis, based on the signal equivalent to respiratory vibration, moving average processing is performed, and based on this moving average beat interval data, stress analysis is performed. Disturbances, such as the effects of sudden changes in heart rate caused by fluctuations in surrounding sounds, and disturbances caused by the subject, such as respiratory vibrations and body vibrations caused by deep breathing such as underextension, sitting back, etc. Can remove the influence due to a sudden change in the heart rate by, there is an effect that a reduction in the judgment result accuracy can be prevented.

Since the magnitude of the respiratory vibration equivalent signal is determined and the moving average section of the moving average process is set based on the determined magnitude, the influence based on the respiratory vibration is further removed. It is possible to prevent a decrease in the accuracy of the determination result.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram when a mental stress determination device is applied to a vehicle.

FIG. 2 is a view for explaining a method of acquiring a heartbeat-equivalent signal.

FIG. 3 is a view for explaining a method of acquiring a heartbeat-equivalent signal.

FIG. 4 is an explanatory diagram of a method for acquiring a signal equivalent to respiratory vibration.

FIG. 5 is a diagram illustrating the configuration of a mental stress determination device.

FIG. 6 is a diagram illustrating a relationship between distribution and stress.

FIG. 7 is a diagram illustrating the influence of disturbance.

FIG. 8 is an explanatory diagram of a method of analyzing / determining a heartbeat-equivalent signal.

FIG. 9 is an explanatory diagram of a method for analyzing / determining a heartbeat-equivalent signal.

FIG. 10 is an explanatory diagram of a method of analyzing / determining a heartbeat equivalent signal.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Vehicle 5 Seat 10 Steering 15 Electrode 20 Seat belt 25 Piezoelectric sensor 30 Analysis / judgment device 40 Vehicle speed sensor 50 Wristband 55 Electrode

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) A61B 5/04 312R

Claims (2)

[Claims] An acquisition means for acquiring a heartbeat-equivalent signal and a respiratory vibration-equivalent signal of a subject when a load is applied to the subject, and time-frequency conversion of beat interval data on the time axis of the heartbeat-equivalent signal. A first conversion unit for performing data conversion; a data specifying unit for specifying only data in a frequency band equal to or lower than a predetermined frequency among the data converted by the first conversion unit; A second conversion unit for performing frequency-time conversion, and a movement for performing a moving average process on the beat interval data on the time axis converted by the second conversion unit based on the respiratory vibration equivalent signal. Mental stress, comprising: averaging means; and stress analysis means for performing stress analysis based on beat interval data moving averaged by the moving average means. Judgment device. 2. The mental stress judging device according to claim 1, further comprising a judging means for judging a magnitude of the respiratory vibration equivalent signal, wherein the moving average means is based on the magnitude judged by the judging means. And setting a moving average section of a moving average process.