Sleepiness calculation device

Abstract

The present invention provides a sleepiness calculation device that can accurately calculate sleepiness by taking into account the subject's current and past activities. [Solution] In the smartphone 1, the I / F 18 acquires the subject's heart rate and measurement time measured by the heart rate sensor 19, and the state determination unit 11 determines the subject's activity state and transmits it to the server 2. Next, in the server 2, the correction unit 24 acquires the subject's drowsiness criterion heart rate from the drowsiness criterion heart rate table 21, which stores the drowsiness criterion heart rate for each activity state, according to the activity state transmitted from the smartphone 1. Next, the correction unit 24 corrects the drowsiness criterion heart rate acquired based on fatigue level, sleep duration, and meal status. Then, the corrected drowsiness criterion heart rate (corrected drowsiness criterion heart rate) is transmitted to the smartphone 1, and the drowsiness level calculation unit 16 of the smartphone 1 calculates the subject's drowsiness level based on the corrected drowsiness criterion heart rate and the heart rate acquired by the I / F 18.

Description

Technical Field

[0001] The present invention relates to a drowsiness calculation device that calculates the drowsiness of a subject.

Background Art

[0002] For example, in order to detect drowsy driving of a vehicle, it has been proposed to detect a drowsy state from the heart rate of a driver (see, for example, Patent Documents 1 and 2).

[0003] Patent Document 1 describes using, as a threshold value, the average value of the RRI (R-R Interval) values of the heartbeats detected by a heart rate sensor during wakefulness and a predetermined multiple of the integral value of the RRI values exceeding the average value, and integrating the RRI values exceeding the average value, and determining that it is drowsiness when the threshold value is exceeded.

[0004] Patent Document 2 obtains heartbeat interval data in which RRI is serialized based on the heartbeat waveform of a subject in a resting state measured by a sensor. Next, the heartbeat interval data is subjected to frequency analysis to obtain a frequency analysis result of heartbeat fluctuation including the power spectral density (PSD) with respect to the frequency at a certain time and the total power (TP) of the autonomic nerve. Next, the frequency analysis result is set as an initial state, and based on the estimated drowsiness position and the estimated wakefulness position included in the initial state, a drowsiness scale in which the origin of the drowsiness position and the origin of the wakefulness position are set is determined. And it describes determining the drowsiness of the subject based on the drowsiness scale.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0006] In the method described in Patent Document 1, data from the initial awakening state is used as a reference. However, the initial state varies from person to person, and depending on the degree of awakening in the initial state, it was sometimes not possible to appropriately set the average RRI value, etc. Therefore, there was a problem in that it could not cope with the differences in the initial state from person to person.

[0007] The method described in Patent Document 2 has the following problems: The correspondence between heart rate information and the sleepiness scale changes due to various factors. Even based on a resting state, there are daily and time-of-day variations. For example, it is known that sleepiness is strongest between 3 and 4 in the early morning and between 3 and 4 in the afternoon. Therefore, it is difficult to accurately estimate sleepiness using only heart rate information. In addition, driving is a kind of stressful state, and the baseline heart rate differs depending on the road conditions (city or highway). Furthermore, since the initial state at the start of measurement is not constant, it could not accommodate changes such as being sleepy from the beginning or not being sleepy at first.

[0008] Furthermore, in order to determine drowsiness, it is necessary to set some kind of standard heart rate, etc., as described in Patent Documents 1 and 2. However, heart rate not only varies from person to person, but also fluctuates depending on the time of day and activity level. Moreover, drowsiness is also influenced by past actions (activities) such as the state of sleep (whether one slept well the previous night, etc.), eating habits (whether one ate or not, etc.), or fatigue level (whether one is tired or not).

[0009] The methods described in Patent Documents 1 and 2 do not take into account the subject's activities, so depending on the subject's activities, accurate detection of drowsiness may not be possible.

[0010] Therefore, in view of the above-mentioned problems, the object of the present invention is to provide a sleepiness calculation device that can calculate sleepiness with high accuracy by taking into account, for example, the subject's current activities and past activities. [Means for solving the problem]

[0011] To solve the above problems, the invention described in claim 1 is a sleepiness calculation device characterized by comprising: a biological information acquisition unit that acquires biological information relating to the heart rate of a subject; a fatigue level information acquisition unit that acquires fatigue level information relating to the fatigue level of the subject; and a sleepiness level calculation unit that calculates information relating to the current sleepiness of the subject based on the biological information and the fatigue level information. [Brief explanation of the drawing]

[0012] [Figure 1] This is a schematic diagram of a sleepiness calculation device according to the first embodiment of the present invention. [Figure 2] Figure 1 is a schematic diagram of the smartphone's configuration. [Figure 3] Figure 1 is a schematic diagram of the server configuration. [Figure 4] Figure 2 shows a table illustrating examples of activity status determination in the state determination unit. [Figure 5] Figure 2 shows an example of a parameter table. [Figure 6] Figure 2 is an explanatory diagram illustrating an example of the display of the drowsiness indicator unit. [Figure 7] Figure 3 shows an example of a heart rate table based on drowsiness criteria. [Figure 8] This is another example of a configuration for calculating and setting a sleepiness threshold heart rate table. [Figure 9] This is an explanatory diagram about the interpolation of the sleepiness criterion heart rate table. [Figure 10] This is a table of correction values ​​corresponding to fatigue levels. [Figure 11] This is an example of a screen where the subject inputs their fatigue level. [Figure 12] This is an example of a screen where the subject inputs their sleep status. [Figure 13] This is a table of correction values ​​corresponding to the time elapsed since eating. [Figure 14] This is a flowchart for determining whether someone has eaten or not based on location information. [Figure 15] This is a table of correction values ​​corresponding to fatigue levels and sleep levels. [Figure 16] It is a flowchart of the operation of the drowsiness calculation device shown in FIG. 1. [Figure 17] It is a flowchart of the operation of correcting the drowsiness reference heart rate shown in FIG. 16. [Figure 18] It is a table of correction values corresponding to the destination of the drowsiness calculation device according to the second embodiment of the present invention. [Figure 19] It is a table of correction values corresponding to the distance to the destination. [Figure 20] It is a table of correction values corresponding to the road shape. [Figure 21] It is an explanatory diagram of a display example before subjective evaluation of the drowsiness display unit of the drowsiness calculation device according to the third embodiment of the present invention. [Figure 22] It is an explanatory diagram of a display example after subjective evaluation of the drowsiness display unit of the drowsiness calculation device according to the third embodiment of the present invention.

Mode for Carrying Out the Invention

[0013] Hereinafter, a drowsiness calculation device according to an embodiment of the present invention will be described. The drowsiness calculation device according to an embodiment of the present invention acquires biological information regarding the heartbeat of a subject by a biological information acquisition unit, acquires the current activity state of the subject by an activity state acquisition unit, and acquires information regarding the past activity state of the subject by an activity history acquisition unit. Next, a drowsiness reference heart rate acquisition unit acquires a drowsiness reference heart rate from a drowsiness reference heart rate storage unit in which the drowsiness reference heart rate of the subject is stored for each activity state according to the activity state acquired by the activity state acquisition unit. Next, a correction unit corrects the drowsiness reference heart rate acquired by the drowsiness reference heart rate acquisition unit based on the information regarding the past activity state acquired by the activity history acquisition unit. Then, a drowsiness degree calculation unit calculates information regarding the drowsiness of the subject based on the drowsiness reference heart rate corrected by the correction unit and the biological information acquired by the biological information acquisition unit. By doing so, since the drowsiness reference heart rate can be corrected based on the past activity state of the subject, it is possible to calculate the drowsiness with high accuracy considering not only the current activity but also the past activity.

[0014] Furthermore, the activity history acquisition unit may acquire at least one of the following: the subject's past sleep information, past meal information, and past activity-related fatigue information. By doing so, the sleepiness reference heart rate can be corrected by considering the subject's past sleep information, past meal information, and past activity-related fatigue information. Therefore, the degree of sleepiness can be calculated taking these factors into account.

[0015] Furthermore, the activity history acquisition unit may acquire the examiner's past stay history at specific locations. By doing so, past activities can be estimated based on specific locations such as restaurants or hot springs where the subject has stayed in the past, and the sleepiness reference heart rate can be corrected.

[0016] Furthermore, the system may include an activity prediction unit that predicts the subject's activity, and the correction unit may further correct the drowsiness reference heart rate acquired by the drowsiness reference heart rate acquisition unit based on the activity predicted by the activity prediction unit. In this way, the drowsiness reference heart rate can be corrected not only by past activity but also by taking into account future plans, such as going out for leisure.

[0017] Furthermore, the system may include a destination information acquisition unit that obtains information about the subject's destination, and the activity prediction unit may predict the subject's activities based on the destination information. In this way, it is possible to predict future plans, such as whether the destination is home or a trip.

[0018] Furthermore, the system may include a current location information acquisition unit that acquires information about the subject's current location. If the current activity status acquired by the activity status acquisition unit indicates that the subject is driving a vehicle, the correction unit may correct the drowsiness reference heart rate acquired by the drowsiness reference heart rate acquisition unit based on the current location information acquired by the current location information acquisition unit. In this way, when the subject is driving a vehicle, the drowsiness reference heart rate can be corrected according to the road conditions and other factors the subject is currently driving on.

[0019] Furthermore, the biological information acquisition unit may acquire the measurement time of the acquired biological information, and the sleepiness reference heart rate acquisition unit may acquire the sleepiness reference heart rate from the sleepiness reference heart rate storage unit, which stores the subject's sleepiness reference heart rate for each time and activity state, according to the measurement time and activity state. In this way, the sleepiness reference heart rate can be selected considering not only the activity state but also the time. Therefore, sleepiness can be calculated with even greater accuracy.

[0020] Furthermore, the sleepiness calculation unit may calculate information about the subject's sleepiness based on sleepiness prediction parameters that are set in advance for each activity level. In this way, multiple elements included in the biometric information related to heart rate can be weighted according to the activity level.

[0021] The system also includes a display unit that shows information about drowsiness to the subject, and an input unit that receives the subject's evaluation of the drowsiness information displayed on the display unit. The drowsiness level calculation unit may update the drowsiness reference heart rate stored in the drowsiness reference heart rate storage unit based on the evaluation entered by the input unit, and calculate information about drowsiness. In this way, the subject can subjectively evaluate the drowsiness information calculated by the drowsiness level calculation unit. The subjective evaluation result can then be fed back to the drowsiness level calculation unit to recalculate information about drowsiness. Therefore, information about drowsiness can be calculated with greater accuracy to match the individual subject's perception.

[0022] Furthermore, in the information processing device according to one embodiment of the present invention, the activity status acquisition unit acquires the subject's activity status, and the activity history acquisition unit acquires information about the subject's past activity status. Then, the correction unit selects a drowsiness reference heart rate from a drowsiness reference heart rate storage unit, which stores the subject's drowsiness reference heart rate for each activity status, according to the activity status acquired by the activity status acquisition unit, and corrects it based on the information about the subject's past activity status acquired by the activity history acquisition unit. In this way, the drowsiness reference heart rate can be corrected based on the subject's past activity status.

[0023] Furthermore, in another embodiment of the present invention, the sleepiness calculation device acquires biological information related to the subject's heart rate in the biological information acquisition unit and acquires the subject's current activity state in the activity state acquisition unit. Next, the activity prediction unit predicts the subject's activity state. Next, the sleepiness reference heart rate acquisition unit acquires the sleepiness reference heart rate from the sleepiness reference heart rate storage unit, which stores the subject's sleepiness reference heart rate for each activity state, according to the current activity state. Next, the correction unit corrects the sleepiness reference heart rate acquired by the sleepiness reference heart rate acquisition unit based on the activity state predicted by the activity prediction unit. Then, the sleepiness level calculation unit calculates information regarding the subject's sleepiness based on the sleepiness reference heart rate corrected by the correction unit and the biological information acquired by the biological information acquisition unit. In this way, the sleepiness reference heart rate can be corrected considering the subject's future plans, so sleepiness can be calculated with high accuracy by considering not only current activities but also future activities.

[0024] Furthermore, in the sleepiness calculation method according to one embodiment of the present invention, a biological information acquisition step acquires biological information related to the subject's heart rate, an activity state acquisition step acquires the subject's current activity state, and an activity history acquisition step acquires information about the subject's past activity state. Next, a sleepiness reference heart rate acquisition step acquires the sleepiness reference heart rate from a sleepiness reference heart rate storage unit, which stores the subject's sleepiness reference heart rate for each activity state according to the current activity state. Next, a correction step corrects the sleepiness reference heart rate acquired in the sleepiness reference heart rate acquisition step based on the information about past activity state acquired in the activity history acquisition step. Then, a sleepiness level calculation step calculates information about the subject's sleepiness based on the sleepiness reference heart rate corrected in the correction step and the biological information acquired in the biological information acquisition step. In this way, the sleepiness reference heart rate can be corrected based on the subject's past activity state, so sleepiness can be calculated with high accuracy by considering not only the current activity but also the past activity.

[0025] Alternatively, the sleepiness calculation method described above may be implemented as a computer-based sleepiness calculation program. By doing so, the computer can be used to correct the sleepiness reference heart rate based on the subject's past activity levels, allowing for highly accurate sleepiness calculation that takes into account not only current activity but also past activity.

[0026] Furthermore, the drowsiness calculation program described above may be stored on a computer-readable recording medium. This allows the program to be distributed independently, in addition to being incorporated into a device, and facilitates version upgrades and other modifications. [Examples]

[0027] A sleepiness calculation device according to the first embodiment of the present invention will be described with reference to Figures 1 to 13. The sleepiness calculation device 50 according to this embodiment comprises a smartphone 1 and a server 2 as an information processing device, as shown in Figure 1. The server 2 is also configured to communicate with a stationary measuring device 3. The stationary measuring device 3 is installed, for example, indoors and is capable of measuring the heart rate and blood glucose level of a subject. When these measurements are taken, the measurement results are displayed to the subject and transmitted to the server 2. The server 2 also stores the measurement results of the stationary measuring device 3.

[0028] As shown in Figure 2, smartphone 1 includes a status determination unit 11 as an activity status acquisition unit, a GPS receiver 12, a G sensor 13, a parameter table 14, a communication unit 15, a sleepiness level calculation unit 16, a sleepiness level display unit 17 as a display unit, and an I / F 18 as a biometric information acquisition unit. A heart rate sensor 19 is also connected to smartphone 1.

[0029] The configuration of the smartphone 1 shown in Figure 2 can be configured, for example, as an application program (app). Alternatively, the smartphone 1 may be configured to include a heart rate sensor 19.

[0030] Although this embodiment uses smartphone 1 as an example, it goes without saying that it is not limited to smartphone 1; it could be a mobile phone, a tablet-type terminal, or a smartwatch, or even an in-vehicle device such as a car navigation system.

[0031] The heart rate sensor 19 can be any known sensor that can acquire at least heart rate as biometric information related to heart rate. For example, various forms such as wristwatch-type sensors can be used. Furthermore, the heart rate sensor 19 is not limited to one type, but multiple types may be used depending on the activity status described later.

[0032] In the smartphone 1 with the above configuration, the state determination unit 11 and the drowsiness level calculation unit 16 function as a computing device such as a CPU (Central Processing Unit). The parameter table 14 functions as a storage medium such as flash memory. The drowsiness level display unit 17 functions as a display device such as a liquid crystal display. The I / F 18 functions as a communication control unit that communicates with the heart rate sensor 19, etc. The GPS receiver 12 and G sensor 13 can be those built into the smartphone 1.

[0033] The state determination unit 11 determines the subject's current activity state (current activity status), such as resting, driving, working, or sleeping, based on the current location information and acceleration information (the subject's current location and acceleration) input from the GPS receiver 12 and G sensor 13 of the smartphone 1. In other words, the state determination unit 11 acquires the subject's current activity status.

[0034] As is well known, the GPS receiver 12 receives radio waves transmitted from multiple GPS (Global Positioning System) satellites, determines the current location information (latitude and longitude), and outputs it to the status determination unit 11.

[0035] The G-sensor 13 is a so-called acceleration sensor, and if it is a 3-axis acceleration sensor, for example, it can measure acceleration in the three directions of the X, Y, and Z axes. The G-sensor 13 outputs the measured acceleration to the state determination unit 11.

[0036] Figure 4 shows an example of activity state determination by the state determination unit 11. Figure 4 is an example of a table for determining activity states. In the table in Figure 4, state sta1 represents rest, state sta2 represents driving, state sta3 represents working, and state sta4 represents sleeping. Of course, activity states may include items other than those shown, such as exercise or heavy labor.

[0037] In the table in Figure 4, state sta1 (resting) is determined when the location information is the living room, the acceleration information is low, the velocity is low, and the device is a wristwatch. Here, the velocity can be calculated, for example, from the change in current location information. The device indicates the type and location of the heart rate sensor 19, which may be obtained from the heart rate sensor 19 or set separately by the subject or others.

[0038] Furthermore, if the current location information is a vehicle, the acceleration information is medium, the speed is high, and the device is in the driver's seat, it is determined to be state sta2 (driving). If the current location information is an office, the acceleration information is high, the speed is medium, and the device is in a chair, it is determined to be state sta3 (working). The state is identified as sta4 (sleep) when the report is in bed, acceleration information is low, velocity is low, and the device is in bed. Note that the acceleration information and velocity are examples only, and specific values ​​can be changed as needed in the settings.

[0039] In Figure 4, the table uses location information, acceleration information, speed information, and device information for identification, but it is also possible to use only one of these items, or two or three items. Alternatively, the heart rate measured by the heart rate sensor 19 may be used for identification. For example, if the heart rate changes little, it may indicate a resting state or sleep, and if the heart rate is low but fluctuating, it may indicate driving or working, allowing for a rough determination.

[0040] The parameter table 14 includes a table in which sleepiness prediction parameters, described later, are set for each activity state. Based on the determination result of the state determination unit 11, the sleepiness prediction parameters corresponding to that activity state are output to the sleepiness level calculation unit 16. An example of the table is shown in Figure 5.

[0041] As shown in Figure 5, there are two types of drowsiness prediction parameters, a and b. In the example in Figure 5, when the state is sta1 (resting), drowsiness prediction parameter a is set to 0.2 and drowsiness prediction parameter b is set to 0.2. When the state is sta2 (driving), drowsiness prediction parameter a is set to 0.5 and drowsiness prediction parameter b is set to 0. Also, when the state is sta3 (working), drowsiness prediction parameter a is set to 0.1 and drowsiness prediction parameter b is set to 0.2. When the state is sta4 (sleeping), drowsiness prediction parameter a is set to 0.2 and drowsiness prediction parameter b is set to 0.2. As is clear from the table in Figure 5, drowsiness prediction parameters a and b are set to values ​​in the range of 0 to 1.

[0042] The communication unit 15 communicates with the server 2. The communication unit 15 transmits to the server 2 the heart rate (including heart rate variability) measured by the heart rate sensor 19, the measurement time, and the activity status determined by the status determination unit 11. The communication unit 15 receives the corrected drowsiness reference heart rate transmitted from the server 2. The corrected drowsiness reference heart rate is the drowsiness reference heart rate that has undergone the correction described later, and the drowsiness reference heart rate will be described later.

[0043] The drowsiness level calculation unit 16 calculates the drowsiness level (information related to drowsiness) based on the heart rate measured by the heart rate sensor 19, the drowsiness prediction parameters output from the parameter table 14, and the corrected drowsiness reference heart rate received by the communication unit 15 from the server 2.

[0044] The level of drowsiness is the difference between the current heart rate and the drowsiness threshold heart rate when the current heart rate is lower than the drowsiness threshold heart rate. The level of drowsiness increases in proportion to the percentage by which the current heart rate falls below the drowsiness threshold heart rate.

[0045] Here, we will explain the premise for calculating the degree of drowsiness. In order to estimate the balance of the autonomic nervous system from heart rate variability, we extract the high-frequency variability component (HF component) corresponding to respiratory variability and the low-frequency component (LF component) corresponding to the Mayer wave, which is blood pressure variability, from the time-series data of heart rate variability, and compare the magnitudes of both. The HF component, which reflects respiratory variability, appears in heart rate variability only when the parasympathetic nervous system is tense (activated). On the other hand, the LF component appears in heart rate variability when the sympathetic nervous system is tense, and also when the parasympathetic nervous system is tense.

[0046] The HF component is typically calculated by taking the square root of the sum of the intensities in the HF component region of the power spectrum (from 0.15 Hz to 0.40 Hz). Because the HF component (HFS) varies significantly from person to person, it is normalized for each individual and converted to a value between 1 and 5. HFS = β × √(HF)···(1)

[0047] Here, the value of β is determined such that the average HFS is 2. Values ​​of HFS less than 1 are set to 1, and values ​​of HFS greater than 5 are set to 5. As a result, HFS values ​​range from 1 to 5.

[0048] In the sleepiness level calculation unit 16, if sleepiness prediction parameters are a and b, RR is the RR interval obtained from the current heart rate HR, RR_ref is the reference RR interval obtained from the reference heart rate HR_ref, and HF is the heart rate variability, then the sleepiness level D is calculated by the following equation (2). D=a×(RR-RR_ref)+b×HFS...(1)

[0049] Here, since heart rate (HR) is the number of heartbeats per minute, the RR interval is calculated from heart rate (HR) using the following equation (3). RR = 60000 / HR (milliseconds) ... (3)

[0050] Similarly, RR_ref is calculated using the following equation (4). RR_ref = 60000 / HR_ref (milliseconds) ... (4)

[0051] In this embodiment, HFS is calculated from heart rate by the sleepiness level calculation unit 16, but it may also be obtained from the heart rate sensor 19. In that case, the biometric information obtained by I / F 18 will be heart rate and the high-frequency component of heart rate variability.

[0052] Here, we will explain how to set the drowsiness prediction parameters a and b. For example, drowsiness while driving is more likely to occur during monotonous driving (such as on highways), and drowsiness is accompanied by a decline in function. In terms of physiological function, such a decline in function is characterized by a decrease in heart rate and blood pressure, and fluctuations in eye movements and brain waves. Subjective symptoms include a significant increase in fatigue, mainly due to drowsiness, lethargy, and fatigue in the limbs, and a strong feeling of decreased concentration. In terms of behavioral ability, there is a significant increase in reaction time and variability, and a decrease in accuracy, which can lead to dangerous states such as closed eyes and drowsiness.

[0053] The state of drowsiness can be classified as follows based on the function of the autonomic nervous system. <1. If you are not feeling sleepy> This state involves increased sympathetic nervous system activity and suppressed parasympathetic nervous system activity. The heart rate (HR) is high, and the high-frequency component (HF) of heart rate variability is low. <2. If you are showing signs of drowsiness> Although psychological drowsiness may be minimal due to monotonous driving or fatigue, physiological signs of drowsiness may appear. As sympathetic nervous system activity shifts from an elevated state to an inhibited state, heart rate decreases. <3. If you experience drowsiness> While the sympathetic nervous system remains suppressed, parasympathetic nervous system activity increases, resulting in a decrease in heart rate (HR) and an increase in the high-frequency component (HF) of heart rate variability. <4. A state of conflict against sleepiness> When we sense danger, we create a state of tension to resist drowsiness. When we feel a chill, for example, sympathetic nervous system activity is intermittently increased, and the high-frequency component (HF) of heart rate variability decreases. <5. A state where one cannot resist sleepiness> The tension disappears, and drowsiness begins. Sympathetic nervous system activity is suppressed, so heart rate (HR) decreases.

[0054] In the above classification, typical drowsy driving progresses from state 1 (no drowsiness) to state 2 (signs of drowsiness), and often to state 4 (a state of conflict against drowsiness). On the other hand, in a resting state, it progresses from state 1 to state 2, then to state 3 (drowsiness occurs), and finally to state 5 (a state where drowsiness cannot be resisted) if one falls asleep.

[0055] Therefore, in equation (2), as shown in Figure 4, the drowsiness prediction parameter b for heart rate variability is set to 0 during driving. In other words, during driving, the drowsiness prediction parameter a for heart rate change is increased, and the drowsiness prediction parameter b for heart rate variability is decreased. On the other hand, in a resting state, the drowsiness prediction parameter a for the heart rate change element is decreased, and the drowsiness prediction parameter b for heart rate variability is increased.

[0056] In other words, the sleepiness prediction parameters a and b are coefficients used to weight the RR interval and heart rate variability, respectively. As mentioned above, the contribution of the RR interval and heart rate variability to sleepiness differs depending on the activity level, so weighting each of them improves the accuracy of the sleepiness prediction.

[0057] The drowsiness level calculated by equation (2) will be in the range of 1 to 5. A drowsiness level of 1 indicates low drowsiness, and a drowsiness level of 5 indicates high drowsiness. Also, since the drowsiness level D is the difference between the current heart rate and the drowsiness threshold heart rate when the current heart rate is lower than the drowsiness threshold heart rate, RR <RR_re If f(HR>HR_ref), the drowsiness level is set to 1.

[0058] In other words, since the sleepiness prediction parameter changes depending on the activity level, the sleepiness level calculation unit 16 calculates the sleepiness level based on heart rate (biological information), activity level, and sleepiness reference heart rate.

[0059] The drowsiness level display unit 17 displays the drowsiness level calculated by the drowsiness level calculation unit 16. The drowsiness level may simply display the numerical value at that time, or it may be displayed as a bar graph or line graph, etc., to show how it changes over time.

[0060] An example of the display of the sleepiness level display unit 17 is shown in Figure 6. Figure 6(a) is a table showing the heart rate, sleepiness threshold heart rate, and calculated sleepiness level for each time period. Figure 6(b) is a bar graph of Figure 6(a). That is, if the heart rate is measured as in Figure 6(a), it will be displayed to the subject as in Figure 6(b).

[0061] I / F18 is the interface (I / F) to which the heart rate sensor 19 is connected. I / F18 is the interface that supports wired connections if the heart rate sensor 19 is connected via a wired connection, and the interface that supports wireless connections if the heart rate sensor 19 is connected via a wireless connection. In other words, I / F18 acquires biometric information (heart rate, measurement time, etc.) related to the subject's heart rate measured in the subject.

[0062] As shown in Figure 3, Server 2 includes a sleepiness reference heart rate table 21 as a sleepiness reference heart rate storage unit, a terminal communication unit 22, an external communication unit 23, a correction unit 24, and an individual history data storage unit 25.

[0063] As is well known, Server 2 is installed in a business office or similar location and can communicate with multiple terminals such as smartphones 1 via a network such as the Internet. Furthermore, as described above, Server 2 can also communicate with the stationary measuring instrument 3. In Server 2 with the above configuration, the drowsiness reference heart rate table 21 and the individual history data storage unit 25 function as storage media such as a hard disk. The terminal communication unit 22 and the external communication unit 23 function as network control boards and semiconductor circuits. The correction unit 24 functions as an arithmetic unit such as a CPU.

[0064] The drowsiness reference heart rate table 21 includes a table in which the subject's drowsiness reference heart rate is stored for each time and activity level. Based on the determination result of the state determination unit 11 received from the smartphone 1, the drowsiness reference heart rate corresponding to that activity level is output to the correction unit 24. An example of the table is shown in Figure 7.

[0065] Here, the drowsiness threshold heart rate is the heart rate at which drowsiness occurs. In other words, it is the heart rate below which a person feels drowsy. The method for calculating the drowsiness threshold heart rate is to find the lowest heart rate and its standard deviation during a certain activity state, such as driving, and then add the standard deviation to the lowest heart rate.

[0066] In the example shown in Figure 7, a drowsiness threshold heart rate is set for each state (sta1-sta3) at one-hour intervals from time t1 to t4. Time t1-t4 refers to times such as 12:00 PM or 1:00 PM.

[0067] The drowsiness threshold heart rate is calculated based on the standard deviation calculated from information such as heart rate obtained from the heart rate sensor 19. For example, to calculate the threshold heart rate at noon, the heart rate for a predetermined period around noon is obtained, the minimum heart rate and standard deviation are determined as described above, and the threshold heart rate is calculated by adding the standard deviation to the minimum heart rate. The activity status at that time is then obtained from the status determination unit 11 of the smartphone 1 and set in the table as the drowsiness threshold heart rate at noon for that activity status.

[0068] Furthermore, the drowsiness threshold heart rate may be calculated and set using the smartphone 1 shown in Figure 1, or it may be calculated by sending information such as heart rate to the server 2. Alternatively, it may be calculated and set separately using the configuration shown in Figure 8 and then transferred to the drowsiness threshold heart rate table 21 on the server 2. The initial calibration unit 31 calculates the drowsiness threshold heart rate using the method described above.

[0069] Furthermore, since people engage in various activities throughout the day, even if heart rate is measured for 24 hours, it is difficult to fill in all the time periods and activity levels in the sleepiness criterion heart rate table 21 (Figure 9(a)). Therefore, the data in the blank areas may be interpolated to smooth out the gaps based on the data before and after (Figure 9(b)).

[0070] Furthermore, the portion initially filled in by interpolation may be updated (added) to the calculated value if a drowsiness criterion heart rate is calculated through subsequent measurements. In addition, other interpolated values ​​may be updated based on that update (addition). In this way, the accuracy of the drowsiness criterion heart rate can be improved. That is, the drowsiness criterion heart rate table 21 also functions as a drowsiness criterion heart rate setting unit that adds or updates the drowsiness criterion heart rate based on the heart rate measurement time and the subject's activity status detected by the status determination unit 11 (activity status acquisition unit).

[0071] The terminal communication unit 22 receives heart rate (including heart rate variability), measurement time, and activity status determined by the status determination unit 11 from the smartphone 1. The received heart rate, measurement time, and activity status are used by the correction unit 24 and stored in the individual history data storage unit 25. The terminal communication unit 22 also transmits the corrected drowsiness reference heart rate corrected by the correction unit 24 to the smartphone 1. In other words, the terminal communication unit 22 functions as an activity status acquisition unit in the server 2.

[0072] The external communication unit 23 acquires heart rate and blood glucose level information (including measurement time) from the stationary measuring device 3. The acquired heart rate and blood glucose level data is stored in the individual history data storage unit 25.

[0073] The correction unit 24 reads and obtains the sleepiness reference heart rate from the sleepiness reference heart rate table 21 based on the current time and the activity status determined by the status determination unit 11. Then, it corrects the read sleepiness reference heart rate based on information about past activities, such as heart rate, activity status, and blood glucose levels, stored in the individual history data storage unit 25. The current time may be referenced from the clock function built into the server 2, or it may be obtained from an external NTP (Network Time Protocol) server, etc. In other words, the correction unit 24 also functions as a sleepiness reference heart rate acquisition unit and a date and time acquisition unit.

[0074] Here, we will explain how the correction unit 24 acquires (calculates) information about past activities. In this embodiment, information about past activities used to correct the sleepiness reference heart rate includes fatigue level, sleep duration, and meal status (whether or not it was after a meal).

[0075] First, let's explain fatigue level. Fatigue level indicates the current degree of fatigue due to past activities (fatigue information associated with past activities). Since a higher fatigue level makes one more prone to drowsiness, the accuracy of calculating drowsiness level can be improved by correcting the drowsiness threshold heart rate based on the fatigue level. Fatigue level can be calculated using the high-frequency component HF and the low-frequency component LF of heart rate variability, as described above. The low-frequency component LF can be obtained by frequency analysis of the RR interval, similar to the high-frequency component HF. Since the RR interval is the reciprocal of the heart rate, as shown in equation (3), it can be calculated from the heart rate stored in the individual history data storage unit 25. Fatigue level is calculated using the following equation (5). Fatigue level HRV = LF / HF····(5)

[0076] And if we let the average value of fatigue HRV be av, If HRV > av, then you are not tired at all. If av ≥ HRV > 0.75 × av, then you are slightly tired. If 0.75 × av ≥ HRV > 0.5 × av, then you are tired. If 0.5 × av ≥ HRV > 0.25 × av, then you are quite tired. If 0.25 × av ≥ HRV, then you are very tired. It can be classified into five stages.

[0077] Then, the above classification is set as follows: "Not tired at all" is Level 0, "Slightly tired" is Level 1, "Tired" is Level 2, "Quite tired" is Level 3, and "Very tired" is Level 4. As shown in Figure 10, the sleepiness reference heart rate is corrected according to the fatigue level. In the example in Figure 10, if the fatigue level is 4, 10 is added to the read sleepiness reference heart rate. This added value becomes the corrected sleepiness reference heart rate. The added correction value is just an example and can be changed as appropriate in the settings.

[0078] Furthermore, fatigue levels can also be calculated from results obtained from activity trackers (activity sensors), etc., which are not shown in the diagram. Activity trackers detect changes in acceleration due to body movements with a frequency of 2-3 Hz and a force of 0.01 G or higher. Measurements using such activity trackers show that the average activity level while awake is approximately 200 times / minute. Therefore, current activity levels are classified as follows. If activity level > 400, you are not tired at all. If 400 ≥ activity level > 350, you are slightly tired. If 350 ≥ activity level > 250, then you are tired. If 250 ≥ activity level > 200, you are quite tired. If your activity level is 200 or higher, you are extremely tired.

[0079] By classifying in this way, the drowsiness reference heart rate can be corrected using Figure 10. Therefore, fatigue levels can also be calculated by periodically sending the measurement results of the activity tracker to the server 2 and accumulating them in the individual history data storage unit 25.

[0080] Alternatively, a screen like the one shown in Figure 11 may be displayed on the smartphone 1, allowing the subject to make a selection.

[0081] Next, let's discuss sleep duration. Sleep duration indicates whether or not you slept well the previous night. In other words, it shows past sleep information. If you didn't sleep well, you are more likely to feel sleepy afterward, so by correcting the sleepiness threshold heart rate based on sleep duration, the accuracy of calculating sleepiness can be improved. Since heart rate decreases during sleep, sleep duration can be estimated from the heart rate stored in the individual history data storage unit 25. Then, this sleep duration is classified as follows. If sleep duration is >8 hours, then you slept well. If 8 hours or more of sleep time > 6 hours, I was able to sleep normally. If 6 hours ≥ 2 hours of sleep, then you didn't sleep very well. If 2 hours ≥ sleep time > 0 hours, then you couldn't sleep. If 0 hours equals sleep time, I couldn't sleep at all.

[0082] Then, similar to the fatigue level classification, the sleepiness reference heart rate is adjusted according to the sleep level, with "slept well" being level 0, "slept normally" being level 1, "didn't sleep well" being level 2, "didn't sleep at all" being level 3, and "didn't sleep at all" being level 4.

[0083] Furthermore, regarding sleep, a screen like the one shown in Figure 12 may be displayed on the smartphone 1, allowing the subject to make a selection.

[0084] Next, let's explain the meal status. The meal status indicates whether or not a meal has been eaten recently (e.g., satiety level). In other words, it shows past meal information. This allows us to determine whether or not a meal has been eaten by detecting an increase in blood glucose levels. Since blood glucose levels are stored in the individual history data storage unit 25, the determination is made from that data.

[0085] Then, a correction value is added according to the time elapsed since the meal, as shown in Figure 13. For example, if it is within one hour after the meal, 3 is added to the read-out drowsiness threshold heart rate. The correction value is just an example and can be changed as needed through the settings.

[0086] Furthermore, the eating status may be determined from the subject's location information. For example, if the subject stays near a specific location (within a certain range) such as a restaurant for a certain period of time or longer, and their heart rate increases, it can be determined that they are eating. The subject's location information can be determined by acquiring information from the GPS receiver 12 of the smartphone 1. In other words, in this case, the individual history data storage unit 25 also stores the subject's past location information (stay history).

[0087] The meal detection flowchart will be explained with reference to Figure 14. The flowchart in Figure 14 may be performed by the correction unit of Server 2, or it may be performed on Smartphone 1 and only the results sent. If Server 2 performs the operation, it will either have map information and location information on its own or obtain it from an external source. If Smartphone 1 performs the operation, a map application or navigation application (navigation application) must be installed.

[0088] First, in step S11 of Figure 14, the current location is obtained. Next, in step S12, it is determined whether or not there is a restaurant within a certain range. If there is (YES), the duration of stay is obtained in step S13. On the other hand, if there is no restaurant within a certain range (NO), the process returns to step S11 (determining that no meal was eaten).

[0089] Next, in step S14, it is determined whether or not the user stayed for a certain period of time. If the user stayed for a certain period of time (YES), the heart rate is obtained in step S15. Then, in step S16, it is determined whether or not the heart rate has increased. If it has increased (YES), it is determined that the user ate a meal. On the other hand, if the user did not stay for a certain period of time in step S14 (NO), or if the heart rate did not increase in step S16 (NO), the process returns to step S11 (it is determined that the user did not eat a meal). In other words, staying for a certain period of time or longer at a facility where meals are eaten (a specific location) constitutes the user's past stay history.

[0090] Furthermore, adjustments may be made based on information from multiple past activities. For example, adjustments may be made by adding values ​​as shown in Figure 15, depending on both the fatigue level and sleep duration. In Figure 15, if the fatigue level is 4 and the sleep level is 4, 10 is added.

[0091] Furthermore, dietary habits may also be incorporated. For example, in Figure 15, correction values ​​for fatigue level and sleep status could be calculated, then a correction value for dietary habits could be calculated, and both correction values ​​could be added together.

[0092] The individual history data storage unit 25 stores individual history data such as heart rate, measurement time, activity status, etc., received by the terminal communication unit 22, and heart rate, blood glucose level, measurement time, etc., transmitted from the stationary measuring device 3. The server 2 may create or update the sleepiness criterion heart rate table 21 based on the history data stored in the individual history data storage unit 25.

[0093] Figure 16 shows a flowchart of the operation of the sleepiness calculation device 50 with the configuration described above. First, in step S21, the communication unit 15 sends the user ID and password, which are assigned to each user, from the smartphone 1 to the server 2. In step S31, the server 2 performs user authentication using the user ID and password sent from the smartphone 1.

[0094] Next, the communication unit 15 transmits the activity status determined by the status determination unit 11 from the smartphone 1 to the server 2. In step S32, the server 2 receives the activity status transmitted from the smartphone 1. In other words, this step functions as an activity status acquisition step.

[0095] Next, in step S33, server 2 acquires (reads) history data (heart rate, blood glucose level, etc.) stored in the individual history data storage unit 25. Then, in step S34, the correction unit 24 reads the sleepiness reference heart rate from the sleepiness reference heart rate table 21 based on the activity status and current time, corrects the read sleepiness reference heart rate based on the fatigue level etc. mentioned above, and sends the corrected sleepiness reference heart rate to smartphone 1. In other words, step S34 functions as a sleepiness reference heart rate acquisition step, an activity history acquisition step, and a correction step.

[0096] Next, in step S23, the smartphone 1 receives the corrected drowsiness reference heart rate from the server 2. Subsequently, in step S24, the drowsiness level calculation unit 16 calculates the drowsiness level and displays it on the drowsiness level display unit 17. In other words, step S24 functions as a drowsiness level calculation process. Then, in step S25, the history data (heart rate, measurement time, etc.) is sent to the server 2. In other words, this step functions as a biometric information acquisition process. Note that the history data may also be sent when transmitting the activity status in step S22.

[0097] Next, in step S35, server 2 stores the received history data in the individual history data storage unit 25.

[0098] Figure 17 shows a flowchart of the operation of step S34 (correcting and transmitting the sleepiness reference heart rate) in Figure 16. First, in step S41, the fatigue level and sleep time are calculated, and in step S42, a correction value is calculated from the fatigue level and sleep time based on the table shown in Figure 15.

[0099] Next, in step S43, it is determined whether or not the user has eaten based on the blood glucose level and location information as described above. If it is after a meal, the correction value calculated in step S43 is further increased by a correction value based on the table shown in Figure 13 (+α), and in step S45, the corrected drowsiness reference heart rate, which has been corrected using the correction value calculated in step S44, is calculated and transmitted. On the other hand, if it is not after a meal, in step S45, the corrected drowsiness reference heart rate, which has been corrected using the correction value calculated in step S42, is calculated and transmitted.

[0100] In the flowchart in Figure 17, the drowsiness threshold heart rate was corrected each time. However, for fatigue level and sleep duration, the correction values ​​could be determined on the morning of the day, and a drowsiness threshold heart rate table could be generated based on these correction values. Then, the drowsiness threshold heart rate could be read from the generated drowsiness threshold heart rate table based on the activity level and current time. For meals, the values ​​in the generated drowsiness threshold heart rate table could be corrected accordingly.

[0101] In this embodiment, the smartphone 1 uses an interface 18 to acquire the subject's heart rate and measurement time measured by the heart rate sensor 19, and a state determination unit 11 determines the subject's activity status and transmits it to the server 2. Next, the server 2 uses a correction unit 24 to acquire the subject's drowsiness criterion heart rate from a drowsiness criterion heart rate table 21, which stores the drowsiness criterion heart rate for each activity status, according to the activity status transmitted from the smartphone 1. Next, the correction unit 24 corrects the drowsiness criterion heart rate acquired by the correction unit 24 based on fatigue level, sleep duration, and meal status calculated from history data read from the individual history data storage unit 25. Then, the corrected drowsiness criterion heart rate (corrected drowsiness criterion heart rate) is transmitted to the smartphone 1, and the drowsiness level calculation unit 16 of the smartphone 1 calculates the subject's drowsiness level based on the corrected drowsiness criterion heart rate and the heart rate acquired by the interface 18. By doing so, the sleepiness threshold heart rate can be adjusted based on the subject's past activity level, allowing for highly accurate sleepiness calculations that take into account not only current activity but also past activity.

[0102] Furthermore, the correction unit 24 acquires information on the subject's past activity status, including fatigue level, sleep duration, and meal status. By doing so, the sleepiness reference heart rate can be corrected by taking into account the subject's fatigue level, sleep duration, and meal status. Therefore, the sleepiness level can be calculated taking these factors into account.

[0103] Furthermore, Server 2 adds or updates the sleepiness reference heart rate stored in the sleepiness reference heart rate table 21 based on the time the heart rate was measured and the subject's activity status determined by the status determination unit 11. In this way, the changes in the sleepiness reference heart rate for each time period due to the subject's daily activity status can be updated or added as needed.

[0104] Furthermore, the sleepiness level calculation unit 16 calculates the subject's sleepiness level based on sleepiness prediction parameters that are set in advance for each activity state. In this way, heart rate and heart rate variability can be weighted according to the activity state.

[0105] Furthermore, since it is equipped with a drowsiness level display unit 17, the subject can specifically perceive their own drowsiness and take appropriate action, such as taking a break or exercising, based on the displayed drowsiness level.

[0106] Furthermore, it is not necessary to obtain all three pieces of information—the subject's fatigue level, sleep duration, and dietary habits—but it is acceptable to obtain only one of them.

[0107] Furthermore, while we explained that location information can be used to determine whether or not a meal has just been eaten, this is not limited to post-meal assessments. For example, after staying at a relaxation facility such as a hot spring, tension is relieved, making one more likely to feel sleepy. Therefore, the sleepiness threshold heart rate may be adjusted if a person has stayed at a relaxation facility for a certain period of time or longer. Whether or not a person has stayed at a relaxation facility can be determined in the same way as the flowchart in Figure 14. In other words, having stayed at a relaxation facility (specific location) for a certain period of time or longer constitutes past stay history.

[0108] Furthermore, in the above-described embodiment, sleepiness prediction parameters and sleepiness reference heart rate were selected based on activity level and measurement time. However, sleepiness prediction parameters and sleepiness reference heart rate may also be selected based solely on activity level. However, it is preferable to also consider the measurement time, as this allows for the selection of appropriate sleepiness prediction parameters and sleepiness reference heart rate.

[0109] In addition to the drowsiness level display unit 17, the drowsiness level may also be notified by voice. Alternatively, voice notification may be given when the drowsiness level exceeds a certain level.

[0110] Furthermore, the server 2 may provide either or both of the parameter table 14 and the sleepiness level calculation unit 16, or the smartphone 1 may provide either or both of the sleepiness reference heart rate table 21 and the correction unit 24. Alternatively, the server 2 may not be used, and the smartphone 1 may provide all of these components. [Examples]

[0111] Next, a sleepiness calculation device according to the second embodiment of this model will be described with reference to Figures 18 to 20. Note that parts identical to those in the first embodiment described above are denoted by the same reference numerals and their descriptions are omitted.

[0112] In the first embodiment, the drowsiness threshold heart rate was corrected based on information about past activities, but in this embodiment, in addition to that, future activities are predicted and corrected accordingly. The configuration is the same as that shown in Figures 1 to 3. That is, the correction unit 24 functions as an activity prediction unit. In this embodiment, a smartphone 1 is used as an example, but as will be clear from the following description, a car navigation system may also be used.

[0113] For example, if a subject drives a car or other vehicle home, it is predicted that they will engage in relatively calm and relaxing activities such as resting and sleeping, which will ease their tension and make them more likely to feel sleepy. On the other hand, if they go on a trip to a leisure destination, it is predicted that they will engage in relatively active activities such as shopping and sightseeing, which will likely elevate their mood and make them less likely to feel sleepy.

[0114] Therefore, when a destination is set in the navigation app or the like on smartphone 1, that information is sent to server 2. Then, based on the destination, a correction is performed by adding a correction value, for example, as shown in Figure 18, to the corrected drowsiness criterion heart rate corrected in the first embodiment. Note that smartphone 1 may also send information such as "home" or "leisure spot" instead of destination information. In other words, terminal communication unit 22 functions as a destination information acquisition unit that acquires information about the destination the subject is heading to, and correction unit 24 predicts the subject's activity based on the destination information.

[0115] Furthermore, even when heading to a leisure destination, drowsiness is less likely at the start of the journey, but may occur at the halfway point. Therefore, in addition to the correction in Figure 18, it is also advisable to make a correction based on the distance from the halfway point to the destination, as shown in Figure 19. In Figure 19, 0% represents the starting point and 100% represents the destination. Alternatively, the correction can be made not only based on distance, but also based on the ratio of the time taken from the halfway point to the destination to the time taken from the starting point to the destination. In that case, the same correction can be made by changing the distance in Figure 19 to the time taken.

[0116] In this embodiment, the correction unit 24 predicts the subject's activity and corrects the drowsiness reference heart rate read from the drowsiness reference heart rate table 21 based on the predicted activity. In this way, the drowsiness reference heart rate can be corrected by taking into account not only past activities but also future plans, such as going out for leisure.

[0117] Furthermore, Server 2 obtains information about the subject's destination from Smartphone 1 via Terminal Communication Unit 22, and Correction Unit 24 predicts the subject's activities based on that destination information. In this way, it is possible to predict future plans, such as whether the destination is home or a trip.

[0118] In the above explanation, corrections were made in addition to the past activity status described in the first embodiment, but it is also possible to make corrections only to the future predictions described in this embodiment.

[0119] Furthermore, when driving a vehicle, one is more likely to become sleepy if the road being driven on is monotonous, and more alert if the road is complex, requiring concentration. Therefore, in addition to corrections based on past and future predictions, corrections may also be made based on the shape of the road being driven on, as shown in Figure 20.

[0120] Specifically, the terminal communication unit 22 acquires information regarding the subject's current location. Next, the correction unit 24 corrects the drowsiness reference heart rate acquired according to the subject's current activity status, based on the information regarding the current location, if the subject is driving a vehicle, as acquired by the terminal communication unit 22. As mentioned above, the status determination unit 11 can determine whether the subject is driving a vehicle.

[0121] Furthermore, the navigation app may change the route guided by the user based on the drowsiness level calculated by the smartphone 1. That is, the system may further include a guidance unit (navigation app) that searches for and guides the user to a pre-set destination, and the guidance unit may search for the route to the destination based on the drowsiness level calculated by the drowsiness level calculation unit 16. [Examples]

[0122] Next, a sleepiness calculation device according to the third embodiment of this model will be described with reference to Figures 21 and 22. Note that parts identical to those in the first and second embodiments described above are denoted by the same reference numerals and their descriptions are omitted.

[0123] This embodiment has the same basic configuration as the first embodiment (Figures 1 to 3). In the first embodiment, the level of drowsiness is calculated based on the heart rate measured by the heart rate sensor 19, but there may be discrepancies between the calculated value and the subject's perception. Therefore, in this embodiment, the subject subjectively evaluates the level of drowsiness displayed on the drowsiness level display unit 17, and the drowsiness reference heart rate is corrected (updated) based on that evaluation.

[0124] A specific example will be explained with reference to Figures 21 and 22. Figures 21 and 22 show examples of the display of the sleepiness level display unit 17, showing a graph 171 of heart rate changes and sleepiness reference heart rate, a graph 172 of sleepiness level, and the sleepiness level. The sleepiness level can be entered by touching the displayed value. In other words, in the examples of Figures 17 and 18, the sleepiness level display unit 17 is a touch panel.

[0125] In Figure 21, the basal heart rate for drowsiness is 74. The current level of drowsiness is calculated as 4 by the drowsiness level calculation unit 16. At this point, the subject subjectively evaluates their current level of drowsiness.

[0126] If, based on the subjective assessment, the subject rates their level of sleepiness as 2, they select the sleepiness level 2 on the input unit 9, as shown in Figure 21. The sleepiness reference heart rate correction unit 10 then corrects the sleepiness reference heart rate to match the subjective assessment. In Figure 21, it is corrected to 72. The sleepiness level is then recalculated based on the corrected sleepiness reference heart rate, and the sleepiness level graph 6b is also corrected (Figure 22).

[0127] Next, we will explain how the sleepiness reference heart rate is corrected in the sleepiness reference heart rate correction unit 10. Equation (2) explained in the first embodiment is transformed into the following equation (6). RR_ref=RR-(Db×HFS) / a...(6)

[0128] Therefore, by substituting the subjective evaluation value into D in equation (6), the corrected RR_ref can be calculated. Then, based on the corrected RR_ref, the level of sleepiness is recalculated using equation (2). Note that RR_ref and the sleepiness criterion heart rate HR_ref have the relationship shown in equation (4) explained in the first embodiment, so the sleepiness criterion heart rate HR_ref can be easily calculated by rearranging equation (4).

[0129] The corrected sleepiness threshold heart rate (HR_ref) is used to display the heart rate changes and sleepiness threshold heart rate in graph 6a, and is also sent to server 2 to correct the sleepiness threshold heart rate in the corresponding activity state and time in sleepiness threshold heart rate table 21. In addition, various correction values ​​other than sleepiness threshold heart rate may also be corrected. Since the correction values ​​described in the first and second embodiments may also vary from person to person, it is advisable to provide feedback on the results of subjective evaluations.

[0130] This embodiment includes a touch panel into which the subject's subjective evaluation of the sleepiness level displayed on the sleepiness level display unit 17 is input. The sleepiness level calculation unit 16 then corrects the sleepiness reference heart rate based on the subjective evaluation input via the touch panel and recalculates the sleepiness level based on the corrected sleepiness reference heart rate. In this way, the subject can subjectively evaluate the sleepiness level calculated by the sleepiness level calculation unit 16. The subjective evaluation result is then fed back into the sleepiness level calculation unit 16, which can then recalculate the sleepiness level. Therefore, information regarding sleepiness tailored to each individual subject can be calculated with greater accuracy.

[0131] It should be noted that the present invention is not limited to the embodiments described above. That is, those skilled in the art can implement the invention in various ways, without departing from the core principles, in accordance with conventionally known knowledge. As long as such modifications still incorporate the configuration of the sleepiness calculation device of the present invention, they are of course included within the scope of the present invention. [Explanation of symbols]

[0132] 1. Smartphone 2. Server (Information Processing Device) 3. Stationary measuring instruments 11. State determination unit (activity state acquisition unit) 12 GPS receiver (current location information acquisition unit) 13G sensor 14 Parameter Table 15 Communications Department 16. Sleepiness level calculation unit 17. Drowsiness level display unit (display unit, destination information acquisition unit, input unit) 18 I / F (Biometric Information Acquisition Unit) 19 Heart rate sensor 21. Sleepiness-Based Heart Rate Table (Sleepiness-Based Heart Rate Storage Section) 22 Terminal Communication Unit (Biometric Information Acquisition Unit, Activity Status Acquisition Unit, Current Location Information Acquisition Unit, Destination Information Acquisition Unit) 23 External Communications Department 24. Correction Unit (Sleepiness Reference Heart Rate Acquisition Unit, Activity History Acquisition Unit, Activity Prediction Unit) 25. Individual history data storage unit 50 Sleepiness Calculation Device S25 Receiving historical data (biometric information acquisition process) S22 Activity status acquisition (Activity status acquisition process) S23 Historical data acquisition (Activity history acquisition process) S24 Correct and transmit reference heart rate (sleepiness reference heart rate acquisition process, correction process) S14 Drowsiness level calculation (Drowsiness level calculation process)

Claims

[Claim 1] A biometric information acquisition unit that acquires biometric information related to the subject's heart rate, A fatigue level information acquisition unit that acquires fatigue level information regarding the fatigue level of the subject, A sleepiness level calculation unit calculates information regarding the subject's current sleepiness based on the aforementioned biological information and fatigue level information, A sleepiness calculation device characterized by being equipped with the following features.

Images