Fatigue level estimation system, fatigue level estimation method, and program

Abstract

A fatigue level estimation system (100) for estimating an accumulated fatigue level, which is the level of fatigue that is likely to have accumulated in a user on a particular day, the fatigue level estimation system (100) comprising: an information acquisition unit (40) for acquiring information including at least one of a physiological amount and context information on an acquisition date that is not a holiday of the user and is earlier than the particular day; an accumulation calculation unit (12) for calculating the accumulated fatigue level on the basis of the acquired information; and a presentation unit (60) for presenting the calculated accumulated fatigue level together with the level of fatigue that has accumulated in the user on the same day of the week as the particular day in the past.

Description

Fatigue level estimation system, fatigue level estimation method, and program

[0001] The present disclosure relates to a fatigue level estimation system, a fatigue level estimation method, and a program.

[0002] There has been a need for estimating a person's fatigue level. For example, if a user working can know his or her own fatigue level, it would be useful information for dealing with the fatigue level, such as taking a rest at an appropriate time. Meanwhile, one way of thinking about fatigue is fatigue that accumulates daily due to continuous work, etc. (also known as subacute fatigue or diurnal fatigue). Such fatigue is estimated based on the amount of fatigue accumulated daily and how well the user is getting sleep to recover from fatigue. Patent Document 1 discloses a sleep index calculation device that calculates a sleep index from sleep time, fatigue level immediately before sleep, and fatigue level immediately after sleep.

[0003] Japanese Patent Application Laid-Open No. 2020-121035

[0004] However, in conventional fatigue level estimation systems, the fatigue accumulated by a user is based on daily tasks, and therefore the fatigue level may not be appropriately displayed when the user is not performing tasks. Therefore, the present disclosure provides a fatigue level estimation system that can display a more appropriate fatigue level.

[0005] In order to solve the above problem, a fatigue level estimation system according to one embodiment of the present disclosure is a fatigue level estimation system that estimates an accumulated fatigue level, which is the level of fatigue accumulated in a user on a target day, and includes an information acquisition unit that acquires information including at least one of physiological values ​​and context information on an acquisition date excluding a holiday of the user and prior to the target day, an accumulation calculation unit that calculates the accumulated fatigue level based on the acquired information, and a presentation unit that presents the calculated accumulated fatigue level together with the fatigue level accumulated in the user on a past day of the week that is the same as the target day.

[0006] In addition, a fatigue level estimation method according to one aspect of the present disclosure is a fatigue level estimation method that is executed by a computer and estimates an accumulated fatigue level, which is the level of fatigue accumulated in a user on a target day, and includes the steps of acquiring information including at least one of physiological values ​​and context information on an acquisition date excluding a holiday of the user and prior to the target day, calculating the accumulated fatigue level based on the acquired information, and presenting the calculated accumulated fatigue level together with the fatigue level accumulated in the user on the same day of the week as the target day in the past.

[0007] Furthermore, one aspect of the present disclosure can be realized as a program for causing a computer to execute the fatigue level estimation method, or as a computer-readable recording medium storing the program.

[0008] According to the present disclosure, a more appropriate fatigue level can be presented.

[0009] FIG. 1 is a block diagram showing the functional configuration of a fatigue level estimation system according to an embodiment. FIG. 2 is a diagram for explaining fatigue level estimation according to an embodiment. FIG. 3 is a diagram for explaining fatigue level estimation according to an embodiment. FIG. 4 is a diagram for explaining fatigue level estimation according to an embodiment. FIG. 5 is a flowchart showing a fatigue level estimation method according to an embodiment. FIG. 6 is a diagram for explaining an example of an acquisition date and a target date in fatigue level estimation according to an embodiment. FIG. 7 is a diagram for explaining a presentation example of the results of fatigue level estimation according to an embodiment. FIG. 8 is a diagram for explaining another example of an acquisition date and a target date in fatigue level estimation according to an embodiment.

[0010] (Knowledge that led to the present disclosure) As described in Patent Literature 1, there has been a need for estimating a person's fatigue level, particularly fatigue accumulated over a medium period, also known as subacute fatigue or diurnal fatigue, and various technological developments have been underway. In such estimations, the accumulated fatigue level is calculated based on daily tasks, such as work, learning, and tasks, which relate to a person's fatigue level. Therefore, it is common to provide a configuration necessary for calculating the fatigue level in the task environment. That is, it is common to acquire user information while the user is performing a task and estimate the user's accumulated fatigue level based on that information. However, when the user is not performing a task, it is difficult to acquire user information, and an appropriate fatigue level estimation may not be possible. In the present disclosure, fatigue level estimation is performed by calculating the accumulated fatigue level on a target day, for example, as of the morning. Therefore, it is necessary to acquire user information on an acquisition date prior to the target day and calculate the user's accumulated fatigue level (also referred to as accumulated fatigue level) on the target day. Here, days when it is difficult to acquire user information include days when the user is not performing a task, i.e., the user's days off. User information includes, for example, the user's physiological values ​​and context information (both of which will be described in detail later). Acquiring physiological and contextual information requires sensing using dedicated sensing devices such as a camera, smartphone, smartwatch, or wearable sensor that captures the user's image. However, on holidays, users are often out of camera range and often do not wear dedicated sensing devices such as smartphones, smartwatches, or wearable sensors. Therefore, it is difficult to acquire a user's physiological and contextual information on holidays. Instead of using user information on holidays, a system can calculate a user's accumulated fatigue level on holidays by having the user voluntarily input their activities on holidays. However, this input is cumbersome and requires the user to take voluntary actions, making it difficult to continue. As described above, when calculating a user's accumulated fatigue level, the presence of holidays reduces the accuracy of the calculated accumulated fatigue level.Furthermore, since there are times when a user's information on a holiday is used to calculate accumulated fatigue levels (days close to immediately after the holiday) and times when it is not (days close to immediately before the holiday), there is a possibility that the accuracy of the calculated accumulated fatigue levels will vary, and the fatigue levels may not be presented appropriately.

[0011] Therefore, the present disclosure calculates a user's accumulated fatigue level with a certain degree of accuracy using user information acquired on at least one acquisition date, including days excluding holidays, and presents the accumulated fatigue level together with the user's accumulated fatigue level on the same day of the week in the past, allowing the user to compare fatigue levels with similar accuracy and reliability. Here, the "same day of the week" preferably refers to a day of the week commonly used in the community to which the user belongs. Since users living in such a community are likely to live according to the cycle of the day of the week commonly used in that community, the accumulated fatigue level of a user for each day of the week (e.g., a seven-day calendar) is likely to follow the same trend (every seven days) according to the seven-day calendar. This is because many users alternate between holidays and weekdays in the same cycle based on the seven-day calendar. As a result, by presenting the accumulated fatigue level in the above manner, the user can easily understand the progress of their so-called diurnal fatigue by comparing it with the fatigue level on days with the same trend in the past. The present disclosure provides a fatigue level estimation system and the like that can present more appropriate fatigue levels in this way.

[0012] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. Note that the embodiments described below all show comprehensive or specific examples of the present disclosure. Therefore, the numerical values, components, the arrangement and connection of the components, steps, and the order of steps shown in the following embodiments are merely examples and are not intended to limit the present disclosure. Therefore, among the components in the following embodiments, components that are not described in the independent claims of the present disclosure will be described as optional components.

[0013] Furthermore, each figure is a schematic diagram and is not necessarily an exact illustration. Therefore, the scales and the like do not necessarily match in each figure. In each figure, the same reference numerals are used to denote substantially the same components, and redundant explanations will be omitted or simplified.

[0014] (Embodiment) [Configuration of Fatigue Level Estimation System] First, the functional configuration of a fatigue level estimation system 100 according to this embodiment will be described in detail with reference to Fig. 1. Fig. 1 is a block diagram showing the functional configuration of a fatigue level estimation system according to this embodiment.

[0015] 1 , the fatigue level estimation system 100 of the present embodiment includes a fatigue level estimation unit 10, an information acquisition unit 40, a sleep time estimation unit 50, and a presentation unit 60. These components constituting the fatigue level estimation system 100 may be integrated into a single housing or the like, or may be realized as multiple individual devices connected to each other via a communication line. The fatigue level estimation system 100 is realized by a computer including a presentation device such as a smartphone, a PC, or a tablet terminal, and by a processor, a memory, and a program executed using these.

[0016] The fatigue level estimation unit 10 includes a recovery calculation unit 11, an accumulation calculation unit 12, and an output unit 13. The fatigue level estimation unit 10 is installed in the fatigue level estimation system 100 as one of its functions, for example.

[0017] The recovery calculation unit 11 is a functional unit that calculates the degree of fatigue recovered according to the sleep duration (hereinafter simply referred to as the recovery fatigue degree). The recovery calculation unit 11 calculates the degree of fatigue recovered according to, for example, an input sleep duration. By inputting the user 99's actual sleep duration in the past into the recovery calculation unit 11, the degree of fatigue recovered from the sleep taken by the user 99 can be calculated. The recovery calculation unit 11 has, for example, multiple data tables and calculates the degree of fatigue recovered by referring to the data tables. Figure 2 is a diagram for explaining fatigue degree estimation according to the embodiment. Figure 2 shows an example of multiple data tables.

[0018] 2, the data table shows the degree of fatigue recovered from each hour of sleep (fatigue recovery level). In this embodiment, the fatigue level is described as being on a 100-point scale, with the higher the number, the greater the fatigue level. However, this number is merely an example, and other scales and resolutions may be used, and the significance of the numbers is not particularly limited.

[0019] The figure shows that a sleep duration longer than three hours but shorter than four hours results in a fatigue level of 40 being recovered (fatigue level is reduced by 40). The data table may be prepared, for example, for each age group, or may be prepared for each attribute value of the other user 99 attributes. For example, even if the sleep duration is the same, the recovery fatigue level may be higher for younger users and lower for older users. Furthermore, even if the sleep duration is the same, the recovery fatigue level may be lower for female users and higher for male users. The attributes of the user 99, including age, will be described later. In this way, the recovery calculation unit 11 calculates the fatigue level recovered according to the sleep duration based on information other than the sleep duration. While the recovery calculation unit 11 may use the data table as described above, it may also calculate the recovery fatigue level from the input sleep duration by inference using a trained model or a predetermined arithmetic formula.

[0020] The accumulation calculation unit 12 is a functional unit that calculates the accumulated fatigue level of the user 99 on a target day based on information about the user 99 on the acquisition date acquired by the information acquisition unit 40, which will be described later. The accumulation calculation unit 12 calculates the fatigue level of the user 99 based on information including at least one of the attributes of the user 99, physiological values ​​of the user 99, and context information of the user 99.

[0021] In this embodiment, fatigue level estimation system 100 estimates accumulated fatigue level by calculating the accumulated fatigue level as a result of accumulation and recovery from the previous day remaining at the reference point in time of the target day. At this time, fatigue level estimation system 100 stores the calculated accumulated fatigue level together with the day of the week to which the target day belongs in a storage unit or the like (not shown). When presenting the accumulated fatigue level, fatigue level estimation system 100 can present the stored past accumulated fatigue level together with the newly calculated accumulated fatigue level.

[0022] The accumulation calculation unit 12, for example, has multiple data tables and calculates the fatigue level to be recovered by referring to the data tables. FIG. 3 is a diagram for explaining fatigue level estimation according to an embodiment. FIG. 3 shows an example of multiple data tables. The data table in the figure shows a table in which the correlation between physiological amounts and context information is discretized. Specifically, FIG. 3 shows the correlation between the integrated value of LF / HF as a physiological amount and the step count as context information. It is known that the LF / HF value is low in a state of low stress (fatigue level is unlikely to increase) and high in a state of high stress (fatigue level is likely to increase). The figure shows that when the step count is more than 4,000 steps but less than 7,000 steps, if the integrated value of LF / HF is more than 30 but less than 40, a fatigue level of 50 is accumulated (fatigue level is increased by 50).

[0023] The data table may be provided for each age group, or may be provided for each attribute value for each of the other user 99 attributes. For example, even for the same combination of physiological quantity and context information, the accumulated fatigue level may be smaller for younger ages and larger for older ages. Furthermore, even for the same combination of physiological quantity and context information, the accumulated fatigue level may be larger for female gender and smaller for male gender. Attributes of the user 99, including age, will be described later. In this way, the accumulation calculation unit 12 calculates the accumulated fatigue level based on information (including attributes, physiological quantity, context information, etc.). While the accumulation calculation unit 12 may use a data table as described above, it may also calculate the accumulated fatigue level from input information (including attributes, physiological quantity, context information, etc.) by inference using a trained model or by a predetermined arithmetic formula.

[0024] The accumulation calculation unit 12 further subtracts the recovery fatigue level from the accumulated fatigue level before deduction calculated as described above, thereby applying recovery equivalent to the recovery fatigue level and calculating the accumulated fatigue level, which is the estimated result at the reference point in time.

[0025] The output unit 13 is a functional unit that outputs the calculated accumulated fatigue level. The accumulated fatigue level output here includes the fatigue level accumulated at the reference time point on the target day, as well as the accumulated fatigue level at the same reference time point on the same day of the week as the target day in the past. For example, if the accumulated fatigue level is calculated on a weekly basis, the accumulated fatigue level can be output as the change in accumulated fatigue level from week to week.

[0026] The information acquisition unit 40 is a functional unit that acquires information including the attributes, physiological values, context information, etc. of the user 99 on an acquisition date that is in the past relative to the target date. Due to activities on such an acquisition date, the user 99 has accumulated fatigue as of the target date. The information acquisition unit 40 acquires information about the user 99 on a past acquisition date in order to calculate the accumulated fatigue level of the user 99 as of the target date. The information acquisition unit 40 is connected via a communication module to various modules corresponding to the information, such as an input device and a memory unit for acquiring attributes, a sensor module and a calculation unit (e.g., a processor) for acquiring physiological values, and a sensor module and a calculation unit (e.g., a processor) for acquiring context information. The information acquisition unit 40 is capable of acquiring information about the past user 99 on the target date. Specifically, the information acquisition unit 40 accesses a memory unit (not shown) in which information about the past user 99 is stored to acquire the information, or acquires information about the past user 99 one by one and stores it in an accessible memory unit (not shown).

[0027] Physiological quantities are the results of sensing and the like, and are actual measured values ​​used to calculate fatigue levels. Physiological quantities include, for example, facial images (which can be used for pulse wave analysis and emotion analysis from facial feature positions), electrocardiograms (heart rate information), sweat rate, skin temperature, pupil diameter, etc. For example, when LF / HF is calculated from the RR interval of an electrocardiogram and an LF / HF integrated value is acquired as a physiological quantity, the information acquisition unit 40 is connected to an electrocardiograph and a calculation unit, calculates LF / HF from the RR interval, and acquires the LF / HF integrated value. Furthermore, context information is information that provides a background interpretation for calculating fatigue levels to the results of sensing and the like, and is an actual measured value. The context information includes information about the user's exercise, activity, and surrounding environment, such as the number of steps, activity level, season, date and time, and ambient temperature. For example, when a step count is acquired as context information, the information acquisition unit 40 is connected to an acceleration sensor, a GPS receiver, and a calculation unit, calculates the number of steps, and acquires the total step count for that day. The menstrual flow and context information may be acquired throughout the entire day of the acquisition date, or may be acquired by sampling a predetermined time period for multiple time periods within the day. Specifically, the menstrual flow and context information for the acquisition date may be the average values ​​of the menstrual flow and context information for each of a first time period (e.g., 9:00 to 10:00), a second time period (e.g., 12:00 to 13:00), a third time period (e.g., 15:00 to 16:00), and a fourth time period (e.g., 18:00 to 19:00) of the day. Alternatively, the menstrual flow and context information for the acquisition date may be the sum of the values ​​obtained by multiplying the menstrual flow and context information for each of the first, second, third, and fourth time periods by a weighting coefficient that increases with the later time period (the closer the time period is to the target day).

[0028] Additionally, the information acquisition unit 40 may acquire, for example, a facial image, sweat rate, skin temperature, pupil diameter, etc. Sweat rate can be measured using a moisture sensor worn by the user 99. Stress increases sweat rate. Skin temperature can be measured using a thermal image sensor that captures an image of the user 99 to measure nose skin temperature and forehead skin temperature, and the stress level can be calculated from the difference between the nose skin temperature and the forehead skin temperature. Stress reduces nose skin temperature relatively. Pupil diameter can be measured using a visible image sensor that captures an image of the user 99. It is known that pupil diameter constricts when stressed. The information acquisition unit 40 may also acquire activity level, season, date and time, ambient temperature, etc. Similar to step counts, fatigue can be quantified from the amount of activity. Since people are less likely to get tired in a comfortable environment around 20°C and more likely to get tired at ambient temperatures above 30°C, using the season, date and time, ambient temperature, etc. allows for more accurate quantification.

[0029] The context information includes information about sleep duration. In other words, the information acquisition unit 40 is also a functional unit that acquires the sleep duration of the user 99. The information acquisition unit 40 may acquire the sleep duration input by the user 99 to the fatigue level estimation system 100, or may acquire the user 99's average sleep duration. For example, the information acquisition unit 40 acquires the sleep duration during the nighttime from the acquisition date to the next day based on the average actual sleep duration of the user 99 over a predetermined number of past days (e.g., 1 day, 2 days, 3 days, 5 days, 7 days, or 30 days). When estimating the planned sleep duration, the predetermined number of past days may be changed taking into account attributes of the day, such as the day of the week or a public holiday, or days to be excluded from the predetermined number of past days may be set. The information acquisition unit 40 may also acquire the average sleep duration of people of the same age group.

[0030] The sleep time estimation unit 50 is a functional unit that estimates sleep time from information related to sleep time when the information related to sleep time included in the context information does not directly indicate sleep time. For example, a description will be given assuming that the charging time of a mobile device (e.g., a smartphone) owned by the user 99 is acquired as information related to sleep time included in the context information. The sleep time estimation unit 50 estimates the user's sleep time based on the acquired charging time. Specifically, the sleep time estimation unit 50 estimates the user's sleep time based on the start and end times of charging. FIG. 4 is a diagram for explaining fatigue level estimation according to an embodiment. FIG. 4 shows the transition of the charge amount of the mobile device of the user 99 over a portion of one day. As shown in FIG. 4, the charge amount of the mobile device starts to increase after 11:00 PM and reaches 100% within a few hours. The point at which this increase in charge amount begins, i.e., the start time of charging the mobile device, is close to the user's bedtime. The time at which the charge amount of the mobile device starts to increase can be determined as the sleep start time.

[0031] 4, the charge level of the portable terminal begins to decrease at 6:00. The point at which this decrease in charge level begins, i.e., the time at which charging of the portable terminal ends, is close to the wake-up time of the user 99. The time at which the charge level of the portable terminal begins to decrease can be taken as the wake-up time. Using the bedtime and wake-up time, it is possible to estimate the sleep time of the user 99. This allows for a simple and relatively accurate estimation of sleep time without using a dedicated tracking device.

[0032] It should be noted that since the user 99 may not have gone to bed yet even when the charge amount starts to increase, the sleep time may be calculated by subtracting the time difference (for example, 30 minutes) until the user 99 actually goes to bed.

[0033] This time may be set based on attribute information such as age and gender. Attribute information may be information indicating, for example, the user 99's age, gender, place of residence, family structure, occupation, etc. The time difference until actual bedtime may be a long time, such as one hour, for a young age group, such as those in their teens or twenties. The time difference until actual bedtime may also be changed depending on the time at which the charge amount begins to increase. For example, if the charge amount begins to increase before 8:00 PM, the time difference may be a longer time, such as 1.5 hours. The above-mentioned threshold time of 8:00 PM may also be set (changed) based on the sunset time at the user 99's place of residence. Furthermore, for a user 99 with the attribute of being a parent with children, the user 99 is likely to fall asleep immediately after starting to charge the mobile device, and the time difference until actual bedtime may be zero. Furthermore, for a user 99 with the attribute of being a parent with children, the user 99 is likely to fall asleep immediately after starting to charge the mobile device, and the time difference until actual bedtime may be zero.

[0034] The sleep duration is used to ensure the fatigue level at the reference point. Specifically, by inputting the acquired sleep duration into the recovery calculation unit 11, the fatigue level recovered during that sleep duration can be calculated. This recovery fatigue level is accurate because it is based on the sleep duration. In other words, instead of using a sleep duration based on uncertain past trends as the sleep duration, a recovery fatigue level based on more accurate sleep duration is used. This allows a more accurate fatigue level to be calculated as the fatigue level accumulated at the reference point, and therefore the fatigue level estimation system 100 can present an accurate fatigue level. Note that the mobile device may be a computer in which the fatigue level estimation system 100 is installed.

[0035] The presentation unit 60 is a functional unit that presents the accumulated fatigue level as an estimation result on, for example, a mobile terminal (for example, a smartphone) owned by the user 99. A specific example of presentation by the presentation unit 60 will be described later.

[0036] [Operation of Fatigue Level Estimation System] Next, the operation of the fatigue level estimation system 100 will be described with reference to Figures 5 to 8. Figure 5 is a flowchart showing a fatigue level estimation method according to an embodiment. As shown in Figure 5, when the operation of the fatigue level estimation system 100 starts, the information acquisition unit 40 acquires information about the user 99 on the acquisition date (step S101). If the acquired information about the user 99 includes attributes of the user 99, the settings (data table, inference model, or calculation formula) of the recovery calculation unit 11 and the accumulation calculation unit 12 are changed to match the attributes of the user 99.

[0037] Here, the acquisition date will be described. FIG. 6 is a diagram illustrating an example of the acquisition date and the target date in fatigue level estimation according to the embodiment. The acquisition date includes at least the days before the target date, excluding the user's holidays. Specifically, the acquisition date is, for example, a three-day period including the day before the target date, the day before, and the day before that, and the day before that. In FIG. 6, each row indicates a combination of the target date and the acquisition date for calculating one accumulated fatigue level. Also, in FIG. 6, the holidays for user 99 are Saturday and Sunday.

[0038] In FIG. 6 , grids with numbers entered indicate acquisition dates, and the numbers indicate the weighting coefficients of the information of user 99 on those acquisition dates. As shown in FIG. 6 , if the target day is Friday, the acquisition dates are Tuesday, Wednesday, and Thursday. Similarly, if the target day is Thursday, the acquisition dates are Monday, Tuesday, and Wednesday. Similarly, if the target day is Wednesday, the acquisition dates are Sunday, Monday, and Tuesday. Similarly, if the target day is Tuesday, the acquisition dates are Saturday, Sunday, and Monday. Similarly, if the target day is Saturday, the acquisition dates are Wednesday, Thursday, and Friday. Note that the acquisition dates may be only one day, the day before the target day, or may be only two days, the day before the target day and the day before the day before.

[0039] Alternatively, the number of acquisition days may be changed for each day of the week of the target day. For example, if user 99's holidays are Saturday and Sunday, if the target day is Saturday, the acquisition days may be five days, Monday, Tuesday, Wednesday, Thursday, and Friday, if the target day is Friday, the acquisition days may be four days, Monday, Tuesday, Wednesday, and Thursday, if the target day is Thursday, the acquisition days may be three days, Monday, Tuesday, and Wednesday, if the target day is Wednesday, the acquisition days may be two days, Monday and Tuesday, and if the target day is Tuesday, the acquisition day may be one day, Monday.

[0040] By setting the target days to three, it is possible to ensure that one acquisition date is included, avoiding the two holidays, thereby ensuring a certain degree of reliability in the accuracy of the calculated accumulated fatigue level. Furthermore, the weighting coefficient is larger for acquisition dates closer to the target date among the multiple acquisition dates (here, three days) and smaller for acquisition dates further from the target date among the multiple acquisition dates (here, three days). Since the more recent fatigue is more likely to accumulate (remain) on the target date, the weighting coefficient is set to obtain an accumulated fatigue level that reflects this actual situation. Returning to FIG. 5 , the recovery calculation unit 11 calculates the recovery fatigue level from the sleep time (step S102), and the accumulation calculation unit 12 calculates the accumulated fatigue level based on information about the user 99, more specifically, the user 99's physiological values ​​and context information, and the recovery fatigue level to be subtracted (step S103). Here, the calculation of the accumulated fatigue level when using a weighting coefficient is described.

[0041] As described above, the weighting coefficient is used to increase the influence of the user 99's information on the accumulated fatigue level the closer the acquisition date is to the target date, and decrease the influence of the user 99's information on the accumulated fatigue level the further the acquisition date is from the target date. Specifically, the weighting coefficient is multiplied by the values ​​of the physiological quantities and context information in the acquired information, and the values ​​of all the physiological quantities and context information multiplied by the weighting coefficients for the multiple acquisition dates are summed. For example, assume that the multiple acquisition dates are the three days immediately preceding the target date (the day before, the day before, and the day before the day before). If the step count for the day before was 6,000 steps, the step count for the day before was 7,000 steps, and the step count for the day before the day before was 8,000 steps, then the total is 6,000 x 0.6 + 7,000 x 0.3 + 8,000 x 0.1 = 6,500 steps. Then, when calculating the accumulated fatigue level for the target day, the step count of 6,500 steps can be used to refer to the data table shown in FIG. 3. Similarly, if the integrated value of LF / HF for the previous day is 40, the integrated value of LF / HF for the day before that is 40, and the integrated value of LF / HF for the day before that is 30, the result is 40 x 0.6 + 40 x 0.3 + 30 x 0.1 = 39. When calculating the accumulated fatigue level for the target day, this integrated value of LF / HF of 39 can be used by referring to the data table shown in FIG.

[0042] This weighting factor is also used to calculate the recovery fatigue level based on sleep duration. For example, if the user slept 3 hours the previous day, 4 hours the day before, and 5 hours the day before that, the result is 3 x 0.6 + 4 x 0.3 + 5 x 0.1 = 3.5 hours. When calculating the recovery fatigue level for the target day, the 3.5-hour sleep duration is used to calculate the recovery fatigue level by referring to the data table shown in FIG. 2 . This calculated recovery fatigue level is then subtracted from the accumulated fatigue level before subtraction to calculate the accumulated fatigue level that takes the weighting factor into account. Note that the weighting factor may be calculated by dividing 1 by the number of days in the multiple acquisition dates. This allows the user 99's information on each of the multiple acquisition dates to affect the accumulated fatigue level evenly (on average). While specific numerical values ​​are shown as examples of the weighting factor here, the impact of the accumulated fatigue level of each acquisition date on each target day varies greatly from person to person. For example, the impact of the accumulated fatigue level of each past acquisition date on each target day may be examined for each individual, and the weighting factor may be optimized for each individual based on the results.

[0043] 5 , the output unit 13 outputs the calculated accumulated fatigue level together with the accumulated fatigue level at the same reference point in time on the same day of the week as the target day. The presentation unit 60 presents the accumulated fatigue level of the user 99 together with the accumulated fatigue level at the same reference point in time on the same day of the week as the target day (step S104).

[0044] FIG. 7 is a diagram illustrating an example of the presentation of fatigue level estimation results according to an embodiment. FIG. 7 shows the presented accumulated fatigue level when Friday is the target day and the presented accumulated fatigue level when Thursday is the target day. FIG. 7 also shows the presented accumulated fatigue level when Monday is the target day for reference. In practice, the accumulated fatigue levels for multiple target days are not presented together as in FIG. 7 . Instead, for example, the accumulated fatigue level for one target day selected by the user 99 or selected by the fatigue level estimation system 100 is presented. Since the accumulated fatigue level for the first day of the user 99's holiday has the highest reliability of accuracy and the accumulated fatigue level for the day after the user 99's holiday has the lowest reliability of accuracy, the fatigue level estimation system 100 may select and present the accumulated fatigue level in descending order of reliability of accuracy. Alternatively, the accumulated fatigue levels for multiple target days may be presented together as in FIG. 7 . In this case, the accumulated fatigue level for the day after the holiday, i.e., Monday, may also be presented for reference.

[0045] As shown in FIG. 7 , in this embodiment, the accumulated fatigue level of the user 99 on the same day of the week as the target day in the past is presented along with the newly calculated accumulated fatigue level, allowing the user 99 to grasp the trend in accumulated fatigue level over a week at a glance. The accumulated fatigue levels of the user 99 on the same day of the week as the target day in the past have the same fatigue tendency, which has the advantage of being easy to compare. Furthermore, for example, if the accumulated fatigue level for each day (consecutive days) is presented as a trend, a relatively unreliable day, such as Monday, would be displayed alongside a relatively reliable day, such as Friday. In this case, it is difficult to determine whether the difference between the fatigue level on Monday and the fatigue level on Friday is due to the accumulation of fatigue level over the course of the day or due to variations in the accuracy of the Monday fatigue level. In other words, the fatigue level on Monday may appear to decrease, and the fatigue level may appear to rapidly increase even though the actual fatigue level has not increased significantly toward Friday, or vice versa. In this way, the influence of holidays, which have a low accuracy reliability, on the accumulated fatigue level for each day of the week is less apparent, making it difficult to grasp the trend in accumulated fatigue level. Therefore, by presenting the accumulated fatigue level for each week, which has the same fatigue trend and has the same reliability of accuracy as described above, it is easier to understand the progress of the accumulated fatigue level, and a more appropriate fatigue level can be presented.

[0046] In this embodiment, for example, if Saturday and Sunday are holidays, the reliability of accuracy between the accumulated fatigue level calculated on Friday and the accumulated fatigue level calculated on Monday will differ, and the reliability of accuracy on Monday will be relatively low. Therefore, for example, comparing the accumulated fatigue level calculated on Monday with the accumulated fatigue level calculated on Friday will not be meaningful. This takes into account the above-mentioned viewpoint. Since the comparison is between days of the week with the same reliability of accuracy, the accumulated fatigue level presented as a comparison target for the newly calculated accumulated fatigue level becomes meaningful, allowing the user to accurately grasp the progress of the accumulated fatigue level and its magnitude relationship. In particular, for example, for a person who works weekdays from Monday to Friday, the reliability of accuracy tends to increase as the week progresses (Friday), so comparing the accumulated fatigue level of any day from Tuesday to Saturday with the accumulated fatigue level of the same day has the effect of accurately grasping the progress of the accumulated fatigue level.

[0047] 7, the accumulated fatigue level on the day after a holiday, i.e., Monday, is small because the accumulated fatigue level is low immediately after the holiday, and the holiday, which has low accuracy, has a large effect, making it very difficult to grasp the progress of the accumulated fatigue level. If the acquisition date includes a holiday, not only is it difficult to obtain information about the user 99, but the user 99's sleep time is also likely to be irregular. For example, even if the sleep time is estimated based on the charging time of the smartphone, it is difficult to obtain an accurate sleep time, and the accuracy of the recovery fatigue level and, ultimately, the accumulated fatigue level will be less reliable.

[0048] Therefore, as another example of the embodiment, a method for presenting an accumulated fatigue level that allows a certain degree of understanding of the progress of the accumulated fatigue level by combining the weighting coefficient and a plurality of acquisition dates will be described. Fig. 8 is a diagram illustrating another example of the acquisition date and the target date in the fatigue level estimation according to the embodiment. As in Fig. 6, Fig. 8 shows the acquisition date and the weighting coefficient for each acquisition date when the target day is Monday, i.e., the target day is the day after a holiday.

[0049] As shown in Figure 8, in this example, when Monday is set as the target day, multiple acquisition dates are set so that at least two non-holiday acquisition dates are included, including two holidays. Specifically, three acquisition dates are set for two holidays. Furthermore, rather than increasing the weighting coefficient for days closer to the target day and decreasing the weighting coefficient for days further from the target day as described above, the weighting coefficient is set extremely small when the acquisition date is a holiday and extremely large when the acquisition date is not a holiday. This allows the influence of non-holiday acquisition dates, which have a relatively high reliability in the accuracy of the accumulated fatigue level, to be increased even when Monday is set as the target day, thereby enabling the accumulated fatigue level to be calculated with a certain degree of accuracy and reliability.

[0050] In addition, a dedicated data table for calculating the accumulated fatigue level for the day after a holiday (data table corresponding to Figures 2 and 3) may be prepared as a data table for calculating the accumulated fatigue level for the day after a holiday.

[0051] [Effects, etc.] As described above, the fatigue level estimation system 100 according to the first embodiment of the present invention is a fatigue level estimation system 100 that estimates an accumulated fatigue level, which is the level of fatigue accumulated in a user on a target day, and includes an information acquisition unit 40 that acquires information including at least one of physiological amounts and context information on an acquisition day excluding the user's holidays and prior to the target day, an accumulation calculation unit 12 that calculates the accumulated fatigue level based on the acquired information, and a presentation unit 60 that presents the calculated accumulated fatigue level together with the fatigue level accumulated in the user on the same day of the week as the target day in the past.

[0052] According to this fatigue level estimation system 100, the newly calculated accumulated fatigue level can be evaluated relatively to the fatigue level accumulated in the user 99 on the same day of the week as the target day in the past. The fatigue level accumulated in the user 99 on the same day of the week as the target day in the past is likely to be based on the same fatigue trend as the accumulated fatigue level newly calculated on the target day, and since it is calculated similarly influenced by the user 99's days off, the reliability of accuracy is equivalent. Therefore, it is easy to understand the trend of the accumulated fatigue level by evaluating the newly calculated accumulated fatigue level relatively to the fatigue level accumulated in the user 99 on the same day of the week as the target day in the past. In this way, a more appropriate fatigue level can be presented.

[0053] Furthermore, the fatigue level estimation system 100 according to the second embodiment is the fatigue level estimation system 100 according to the first embodiment, and the target day includes the first day of a holiday and any one of the five days before the holiday.

[0054] This allows the accumulated fatigue level to be calculated for target days including the first day of a holiday and any of the five days before the holiday. In other words, the accumulated fatigue level can be calculated for any target day except for the one day following the holiday. This reduces the influence of noise due to holidays, allowing the progress of accumulated fatigue levels on a weekly basis to be accurately understood.

[0055] Furthermore, the fatigue level estimation system 100 according to the third aspect of this embodiment is the fatigue level estimation system 100 according to the second aspect, in which the holidays are Saturdays and Sundays in the seven-day week, and the target days include any of Saturdays, Tuesdays, Wednesdays, Thursdays, and Fridays in the seven-day week.

[0056] This allows the accumulated fatigue level to be calculated for target days that include any of Saturday, Tuesday, Wednesday, Thursday, and Friday. In other words, the accumulated fatigue level can be calculated for any target day excluding holidays other than the first day (Sunday) and the day after a holiday (Monday).

[0057] Furthermore, the fatigue level estimation system 100 according to a fourth aspect of the present embodiment is the fatigue level estimation system 100 according to any one of the first to third aspects, and the context information includes information regarding the amount of sleep the user has had from the acquisition date to the next day.

[0058] According to this, the accumulated fatigue level can be calculated using the sleeping time of the user 99.

[0059] A fatigue level estimation system 100 according to a fifth aspect of the present embodiment is the fatigue level estimation system 100 according to any one of the first to fourth aspects, in which the physiological data includes a facial image of the user on the acquisition date.

[0060] According to this, the facial image of the user 99 can be used as the physiological quantity.

[0061] Furthermore, the fatigue level estimation system 100 according to a sixth aspect of the present embodiment is the fatigue level estimation system 100 according to any one of the first to fifth aspects, in which the physiological values ​​include the user's heart rate information on the acquisition date.

[0062] According to this, the heart rate information of the user 99 can be used as the physiological amount.

[0063] In addition, a fatigue level estimation method according to a seventh aspect of this embodiment is a fatigue level estimation method that is executed by a computer and estimates an accumulated fatigue level, which is the level of fatigue accumulated in a user on a target day, and includes the steps of: acquiring information including at least one of physiological values ​​and context information on an acquisition date excluding the user's holidays and prior to the target day (step S101); calculating the accumulated fatigue level based on the acquired information (step S103); and presenting the calculated accumulated fatigue level together with the fatigue level accumulated in the user on the same day of the week as the target day in the past (step S104).

[0064] This can achieve the same effects as the fatigue level estimation system 100 described above.

[0065] A program according to an eighth aspect of the present embodiment is a program for causing a computer to execute the fatigue level estimation method according to the seventh aspect.

[0066] This makes it possible to achieve the same effects as those of the fatigue level estimation system 100 described above using a computer.

[0067] (Other Embodiments) While the fatigue level estimation system, fatigue level estimation method, and program according to the present disclosure have been described above based on the above-mentioned embodiments, the present disclosure is not limited to the above-mentioned embodiments. For example, the present disclosure also includes forms obtained by applying various modifications to the embodiments that would occur to those skilled in the art, and forms realized by arbitrarily combining the components and functions of the embodiments within the scope of the present disclosure.

[0068] Furthermore, for example, the present disclosure can be realized not only as a fatigue level estimation system, but also as a program including, as steps, the processes performed by each component of the fatigue level estimation system, and as a computer-readable recording medium on which the program is recorded. The program may be pre-recorded on the recording medium, or may be supplied to the recording medium via a wide area communication network including the Internet.

[0069] In other words, the above-described comprehensive or specific aspects may be realized as a system, an apparatus, an integrated circuit, a computer program, or a computer-readable recording medium, or may be realized as any combination of a system, an apparatus, an integrated circuit, a computer program, and a recording medium.

[0070] REFERENCE SIGNS LIST 10 Fatigue level estimation unit 11 Recovery calculation unit 12 Accumulation calculation unit 13 Output unit 40 Information acquisition unit 50 Sleep time estimation unit 60 Presentation unit 99 User 100 Fatigue level estimation system

Claims

1. A fatigue level estimation system that estimates an accumulated fatigue level, which is the level of fatigue accumulated in a user on a target day, comprising: an information acquisition unit that acquires information including at least one of physiological amounts and context information on an acquisition date excluding holidays of the user and prior to the target day; an accumulation calculation unit that calculates the accumulated fatigue level based on the acquired information; and a presentation unit that presents the calculated accumulated fatigue level together with the fatigue level accumulated in the user on the same day of the week as the target day in the past.

2. The fatigue level estimation system according to claim 1, wherein the target days include the first day of the holiday and any one of the five days before the holiday.

3. The fatigue level estimation system according to claim 2, wherein the holidays are Saturdays and Sundays in a seven-day week, and the target days include any of Saturdays, Tuesdays, Wednesdays, Thursdays, and Fridays in a seven-day week.

4. The fatigue level estimation system according to claim 1, wherein the context information includes information regarding the amount of sleep the user had from the acquisition date to the next day.

5. The fatigue level estimation system according to claim 1, wherein the physiological data includes a facial image of the user on the acquisition date.

6. The fatigue level estimation system according to claim 1, wherein the physiological amount includes the user's heart rate information on the acquisition date.

7. A fatigue level estimation method executed by a computer for estimating an accumulated fatigue level, which is the level of fatigue accumulated in a user on a target day, comprising the steps of: acquiring information including at least one of physiological values ​​and context information on an acquisition date excluding a holiday of the user and prior to the target day; calculating the accumulated fatigue level based on the acquired information; and presenting the calculated accumulated fatigue level together with the fatigue level accumulated in the user on the same day of the week as the target day in the past.

8. A program for causing the computer to execute the fatigue level estimation method according to claim 7.

Images