JP2007000280A - Arousal lowering determination device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an arousal lowering determination device capable of highly precisely determining the lowering state of the arousal of a specific subject. <P>SOLUTION: When measuring detecting signals of brain wave, respiration, body temperature, blinking and heart rate, or a physiological index group of a specific user (ST2), this arousal lowering determination device computes the physiological index feature quantity such as the brain wave α/β and the like of the specific user (ST3), and calculates a first main component obtained by analyzing the main component of the physiological index feature quantity and having no cross-correlation as an arousal lowering determination index (ST4). In this case, this device uses an eigenvector of the first main component prestored in the eigenvector database and corresponding to the specific user. This device thus computes a time mean value of the arousal lowering determination index corresponding to the specific user (ST5), and highly precisely determines and outputs the lowering state of the arousal of the specific user according to the time mean value (ST6 and ST7). <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a wakefulness reduction determination device that determines a state of reduction in wakefulness of a specific subject such as a vehicle driver.

  2. Description of the Related Art Conventionally, there is generally known a vehicle attention level reduction preventing apparatus that includes a driver identification device that identifies a vehicle driver and an arousal level reduction state determination unit that determines a reduction level of a vehicle driver's arousal level. In this vehicle attention level reduction preventing device, the arousal level reduction state determination unit is configured to determine the reduction level of the driver's arousal level based on beat interval data when the heart rate increases instantaneously. (For example, refer to Patent Document 1).

In addition, a sleep state determination apparatus that determines the sleep state of a subject by measuring and integrating biological information such as the subject's body temperature, heart rate, and breathing is generally known. In this sleep state determination apparatus, biological information such as body temperature, heartbeat, and respiration is integrated by a clustering technique such as vector quantization (see, for example, Patent Document 2).
JP 11-314534 A (paragraph number 18, paragraph number 19) JP 2004-254827 A (paragraph number 10, paragraph number 15, paragraph number 31)

  By the way, the arousal level reduction state determination unit described in Patent Document 1 is configured to determine the reduction level of the driver's arousal level simply based on pulsation interval data when the heart rate instantaneously increases. Therefore, there is a problem that it is difficult to accurately determine the state of decrease in the arousal level.

  Moreover, since the sleep state determination apparatus described in Patent Document 2 integrates biological information (physiological indices) such as body temperature, heartbeat, and respiration by a clustering technique such as vector quantization, biological information (physiological indices). There is a problem that mutual correlation is high, it is difficult to pour out a specific change when the arousal level is lowered, and it is difficult to accurately determine the sleep state.

  Then, this invention makes it a subject to provide the wakefulness fall determination apparatus which can determine the fall state of the wakefulness of a specific test subject accurately.

  The arousal level decrease determination device according to the present invention is a device that determines a decrease state of a specific subject's arousal level based on the input measurement data of a specific subject's physiological index group, and the specific subject's arousal level based on the measurement data A plurality of physiological index feature values related to a decrease in the degree, a first principal component obtained by principal component analysis of the obtained plurality of physiological index feature values is obtained as an arousal level decrease determination index, and the arousal level decrease determination index Calculating means for calculating the time average value of the first subject, output means for outputting the state of decrease in the arousal level of the specific subject according to the time average value of the wakefulness level reduction determination index obtained by the calculating means, and the first corresponding to the specific subject Storage means in which an eigenvector of one principal component is stored in advance, and the computing means uses the eigenvector of the first principal component corresponding to the specific subject stored in advance in the storage means. And determining a certain awareness decrease determination indicator.

  In the arousal level decrease determination device according to the present invention, when measurement data of a physiological index group of a specific subject is input, a calculation unit obtains a plurality of physiological index feature amounts based on the measurement data, and the plurality of physiological index features. The first principal component having no cross-correlation obtained by performing principal component analysis of the amount is obtained as a wakefulness reduction determination index. At that time, the computing means determines the arousal level reduction determination index as the first principal component using the eigenvector of the first principal component corresponding to the specific subject stored in advance in the storage means, and this arousal level reduction determination index. Find the time average of. And according to the time average value of this arousal level reduction determination index, the output means outputs the reduction status of the specific subject's arousal level.

  Here, in the wakefulness reduction determination apparatus of the present invention, any of the items of electroencephalogram, respiration, body temperature, blink, heartbeat, and skin potential is included in the physiological index group as a physiological index that easily changes when the wakefulness is lowered. It is characterized by.

  Further, the wakefulness reduction determination apparatus of the present invention is characterized in that the ratio between the electroencephalogram α wave power and the β wave power, which has been found to change easily when the wakefulness is lowered, is included in the physiological index feature amount. To do.

  According to the device for determining arousal level reduction according to the present invention, the calculation means obtains a plurality of physiological index feature values based on the measurement data of the physiological index group of the specific subject, and the mutual analysis is performed by principal component analysis of the plurality of physiological index feature values. Since the first principal component having no correlation is obtained as the arousal level reduction determination index, it is possible to accurately extract the reduction status of the arousal level of the specific user.

  Further, when the calculation means obtains the first principal component obtained by principal component analysis of a plurality of physiological index feature values as a wakefulness reduction determination index, an eigenvector of the first principal component corresponding to the specific subject stored in advance in the storage means is obtained. In order to use the first principal component to determine the arousal level decrease determination index and obtain the time average value of the arousal level decrease determination index, regardless of the individual difference in the calculation data of the physiological index feature amount, It is possible to accurately determine and output the state of decrease in arousal level.

  DESCRIPTION OF EMBODIMENTS Hereinafter, an embodiment of a wakefulness reduction determination apparatus according to the present invention will be described with reference to the drawings. In the drawings to be referred to, FIG. 1 is a block configuration diagram of a wakefulness reduction determination device according to an embodiment.

  The arousal level reduction determination apparatus according to an embodiment reduces a wakefulness level of a specific user based on the input physiological index group of the specific user (specific test subject), that is, measurement data of an electroencephalogram, respiration, body temperature, blink, and heart rate. An apparatus for determining a situation, as shown in FIG. 1, as shown in FIG. 1, a user information input unit 1, a physiological index group measurement unit 2, a physiological index feature value calculation unit 3, a wakefulness level decrease determination index calculation unit 4, an eigenvector database 5, The determination index averaging calculation unit 6 and the arousal level decrease determination signal output unit 7 are provided.

  The user information input unit 1 to the arousal level determination signal output unit 7 include an input / output interface I / O, an A / D converter, a ROM (Read Only Memory) storing programs and data, and a RAM (temporarily storing input data) Random Access Memory), microcomputer (CPU) that executes a program, and the like are configured using microcomputer hardware and software.

  Here, in order to identify the user (subject) to be subjected to the current arousal level determination, the user information input unit 1 is input with the user's name, initials, ID number, and the like. The user information input unit 1 outputs the input user information signal such as the name, initials, and ID number of the specific user to the physiological index group measurement unit 2 and the eigenvector database 5.

  When the information signal of the specific user is input from the user information input unit 1, the physiological index group measurement unit 2 inputs the detection signal of the specific user's brain wave, respiration, armpit body temperature, blink, and heart rate from various sensors (not shown). And measure. Then, the physiological index group measurement unit 2 outputs the measured data to the physiological index feature amount calculation unit 3.

  The physiological index feature value calculation unit 3, the arousal level decrease determination index calculation unit 4, and the arousal level decrease determination index average unit 6 constitute a calculation unit, and the physiological index feature value calculation unit 3 is derived from the physiological index group measurement unit 2. When measurement data of a specific user's physiological index group is input, first to eighth physiological index feature amounts x1 to x8, which will be described later, are calculated based on the measurement data, and the calculated data is used as a wakefulness reduction determination index calculation unit. 4 is output.

  As the 1st to 8th physiological index feature values x1 to x8 calculated by the physiological index feature value calculation unit 3, generally those greatly related to a decrease in the user's arousal level are selected. Here, FIG. 2 is a graph showing the relationship between the level of the user's brain wave α / β (ratio of the brain wave β wave power to the brain wave α wave power) and the level of arousal level, which has been found as a representative example. 3 is a graph showing the relationship between the level of the brain wave α wave power of the user and the level of wakefulness, and FIG. 4 is a graph showing the relationship between the level of the brain wave β wave power of the user and the level of wakefulness. is there.

  In the graphs of FIGS. 2 to 4, the change in the level of wakefulness is indicated by a dotted line, and in the state where the level of wakefulness is greatly reduced and sleepy, the brain wave α / β indicated by the solid line in FIG. The level is rising rapidly. On the other hand, the level of the electroencephalogram α wave power indicated by the solid line in FIG. 3 shows a high level from before the level of wakefulness to the time of reduction, and the level of the electroencephalogram β wave power indicated by the solid line in FIG. Is corresponding to a decrease in the level of arousal.

As a result of such comparative study, it has been found that the level of the brain wave α / β indicated by the solid line in FIG. 2 is greatly related to the decrease in the user's arousal level, and the level of the brain wave α / β is the first level. Is selected as the physiological index feature amount x1. As shown in Table 1 below, as the second to eighth physiological index feature amounts x2 to x8, the respiratory cycle (seconds), armpit body temperature (° C.), number of blinks (per unit time), heart rate fluctuation Heart rate which is the ratio of high frequency component H to high frequency component H of 0.15-0.5 Hz, low frequency component L of 0.01 to 0.15 Hz of heart rate fluctuation, heart rate RRI per minute, low frequency component L of heart rate fluctuation Fluctuation L / H is selected.

  Here, whenever the measurement data of the physiological index group of the specific user is sequentially input from the physiological index group measurement unit 2 at a predetermined time interval, the physiological index feature value calculation unit 3 is shown in Table 1 based on the measurement data. The first to eighth physiological index feature amounts x1 to x8 are calculated, and the calculated data is output to the arousal level decrease determination index calculation unit 4. In this case, the physiological index feature amounts x1 to x8 are calculated as values standardized using the average and standard deviation for the past one hour.

Each time the awakening level decrease determination index calculation unit 4 receives the calculation data of the standardized values of the first to eighth physiological index feature amounts x1 to x8 from the physiological index feature amount calculation unit 3, the awakening degree decrease determination index calculation unit 4 Physiological index feature amounts x1 to x8 of 1 to 8 are subjected to principal component analysis to obtain a first principal component, and this first principal component is used as a wakefulness reduction determination index. If this arousal level decrease determination index is A and the eigenvector of the first principal component is U1 (= u1, u2, u3, u4, u5, u6, u7, u8), the arousal level decrease determination index A is expressed by the following formula (1 ).

Here, the eigenvector database 5 serving as the storage means stores the eigenvector values U1 (= u1, u2, u3, u4, u5, u6) of the first principal component corresponding to a plurality of users including a predetermined specific user. u7, u8) are stored in advance. The eigenvector value U1 (= u1, u2, u3, u4, u5, u6, u7, u8) for each user is based on measurement data of each physiological index group measured in advance by the physiological index group measuring unit 2. The physiological index feature amounts x1 to x8 calculated by the feature amount calculation unit 3 are calculated by the principal component analysis by the arousal level reduction determination index calculation unit 4, and for example, as shown in the following Table 2, Individual differences are recognized for each user.

  When an information signal such as the name, initials, and ID number of a specific user is input from the user information input unit 1, the eigenvector database 5 stores the eigenvector values u <b> 1 to u <b> 8 corresponding to the specific user to the arousal level decrease determination index calculation unit 4. Output. For example, when the name of the specific user O is input as a signal, the values of the eigenvectors u1 to u8 in the column corresponding to the user O in Table 2 are output as the eigenvector U1 of the user O to the arousal level reduction determination index calculation unit 4.

  Here, the arousal level decrease determination index calculation unit 4 performs principal component analysis on the first to eighth physiological index feature amounts x1 to x8 shown in Table 1 as described above, and the arousal level decrease is the first main component. When the determination index A is obtained from the equation (1), the values of the eigenvectors u1 to u8 (see Table 2) corresponding to the specific user O input from the eigenvector database 5 are used as the values of u1 to u8 in the equation (1). .

At that time, the wakefulness level decrease determination index calculation unit 4 uses, for example, the standardized values of the physiological index feature values x1 to x8 of the specific user O calculated by the physiological index feature value calculation unit 3 this time (time t = 0) as follows. In the case of the values shown in Table 3, the value of A (t = 0) = 1.89 is calculated as the arousal level decrease determination index A this time (time t = 0), and the calculated data is averaged for the awakening level decrease determination index. Output to unit 6. Similarly, the arousal level reduction determination index calculation unit 4 responds to the standardized values of the physiological index feature values x1 to x8 of the specific user O that the physiological index feature value calculation unit 3 has previously calculated (time t = −1). As the previous (time t = −1) wakefulness level decrease determination index A, for example, arithmetic data having a value of A (t = −1) = 1.90 is output to the wakefulness level decrease determination index average unit 6.

The determination index averaging calculation unit 6 performs time averaging of a plurality of arousal level decrease determination indexes A sequentially input from the arousal level decrease determination index calculation unit 4 at predetermined time intervals by the following equation (2). The time average value Amean of the decrease determination index A is calculated, and the calculated data is output to the arousal level decrease determination signal output unit 7. In Equation (2), T represents the number of samples, N represents the total number of samples up to the present, and i represents the sample number.

  Here, the time average value Amean of the arousal level decrease determination index A has a characteristic that gradually changes according to the level of the user's arousal level, for example, as shown in the graph of FIG. That is, in a state where the level of the user's arousal level is high and the consciousness is clear, the value of American is a negative side whose center of gravity is smaller than 0, and in a state where the level of the user's arousal level is extremely low and sleepy, the value of American Is the positive side whose center of gravity is greater than zero. In other words, as shown in FIG. 6, the time average value Amean of the wakefulness level decrease determination index A generally takes a negative value when the user's wakefulness level is high and the user's wakefulness level is low. Then take a positive value.

  Therefore, when the time average value Amean of the arousal level decrease determination index A is input from the determination index averaging calculation unit 6, the awakening level decrease determination signal output unit 7 as an output unit receives the value as a predetermined determination threshold value TH ( For example, it is determined whether it is greater than zero. Then, the arousal level decrease determination signal output unit 7 has a reduced level of arousal level of a specific user when the time average value Amean of the arousal level decrease determination index A is larger than a determination threshold TH (for example, zero). And the wakefulness lowering determination signal is output to an alarm device (not shown) or the like to urge the specific user to wake up.

  The wakefulness reduction determination apparatus according to an embodiment configured as described above is equipped with, for example, an arbitrary automobile and the above-described eigenvector values U1 (= u1, u2, u3, u4, unique to a user driving the automobile). By storing the data of u5, u6, u7, u8) in the eigenvector database 5 in advance, it can be used to prevent a specific user from falling asleep. In this case, a plurality of users who drive the car can be set.

  In the arousal level decrease determination apparatus according to the embodiment used in this way, a decrease state of the specific user's arousal level is determined in accordance with the processing procedure of the flowchart shown in FIG. First, when a specific user who is a driver inputs information such as a name, initials, and ID number to the user information input unit 1 in step S1, in step S2, the physiological index group measurement unit 2 detects various sensors (not shown). Measurement is performed by inputting detection signals of brain waves, respiration, body temperature, blinks, and heartbeats, which are physiological index groups of a specific user.

  In the next step S3, based on the measurement data of the brain wave, respiration, body temperature, blink, and heart rate of the specific user, the physiological index feature value calculation unit 3 sets the specific user's physiological index feature value, that is, the first shown in Table 1. Physiological index feature amounts x1 to x8 of ˜8 are calculated.

  In step S4, based on the calculation data of the first to eighth physiological index feature amounts x1 to x8, the arousal level decrease determination index calculation unit 4 performs principal component analysis on the first to eighth physiological index feature amounts x1 to x8. Then, the first principal component having no cross-correlation is calculated as the arousal level decrease determination index A (see Formula (1)).

  Here, the arousal level decrease determination index calculation unit 4 calculates the values of eigenvectors u1 to u8 corresponding to specific users stored in the eigenvector database 5 in advance (table) when calculating the arousal level decrease determination index A of Equation (1). 2) is used as the values of u1 to u8 in the formula (1) to determine the arousal level decrease determination index A of the specific user.

  In subsequent step S5, the determination index averaging calculation unit 6 performs time averaging of the plurality of arousal level decrease determination indexes A sequentially input from the arousal level decrease determination index calculation unit 4 at predetermined time intervals according to Equation (2). The time average value Amean of the arousal level decrease determination index A is calculated.

  In step S6, the wakefulness level decrease determination signal output unit 7 determines whether or not the time average value Amean of the wakefulness level decrease determination index A is greater than a predetermined determination threshold value TH (for example, zero). It is determined whether or not the user's arousal level is reduced.

  Here, if the determination result in step S6 is NO and it is determined that the awakening level of the specific user has not decreased, the process returns to step S2, but the determination result is YES and the awakening of the specific user. If it is determined that the degree is low, in a subsequent step S7, the wakefulness reduction determination signal output unit 7 outputs a wakefulness reduction determination signal to an alarm device (not shown) or the like to urge the specific user to wake up. .

  As described above, according to the device for determining arousal level according to the embodiment, in step S3, physiological index feature amounts x1 to x8 that are largely related to a decrease in the user's arousal level, that is, the first to eighth values shown in Table 1. In addition to calculating the physiological index feature amounts x1 to x8, in step S4, the first to eighth physiological index feature amounts x1 to x8 are subjected to principal component analysis, and the first principal component having no cross-correlation is determined as the arousal level reduction determination index A. Therefore, it is possible to accurately extract the state of decrease in the arousal level of the specific user.

  Further, when calculating the arousal level decrease determination index A of Formula (1) in step S4, the values of the eigenvectors u1 to u8 corresponding to the specific user stored in advance in the eigenvector database 5 (see Table 2) are used. Thus, the wakefulness level decrease determination index A of the specific user is determined, and the time average value Amean of the wakefulness level decrease determination index A is calculated in step S5. Regardless of individual differences in the calculation data of the amounts x1 to x8, it is possible to accurately determine and output the state of decrease in the arousal level of the specific user.

  The arousal level decrease determination device according to the present invention is not limited to the above-described embodiment. For example, the user information input unit 1 is configured to display the names of a plurality of pre-registered users (subjects) on the screen, and select the name of the specific user (specific subject) this time by a switch operation or the like. Also good.

  Further, the arousal level decrease determination signal output unit 7 outputs a wakefulness level decrease determination signal not only to an alarm device but also to an appropriate device that promotes a specific user's arousal, for example, a device that stimulates a skin sensation such as a specific user's face. You may comprise.

It is a block block diagram of the arousal level fall determination apparatus which concerns on one Embodiment of this invention. It is a graph which shows the relationship between the level of a user's brain wave (alpha) / (beta), and the level of arousal level. It is a graph which shows the relationship between the level of a user's brain wave alpha wave power, and the level of arousal level. It is a graph which shows the relationship between the level of a user's brain wave beta wave power, and the level of arousal level. It is a graph which shows the relationship between the level of a user's arousal level, and the time average value Amean of the arousal level fall determination parameter | index. It is a graph which shows the change of the time average value Amean of the arousal level fall determination parameter | index corresponding to the level fall of a user's arousal level. It is a flowchart which shows the outline of the process sequence of the arousal level fall determination apparatus of one Embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 User information input part 2 Physiological index group measurement part 3 Physiological index feature-value calculation part 4 Arousal level fall determination parameter | index calculation part 5 Eigenvector database 6 Arousal level fall determination parameter | index average part 7 Arousal level fall determination signal output part

Claims (3)

A device for determining a decrease in arousal level of a specific subject based on input measurement data of a physiological index group of the specific subject,
Based on the measurement data, a plurality of physiological index feature values related to a decrease in arousal level of a specific subject are obtained, and a first principal component obtained by principal component analysis of the obtained plurality of physiological index feature values is obtained as a wakefulness reduction determination index. And the calculating means for obtaining the time average value of this arousal level decrease determination index,
An output means for outputting a state of decrease in arousal level of the specific subject according to a time average value of the arousal level decrease determination index obtained by the calculation means;
Storage means in which the eigenvector of the first principal component corresponding to the specific subject is stored in advance,
The arithmetic means determines a wakefulness reduction determination index that is a first principal component using an eigenvector of a first principal component corresponding to a specific subject stored in advance in the storage means, Judgment device.   The wakefulness reduction determination apparatus according to claim 1, wherein the physiological index group includes any of the following items: electroencephalogram, respiration, body temperature, blink, heartbeat, and skin potential.   The wakefulness level decrease determination apparatus according to claim 1, wherein the plurality of physiological index feature amounts include at least a ratio of α wave power and β wave power of an electroencephalogram.

Images