Prediction device

Abstract

To enable a proper prediction of a user state.SOLUTION: A prediction device includes: a first acquisition part that acquires a first state during sleep of a user; and a prediction part that predicts a second state during activity of the user from the first state and outputs the second state as prediction information.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to a prediction device and the like. [Background technology]

[0002] Systems for managing the condition of a user have been disclosed. For example, Patent Document 1 discloses an invention for managing the sleeping hours of a user. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2018-166641 A Summary of the Invention [Problem to be solved by the invention]

[0004] The present disclosure provides a prediction device and the like that is capable of appropriately predicting the state of a user. [Means for solving the problem]

[0005] The prediction device of the present disclosure includes a first acquisition unit that acquires biometric information and a first state of a user while sleeping, and a prediction unit that predicts a second state of the user while active from the biometric information and the first state and outputs the second state as predicted information. Effect of the Invention

[0006] According to the present disclosure, it is possible to appropriately predict the condition of a user. [Brief description of the drawings]

[0007] [Figure 1] FIG. 1 is a diagram illustrating an entire prediction system according to a first embodiment. [Diagram 2] FIG. 2 is a diagram illustrating a hardware configuration of a terminal device in the first embodiment. [Diagram 3] FIG. 2 is a diagram illustrating a hardware configuration of a first detection device in the first embodiment. [Figure 4] FIG. 2 is a diagram illustrating a hardware configuration of a second detection device in the first embodiment. [Diagram 5] FIG. 2 is a diagram illustrating a software configuration according to the first embodiment. [Figure 6] FIG. 2 is a diagram illustrating a software configuration according to the first embodiment. [Figure 7] FIG. 4 is a diagram illustrating an example of user information in the first embodiment. [Figure 8] FIG. 2 is a diagram illustrating an example of biological information in the first embodiment. [Figure 9] FIG. 4 is a diagram illustrating an example of first status information in the first embodiment. [Figure 10] FIG. 4 is a diagram illustrating an example of second status information in the first embodiment. [Figure 11] FIG. 4 is a diagram illustrating an example of prediction information in the first embodiment. [Figure 12] FIG. 2 is a diagram illustrating an example of a neural model in the first embodiment. [Figure 13] FIG. 11 is a diagram illustrating an example of a user information acquisition process in the first embodiment. [Figure 14] FIG. 11 is a diagram illustrating an example of a first status information acquisition process in the first embodiment. [Figure 15] FIG. 11 is a diagram illustrating an example of a second state information acquisition process in the first embodiment. [Figure 16] FIG. 2 is a diagram illustrating an example of a learning model generation process in the first embodiment. [Figure 17] FIG. 4 is a diagram illustrating an example of a prediction process in the first embodiment. [Figure 18] FIG. 4 is a diagram illustrating an operation example in the first embodiment. [Figure 19] FIG. 4 is a diagram illustrating an operation example in the first embodiment. [Figure 20] FIG. 4 is a diagram illustrating an operation example in the first embodiment. [Figure 21]FIG. 4 is a diagram illustrating an operation example in the first embodiment. [Figure 22] FIG. 13 is a diagram illustrating an entire prediction system according to a second embodiment. [Diagram 23] FIG. 11 is a diagram illustrating a hardware configuration of a server device in a second embodiment. [Figure 24] FIG. 11 is a diagram illustrating a software configuration according to a second embodiment. [Diagram 25] FIG. 11 is a sequence diagram illustrating a process flow in the second embodiment. [Figure 26] FIG. 11 is a diagram illustrating an operation example in the second embodiment. [Figure 27] FIG. 13 is a diagram illustrating an entire prediction system according to a third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] Hereinafter, one embodiment of the present invention will be described with reference to the drawings. Systems have been disclosed that manage and predict the user's physical condition and sleep state as a condition of the user, for example, a system that suggests bedtime and wake-up times based on the user's work schedule.

[0009] Further, for example, a technique is known in which a user's sleep status is grasped from the user's biometric information or a questionnaire, and a schedule is proposed for pre-sleep activities such as eating and bathing that affect sleep.

[0010] However, such technology only proposes improving sleep conditions based on general behavioral patterns, etc. For example, there has been no proposal from the perspective of what state the user will be in during the course of a day's activities, including during the day.

[0011] In view of such problems, the following embodiment describes a prediction device capable of predicting a user's daily condition based on the user's daily activities including sleep, biological information, etc. Note that the following embodiment is an example for realizing the invention described in the claims, and it goes without saying that the technical scope of the invention described in the claims is not limited to the embodiment.

[0012] [1. First embodiment] [1.1 Overall System] Fig. 1 is a diagram for explaining an overall outline of a prediction system 1. As shown in Fig. 1, the prediction system 1 has a terminal device 10 which is a prediction device. The prediction system 1 also includes a first detection device 20 and a second detection device 30 which can detect and acquire the state of a user.

[0013] The first detection device 20 is, for example, a device placed between the bed device 3 and the mattress 5. The first detection device 20 may be placed under the user P, for example, on the mattress 5, or directly on the floor or tatami mat on which the user P is located.

[0014] The first detection device 20 detects the state of the user P while he / she is asleep. In particular, the first detection device 20 detects the state of sleeping (first state) even while he / she is asleep. The second detection device 30 mainly detects the state of the user P while he / she is active (second state). In this embodiment, the time when the user is asleep refers to the time from when the user enters the bed to when he / she leaves the bed (the time from when he / she is in the bed to when he / she leaves the bed). The user may be asleep (asleep) or awake while he / she is asleep. The time when the user is asleep refers to the time from when the user falls asleep at night to when he / she wakes up the next morning. The time when the user is active refers to the time from when the user wakes up the morning of the day to when he / she falls asleep at night. Depending on the user's life rhythm, the user's activities may be reversed between day and night. In that case, the time when the user sleeps the most continuously in a day may be called sleeping, and the other times may be called active.

[0015] The first detection device 20 is a device that detects the state of the user P on the bed device 3. Therefore, the first detection device 20 mainly detects the sleeping state of the user P among the states during sleep (for example, the state of the user P during night time). To be precise, the time from when the user sits in bed until actually falling asleep and the time from when the user wakes up until when he or she gets out of bed are included in the time during sleep, but any state other than the active state may be simply called the sleeping state.

[0016] The second detection device 30 is, for example, a wearable terminal device that can be worn directly by the user P. The second detection device 30 is, for example, a wristband-type device, a watch-type device, or a glasses-type device. The second detection device 30 is a device that the user P wears after waking up, and mainly detects the active state of the user P (for example, the state from waking up to going to bed, the user's state during the day).

[0017] For the sake of convenience, these devices are described separately, but each description can also realize the function of the other device. For example, if the terminal device 10 has an acceleration sensor or the like, placing the terminal device 10 on the mattress 5 can realize the same function as the first detection device 20.

[0018] Furthermore, the second detection device 30 may function as the first detection device 20 by wearing the wearable device even while sleeping.

[0019] In this way, each device may function as another device depending on how it is used. Also, each device may be composed of multiple devices having further functions. For example, the second detection device 30 may include a pulse oximeter and a respiratory test device in addition to the wearable terminal device.

[0020] For example, when a user P is in bed on the mattress 5, the first detection device 20 detects body vibrations (vibrations emitted from the human body) as a biosignal of the user P. Then, based on the detected vibrations, bioinformation of the user P (e.g., information related to heart rate and breathing) is calculated. In this embodiment, values ​​calculated from the bioinformation (e.g., respiratory rate, heart rate, amount of activity) can be output and displayed as bioinformation values ​​of the user P.

[0021] Furthermore, since the terminal device 10 may be a general-purpose device, it is not limited to an information processing device such as a computer, and may be configured as a device such as a tablet or a smartphone. In this embodiment, a smartphone on which an application capable of realizing the functions of this embodiment is installed will be described as an example. Furthermore, the terminal device 10 is a device that can be connected to the first detection device 20 by wire or wirelessly. For example, the terminal device 10 may be an operation device (e.g., a hand switch, an operation remote control, etc.) that is connected to the first detection device 20 by wire.

[0022] The user may be a person undergoing medical treatment or a person in need of nursing care, or may be a healthy person not in need of nursing care, an elderly person, a child, a person with a disability, or even an animal other than a human.

[0023] Here, the first detection device 20 is formed in a sheet shape so as to be thin. As a result, even if it is placed between the bed device 3 and the mattress 5, it can be used without making the user P feel uncomfortable, and therefore it is possible to measure the biological information in bed for a long period of time. That is, when the user is lying down and resting, the biological information value or the like is acquired as the user's state (first state).

[0024] In addition, the first detection device 20 can determine the position and posture of the user P on the bed apparatus 3. For example, it can determine whether the posture of the user P on the mattress 5 is a recumbent position or a sitting position, and it can also determine in which position on the mattress 5 the user P is sleeping and in which direction the user P is sleeping.

[0025] The first detection device 20 only needs to acquire biosignals (body movement, respiratory movement, ballistocardiogram, etc.) of the user P. In this embodiment, the heart rate and respiratory rate can also be calculated based on body vibrations, but for example, an infrared sensor may be used for detection, the biosignals of the user P may be acquired from acquired images, or an actuator with a strain gauge may be used. In addition, by utilizing a built-in acceleration sensor, for example, the device may be realized by a smartphone, tablet, or the like (terminal device 10) placed on the bed device 3.

[0026] [1.2 Hardware configuration] [1.2.1 Common Configuration] Of the individual devices, the control unit, storage, ROM, RAM, display unit, operation unit, and communication unit will be described below.

[0027] The control unit (e.g., control unit 100, control unit 200, control unit 300) is a functional unit for controlling each device as a whole. The control unit realizes various functions by reading and executing various programs stored in storage or ROM. The control unit may be realized by one or more control devices / arithmetic units (CPU (Central Processing Unit), SoC (System on a Chip)).

[0028] Storage (for example, storage 110, storage 210, storage 310) is a non-volatile storage device capable of storing programs and data. For example, it may be configured as a storage device using an HDD (Hard Disk Drive) or a semiconductor memory. Furthermore, the storage may be an externally connectable storage device or storage medium, and may be configured as a USB memory, a CD-ROM drive, or a BD-drive. Furthermore, the storage may be, for example, a storage area on the cloud.

[0029] The ROM (for example, the ROM 120, the ROM 220, and the ROM 320) is a non-volatile memory that can retain programs and data even when the power is turned off. The ROM may store, for example, firmware for each device, initial applications, and the like.

[0030] RAM (e.g., RAM 130, RAM 230, RAM 330) is a main memory that is mainly used by the control unit of each device when executing processing. RAM is a rewritable memory that temporarily holds data including the storage of each device, programs read from the ROM of each device, and execution results.

[0031] The display unit (for example, display unit 140, display unit 240, display unit 240) is a device capable of displaying various information and an execution screen. The display unit may be a device capable of displaying information such as a liquid crystal display or an organic EL display, or may use a light-emitting element such as an LED. For example, the display unit may display the power state of the device or an error state depending on the lighting state of the LED. The display unit may also be configured by an external device. For example, the terminal device 10 may use a display device connected to a terminal capable of outputting video such as HDMI (registered trademark) or DVI.

[0032] Also, for example, the first detection device 20 may display the state of the device on the display unit 140 of the terminal device 10 instead of the display unit 240. In this case, the first detection device 20 does not need to be provided with the display unit 240.

[0033] The operation unit (e.g., operation unit 150, operation unit 250, operation unit 350) is a device through which a user operates each device. The operation unit may be realized by a hardware switch or a software switch. For example, a touch panel integrated with the display unit may be used to realize software keys as the operation unit. The operation unit may also be configured by an external device. For example, the terminal device 10 may use an operation device (e.g., a keyboard, etc.) connected to an interface such as a USB.

[0034] Also, for example, the first detection device 20 may use the operation unit 150 of the terminal device 10 to operate the device instead of the operation unit 250. In this case, the first detection device 20 does not need to be provided with the operation unit 250.

[0035] The communication unit (e.g., communication unit 170, communication unit 270, communication unit 370) is a communication interface that communicates with other devices. For example, it may be a network interface that can provide a wired connection or a wireless connection. In this embodiment, it is possible to communicate with other devices via a network NW. Furthermore, the communication unit may provide a function that allows connection to a mobile network (LTE (Long Term Evolution) / 4G / 5G / 6G).

[0036] 1.2.2 Terminal Equipment As shown in FIG. 2, the terminal device 10 includes a control unit 100, a storage 110 as a memory unit, a ROM 120, and a RAM 130, a display unit 140, an operation unit 150, and a communication unit 170.

[0037] The terminal device 10 may also have a camera unit and an acceleration sensor unit. The camera unit can capture, for example, the terminal device 10, the state of the user, and the environment around the user. By using the camera unit, the terminal device 10 can recognize the facial expression and the movement of the user.

[0038] Furthermore, the acceleration sensor unit can detect the acceleration occurring in the terminal device 10. For example, the terminal device 10 can obtain the activity state of the user by using the acceleration sensor unit.

[0039] [1.2.3 First detection device] As shown in FIG. 3, the first detection device 20 includes a control unit 200, a storage 210 as a memory unit, a ROM 220 and a RAM 230, a display unit 240, an operation unit 250, a sensor unit 260, and a communication unit 270.

[0040] Here, the display unit 240 and the operation unit 250 may be provided as necessary. The display unit 240 may be an element such as an LED that indicates the power-on state or the communication state. Instead of the display unit 240, an output device capable of outputting sound or voice (e.g., a piezoelectric buzzer, a speaker, etc.) may be provided.

[0041] The sensor unit 260 is a first sensor device capable of acquiring the body movement of the user. For example, the sensor unit 260 may be configured with a pressure sensor or a load sensor. The sensor unit 260 detects the body movement of the user. The sensor unit 260 may output, for example, the detected load value or a change in the load value.

[0042] The sensor unit 260 may be an external sensor device connected to the first detection device 20. For example, the sensor unit 260 may be a sensor device that detects the heartbeat of the user by an electrical signal. Also, for example, the change in load may be acquired from a load sensor provided in the bed apparatus 3.

[0043] The communication unit 270 may communicate with, for example, the terminal device 10. For example, the communication unit 270 may connect to the terminal device 10 using Bluetooth (registered trademark). In addition, the communication unit 270 may connect to a network by wireless LAN or the like and communicate with the terminal device 10 via the network, or may communicate with the terminal device 10 using a mobile communication network.

[0044] [1.2.4 Second detection device] As shown in FIG. 4, the second detection device 30 has a control unit 300, a storage 310 as a memory unit, a ROM 320 and a RAM 330, a display unit 340, an operation unit 350, a sensor unit 360, and a communication unit 370.

[0045] Here, the display unit 340 and the operation unit 350 may be provided as necessary. For example, in the case of a wristwatch-type wearable terminal device, the display unit 340 and the operation unit 350 integrated as a touch panel may be provided. Also, if a sensor is simply used, the display unit 340 and the operation unit 350 may not be provided.

[0046] The sensor unit 360 is a second sensor device used to acquire biometric information and an activity state of the user. For example, the sensor unit 360 may be configured with an optical sensor, a moisture detection sensor, a temperature sensor, an acceleration sensor, etc. That is, the sensor unit 360 may be a sensor such as a respiration sensor, an SpO2 sensor, a blood glucose sensor, a heart rate sensor, a blood pressure sensor, a sweat sensor, or a body temperature sensor.

[0047] The sensor unit 360 may be an external sensor device connected to the second detection device 30. For example, the sensor unit 360 may communicate with an external temperature sensor or heart rate sensor via the communication unit 370 and acquire the sensor values ​​output by each sensor.

[0048] [1.3 Software configuration] Next, the software configuration of the prediction system 1 will be described with reference to Fig. 5. Fig. 5 is a diagram that illustrates the software configuration of the terminal device 10, the first detection device 20, and the second detection device 30, and the data stored in the storage unit (storage).

[0049] [1.3.1 First detection device] The first detection device 20 functions as a biological information acquisition unit 202 and a first state determination unit 204 by the control unit 200 executing a program stored in the storage unit.

[0050] The biological information acquiring unit 202 acquires biological information based on a sensor value acquired by the sensor unit 260 .

[0051] For example, the sensor unit 260 may acquire the body vibration of the patient using a pressure sensor, or may acquire the biosignal from the change in the center of gravity position (body movement) of the patient using a load sensor. Alternatively, a microphone may be provided instead of a pressure sensor to acquire the biosignal based on the sound picked up by the microphone. Alternatively, the biosignal may be acquired based on the displacement of the living body or bedding using microwaves or laser speckles. Alternatively, the biosignal may be acquired from the output value of an acceleration sensor or a gravity sensor. It is sufficient that the sensor unit 260 can acquire the biosignal using any method.

[0052] The bioinformation acquiring unit 202 acquires values ​​related to the bioinformation of the user (such as respiration rate, heart rate, and activity level). For example, the respiration rate and heart rate may be obtained based on the respiration interval and heart rate interval by extracting the respiration component and heart rate component from the body movement (body vibration) acquired by the sensor unit 260. In addition, the periodicity of the body movement acquired by the sensor unit 260 may be analyzed (Fourier transform, etc.) and the respiration rate and heart rate may be calculated and acquired from the peak frequency.

[0053] The biometric information acquiring unit 202 may also acquire information (values) related to the biometric information from a device capable of acquiring other biometric information. The biometric information acquiring unit 202 stores the acquired biometric information in the biometric information storage area 212 in chronological order for each user.

[0054] FIG. 8 is a diagram showing an example of biometric information stored in the biometric information storage area 212. FIG. 8 shows an example of biometric information stored in the biometric information storage area 212. The biometric information storage area 212 stores the biometric information in chronological order in association with an ID (e.g., "PB010025") that is the user's identification information. The biometric information stores a heart rate (e.g., "58"), an SpO2 value (e.g., "99"), and a respiratory rate (e.g., "14") as examples of biometric information values ​​in association with a date and time (e.g., "2023 / 06 / 05 04:20:00").

[0055] Here, the bioinformation value may be stored at a predetermined time interval. The timing at which the bioinformation value is stored may be uniform or may vary depending on the bioinformation value. As an example, the bioinformation storage area 212 in FIG. 8 stores the heart rate every minute, and the SpO2 value and the respiratory rate every five minutes.

[0056] The first state determination unit 204 determines the state of the user mainly when the user is asleep among the states of the user. That is, the first state is the state of the user while the user is asleep (in bed apparatus 3), mainly at night, and is the state during the average time when the user sleeps. Moreover, the night determined by the first state may include night, late night, and early morning depending on the user.

[0057] The user's status determined by the first status determination unit 204 is stored in the first status information storage area 214. Fig. 9 is a diagram showing an example of the first status information stored in the first status information storage area 214. The first status information storage area 214 stores information on the user's first status in chronological order for each ID that is identification information of the user.

[0058] Here, the first state includes the following states: The user is out of bed. The user is in bed. The state of the user when the user is in bed. For example, the state of "awake" or "asleep". The state of "asleep" may also be represented by a level indicating the depth of sleep. For example, the state of light sleep may be represented as LV1, and the state of deepest sleep may be represented as LV4, and the state may be stored in stages. - User's posture. For example, the user's posture is "supine", "side-lying", or "prone"

[0059] The biometric information acquiring unit 302 calculates biometric information from the biometric signal of the user detected by the sensor unit 360, and stores the information in the biometric information storage area 312. When the sensor unit 360 is an optical sensor, the biometric information acquiring unit 302 can acquire biometric information such as heart rate, SpO2 value, and blood glucose level. In addition, the biometric information acquiring unit 302 may acquire blood pressure, sweating, body temperature, and the like. The biometric information acquiring unit 302 may also acquire other conditions of the user. For example, by using an acceleration sensor as the sensor unit 360, it may acquire, for example, that the user has coughed, sneezed, or fallen.

[0060] [1.3.2 Second detection device] The second detection device 30 functions as a biological information acquisition unit 302 and a second state determination unit 304 by the control unit 300 executing a program stored in the storage unit.

[0061] The biometric information acquiring unit 302 acquires biometric information based on the sensor value acquired by the sensor unit 360. For example, the biometric information acquiring unit 302 can acquire information such as a heart rate, a respiratory rate, an activity level, and an SpO2 value from the sensor value. The biometric information acquiring unit 302 may also store the acquired biometric information in a chronological order for each user in the biometric information storage area 312. Here, the data stored in the biometric information storage area 312 may be in the same format as that of the biometric information storage area 212 shown in FIG. 8, for example.

[0062] The second state determination unit 304 determines the state of the user mainly when the user is active. That is, the second state is the state of the user from waking up to going to bed (getting out of bed from the bed device 3), mainly during the daytime, and is the state during the average time when the user is active. In addition, the daytime determined by the second state may include morning (after waking up), evening, and night (before going to bed), depending on the user.

[0063] The user's status determined by the second status determination unit 304 is stored in the second status information storage area 314. Fig. 10 is a diagram showing an example of the second status information stored in the second status information storage area 314. The second status information storage area 314 stores information on the second status of the user in chronological order for each ID that is identification information of the user.

[0064] Here, the second state includes the following states: - A state where the user can "concentrate" and feel "drowsy" A state of "hypervigilance" in which the user is not sleepy but cannot concentrate In this embodiment, a state in which the user is not "drowsy" is described as a state in which the user can "concentrate", but there may be multiple other states.

[0065] For example, the following conditions may be determined as the user's condition: (1) A "normal" state as the state of the user. Here, the normal state of the user refers to a state in which the user is not in a state in which the user's concentration is particularly high, but is not in a state in which the user feels drowsy. For example, the normal state of the user may include a state in which the user is moving, or performing daily activities such as eating or taking a bath. (2) A state in which a user can concentrate compared to a "normal" state may be defined as a "concentrated" state. A "concentrated" state may be, for example, a state in which the user can achieve a better score than in a "normal" state when taking a test to check the user's state using an application or the like, or a state in which the user is aware that he or she can concentrate. Also, for example, a state in which the user's heart rate is slightly higher than normal may be defined as a "concentrated" state. (3) As described above, the user's state may be further determined to be a state of "drowsiness" or a state of "hyperarousal."

[0066] The second state may be indicated by a level. For example, a drowsy state may be indicated as LV1, a state where one can concentrate as LV2, and a state of hyperarousal as LV3.

[0067] The second state may indicate other states. For example, a state such as "moving" where the user is moving, or "in bed" where the user is in the bed apparatus 3 may be determined.

[0068] The second state determination unit 304 may determine another state as the second state. For example, the second state determination unit 304 may determine the second state as "coughing" when the user is coughing, as "taking a bath" when the user is taking a bath, or as "exercising" when the user is exercising.

[0069] Moreover, the second state determination unit 304 determines the second state of the user from, for example, a fluctuation in the heart rate of the user. For example, the following documents disclose the relationship between the fluctuation in the heart rate and drowsiness.

[0070] Sang-Ho Jo et al., "Heart Rate Change While Drowsy Driving" J Korean Med Sci (https: / / doi.org / 10.3346 / jkms.2019.34.e56)

[0071] For example, the above-mentioned literature states that "Driving HR showed a trend for increment as compared to day time mean HR, from 85 ± 5.6 to 89.8 ± 5.6 beats / min (by 7%) (P = 0.093). Mean HR while sleepy driving significantly decreased to 81.5 ± 9.2 beats / min by 9.3% ± 7.4% (P = 0.046)," indicating that when one feels "drowsy," there is a decrease in heart rate.

[0072] Therefore, if the acquired heart rate is within ±5% of the average heart rate, the second state determination unit 304 determines that the user is in a state where he or she can concentrate. Also, if the acquired heart rate is 5% or more lower than the average heart rate, the second state determination unit 304 determines that the user is in a sleepy state. Also, if the acquired heart rate is 5% or more higher than the average heart rate, the second state determination unit 304 determines that the user is not sleepy but cannot concentrate, that is, is in a state of hyperarousal.

[0073] The average heart rate described here is an example, and may be determined for each user. For example, the second state determination unit 304 may make the following determination based on the relationship between the heart rates. (1) If your current heart rate is 8% or more lower than your average heart rate, you are sleepy. (2) You can concentrate if your current heart rate is 5% to 10% higher than your average heart rate. (3) If your current heart rate is 10% or more higher than your average heart rate, you are in a state of hyperarousal. (4) Any other condition than the above is normal. For example, the heart rate is within ±5% of the average heart rate. It may be determined that:

[0074] The second state determination unit 304 may use other methods to determine the drowsiness of the user. For example, the second state determination unit 304 may determine drowsiness from the movement of the user's body, or may determine drowsiness from the state of blood flow (for example, parameters such as oxygenated hemoglobin, deoxygenated hemoglobin, and oxygen saturation). For determining drowsiness, for example, conventional technologies such as those described in JP 2008-35964 A, JP 2008-220424 A, JP 2018-27184 A, and JP 2021-18779 A can be used.

[0075] 1.3.3 Terminal Equipment The terminal device 10 functions as an information receiving unit 102, a user information acquiring unit 104, a sleep score calculating unit 106, a learning unit 107, and a prediction unit 108 by the control unit 100 executing a program stored in the storage unit. In addition, a user information storage area 111 for storing user information, a biological information storage area 112 for storing biological information, a first status information storage area 113 for storing first status information, a second status information storage area 114 for storing second status information, a prediction information storage area 115 for storing prediction information, and a learning model storage area 116 for storing a learning model are secured in the storage 110. In addition, these pieces of information are not limited to the storage 110, and may be stored in an area secured on a cloud service, for example.

[0076] The information receiving unit 102 acquires each piece of information from the first detection device 20 and the second detection device 30 via the communication unit 170. For example, the information receiving unit 102 receives biometric information from the first detection device 20 and the second detection device 30 and stores it in the biometric information storage area 112. Furthermore, the information receiving unit 102 receives first status information from the first detection device 20 via the communication unit 170 and stores it in the first status information storage area 113. Furthermore, the information receiving unit 102 receives second status information from the second detection device 30 via the communication unit 170 and stores it in the second status information storage area 114.

[0077] The user information acquiring unit 104 acquires information about a user. The user information acquiring unit 104 may acquire the user information from the user via, for example, the operation unit 150. The user information acquiring unit 104 may also acquire the user information from, for example, a server device connected via a network. The user information acquiring unit 104 may store the acquired user information in the user information storage area 111.

[0078] An example of the user information will now be described with reference to Fig. 7. Fig. 7 is a diagram showing an example of the user information stored in the user information storage area 111.

[0079] 7(a) stores basic information input by the user. The user information A stores, for example, an ID (e.g., "PB010025") for uniquely identifying the user, a name (e.g., "Suzuki Taro"), a date of birth (e.g., "2000 / 02 / 24"), a sex (e.g., "male"), a height (e.g., "170 cm"), a weight (e.g., "62.0 kg"), and information regarding illness (e.g., "none"). The contents included in the user information A are the contents input by the user via the operation unit 150.

[0080] The user information may also store information by day, such as the user information B in FIG. 7(b). As shown in FIG. 7(b), the user information B stores an ID (e.g., "PB010025"), a date (e.g., "2023 / 06 / 01"), and weight (e.g., "62.5 kg"), blood pressure (e.g., "122 / 70"), and body temperature (e.g., "36.2") measured on the date. For example, the information stored in the user information B may be information input by the user via the operation unit 150, or may be information acquired on an ad hoc basis from another device. For example, the terminal device 10 acquires weight from a weight scale, blood pressure from a blood pressure monitor, and body temperature from a thermometer, and stores them as user information. The user information B may also store information related to the life of the day. For example, in FIG. 7(b), information indicating "drinking alcohol" is stored as life information on the day.

[0081] The time when each piece of user information was acquired may be stored. When multiple pieces of user information were acquired on a given date, multiple pieces of user information may be stored.

[0082] The user information may also store information related to sleep, such as user information C in Fig. 7(c). As shown in Fig. 7(c), the user information C may store an ID (e.g., "PB010025"), a date (e.g., "2023 / 06 / 01"), a bedtime (e.g., "2023 / 05 / 31 22:10"), a wake-up time (e.g., "2023 / 06 / 01 06:45"), and a sleep score (e.g., "89").

[0083] The sleep score is a score for evaluating the quality of sleep based on the user's sleep state throughout the night. Here, the sleep score is information for evaluating sleep, and is calculated by the sleep score calculation unit 106.

[0084] The sleep score calculation unit 106 is a value that evaluates the sleep based on the state of sleep of the user. For example, the sleep score calculation unit 106 calculates a sleep score that evaluates the sleep from parameters such as the depth of sleep, the number of awakenings during sleep, and the sleep duration. Here, the method of calculating the sleep score may be, for example, the method disclosed in Japanese Patent Application Laid-Open No. 2013-45336 (Title of the Invention: Sleep State Evaluation Apparatus, Sleep State Evaluation System and Program, Filing Date: August 25, 2011).

[0085] The learning unit 107 outputs the generated learning model to the learning model storage area 116. The learning model (trained model) in this embodiment is a model that outputs a second state indicating the user's state of concentration, sleepiness, etc. in a day.

[0086] Here, the learning unit 107 may use various methods such as a machine learning model and a deep learning model as the learning model generated. Here, in the machine learning model, methods such as a support vector machine, decision algorithm, random forest, and K-nearest neighbor method may be used. In addition, in the deep learning model, methods such as a convolutional neural network (CNN) and a recurrent neural network (RNN) may be used.

[0087] For example, the learning unit 107 inputs necessary information (parameters) from the user information, the biological information, and the first state information to a neural network composed of multiple layers and neurons included in each layer as shown in FIG. 12. Each neuron receives signals from multiple other neurons and outputs a signal obtained by performing an operation to multiple other neurons. When the neural network has a multi-layer structure, the layers are called an input layer, an intermediate layer (hidden layer), and an output layer in the order in which the signals flow. Various operations (convolution operation, pooling operation, normalization operation, matrix operation, etc.) are performed on the neurons in each layer of the neural network, and the signals flow while changing their shape, and multiple signals are output from the output layer. The multiple output values ​​from the neural network are each output in association with the second state of the user. Then, the learning unit 107 performs a process of predicting (estimating) the state of the user at a certain point in time (for example, the second state such as concentration, drowsiness, or normal) in association with the largest output value among the output values ​​output from the neural network.

[0088] The parameters, which are coefficients used in various calculations of the neural network, are determined by inputting a large number of parameters and the second state of the user that is actually determined based on the parameters into the neural network in advance, propagating the error between the output value and the correct value backward through the neural network using the error backpropagation method, and repeatedly updating the parameters of the neurons in each layer. This process of updating and determining parameters is called learning.

[0089] The structure of the neural network and each of the individual calculations are publicly known techniques explained in books and papers, and any of these techniques may be used.

[0090] The prediction unit 108 predicts the user's state by applying the input parameters to the learning model generated by the learning unit 107. The prediction unit 108 of this embodiment applies, for example, biometric information, user information, first state information, and second state information to the learning model as parameters, and outputs prediction information that predicts the user's second state (for example, daytime state).

[0091] Furthermore, the prediction unit 108 may output a recommendation based on the prediction information. Examples of the recommendation output by the prediction unit 108 include the following.

[0092] (1) The prediction unit 108 displays the prediction information in a one-day timeline. For example, the prediction unit 108 outputs how the user's activities change in one day. (2) The prediction unit 108 outputs the time when the user can concentrate and the time when it is difficult for the user to concentrate. (3) When it is time for the user to concentrate, the prediction unit 108 notifies other devices. For example, if the second detection device 30 is a wristwatch-type device, the prediction unit 108 notifies the display unit of the second detection device 30 that it is time for the user to concentrate. (4) The prediction unit 108 outputs the amount of sleep the user needs and the ideal time to fall asleep (time to be in bed). (5) The prediction unit 108 outputs a report comparing the possible concentration time based on the prediction information with the actually measured concentration time.

[0093] [1.3.4 Alternative Configuration] Note that the specific state of the user may be determined by any device in the prediction system 1. For example, in Fig. 6, the first detection device 20 and the second detection device 30 only acquire biometric information.

[0094] Then, the first detection device 20 transmits the biometric information to the terminal device 10. The second detection device 30 transmits the biometric information to the terminal device 10. The first detection device 20 may further transmit a sensor value output by the sensor unit 260 to the terminal device 10, and the second detection device 30 may further transmit a sensor value output by the sensor unit 360 to the terminal device 10.

[0095] The terminal device 10 has a first state determination unit 103 and a second state determination unit 105. The first state determination unit 103 is the first state determination unit 204 described in FIG. 5. The second state determination unit 105 is the second state determination unit 304 described in FIG.

[0096] Furthermore, the biological information acquisition unit may also be realized by the terminal device 10. In this case, the first detection device 20 and the second detection device 30 may simply be sensor devices capable of outputting sensor values.

[0097] [1.4 Processing flow] The process flow in each device will be described below with reference to the drawings.

[0098] [1.4.1 User information acquisition process] FIG. 13 is an example of an operational flow showing the flow of a user information acquisition process executed by the control unit 100 (user information acquisition unit 104) of the terminal device 10. In FIG.

[0099] The control unit 100 acquires the attributes of the user (step S102). For example, the control unit 100 displays, on the display screen displayed on the display unit 140, an input screen on which the attributes of the user can be input.

[0100] Then, the user uses the input screen to input the user's attributes. Here, the user's attributes include, for example, information such as the user's name and age, as shown in Fig. 7(a). The user's attributes may also include the user's height, weight, etc.

[0101] Next, the user uses the input screen to acquire daily information as information that can be input by the user on a daily basis (Step S104). The daily information is parameters related to the user that change daily, and may be, for example, parameters such as those shown in FIG. 7(b). The daily information may also include measurements acquired from other devices (for example, a weight scale, a blood pressure monitor, a thermometer, etc.).

[0102] The control unit 100 stores the attributes of the user and the daily information of the user as user information in the user information storage area 111 (step S106).

[0103] [1.4.2 First status acquisition process] FIG. 14 is an example of an operational flow showing the flow of a first state acquisition process executed by the control unit 200 (first state determination unit 204) of the first detection device 20.

[0104] When the control unit 200 detects that the user is in bed (step S112), the control unit 200 acquires the biometric information by the biometric information acquisition unit 202 (step S114). The biometric information acquisition unit 202 continues to acquire the biometric information while the user is in bed. That is, the biometric information acquisition unit 202 continues to acquire the biometric information until step S124 in FIG. 14.

[0105] Then, control unit 200 executes a first state determination process (step S116) and determines first state information based on the first state of the user. In the first state determination process, control unit 200 executes first state determination unit 204 to determine the first state of the user. Note that first state determination unit 204 determines the first state of the user and outputs the first state information until the user wakes up. That is, first state determination unit 204 determines the first state and outputs the first state information up to step S124 in FIG. 14.

[0106] When the control unit 200 determines that the user has fallen asleep (step S118; Yes), it stores first state information based on the first state of the user in the first state information storage area 214. Here, the control unit 200 may determine that the user has fallen asleep when, for example, the first state of the user has been determined to be "sleeping" for a predetermined period of time or more, or when it detects that the user is not moving much (no significant body movement is detected).

[0107] When the user falls asleep, the control unit 200 stores the first status information of the user in the first status information storage area 214 (step S120). In addition, the control unit 200 stores the biometric information of the user in the biometric information storage area 212 (step S122).

[0108] Then, the control unit 200 executes steps S120 and S122 until the user wakes up (step S124; No). When the control unit 200 detects that the user has woken up (step S124; Yes), it executes a sleep evaluation process (step S126).

[0109] The sleep evaluation process may be executed in the terminal device 10. At this time, the terminal device 10 may execute the sleep evaluation process, for example, by executing the sleep score calculation unit 106. Furthermore, information related to sleep may be stored in the user information storage area 111 as user information.

[0110] 7(c), the control unit 100 may obtain the "bedtime" and "wake-up time" from the user's sleep onset / asleep state, and store them in the user information storage area 111. The control unit 100 may also store in the user information storage area 111 the sleep score calculated by executing the sleep score calculation unit 106.

[0111] [1.4.3 Second status acquisition process] FIG. 15 is an example of an operational flow showing the flow of the second state acquisition process executed by the control unit 300 (second state determination unit 304) of the second detection device 30.

[0112] The control unit 300 detects a trigger for starting acquisition of the second state (step S132). Here, the trigger for starting acquisition of the second state may be, for example, the following.

[0113] When the second detection device 30 detects that the device is worn by a user When the user performs an operation to start acquiring the second status When a different device (e.g., a sensor device) detects that the user has left the bed and determines to start acquiring the second state. When a preset or predicted start time arrives and it is determined that acquisition of the second state should begin

[0114] When any one of these triggers is detected (step S132; Yes), the biometric information acquisition unit 302 acquires the biometric information (step S134). Furthermore, the second state determination unit 304 executes a second state process (step S136). The second state determination process determines the second state as a state in which the user is active. Here, the second state determination process may determine, for example, "drowsiness," "concentration," "hyperalertness," or "normal" as the second state.

[0115] Then, the control unit 300 stores the biometric information in the biometric information storage area 312, and stores the second status information based on the second status in the second status information storage area 314 (step S138).

[0116] The control unit 300 detects a trigger for ending acquisition of the second state (step S140). Here, the trigger for ending acquisition of the second state may be, for example, the following.

[0117] When the second detection device 30 detects that the device has been removed from the user When a user action is detected to end acquisition of the second state When a different device (e.g., a sensor device) detects that the user is in bed and determines to end acquisition of the second state When a preset or predicted end time is reached and it is determined that acquisition of the second state is to end When the control unit 300 detects any one of these triggers (step S140; Yes), the control unit 300 ends this process.

[0118] [1.4.4 Learning model generation process] FIG. 16 is an example of an operational flow showing the flow of a learning model generation process executed by the control unit 100 (learning unit 107) of the terminal device 10.

[0119] The learning model is a model obtained by performing machine learning in which the sleep information includes user information, a first state indicating the state of the user while sleeping, and biological information of the user while sleeping, and the second state during the user's activity on the next day (daytime while awake). Here, the activity state of the second state is labeled, for example, as a "concentration" state, a "drowsiness" state, a "hyperarousal" state, or a "normal" state. The learning unit 107 may also perform semi-supervised machine learning to generate the learning model. For example, the learning unit 107 performs unsupervised machine learning for a predetermined period from when the user starts using the system (for example, until the learning unit 107 acquires a characteristic state and generates a learning model). Then, the learning unit 107 performs supervised machine learning after the predetermined period has elapsed (for example, after the learning unit 107 generates a learning model). The trained model may be, for example, a model that has learned a correlation between information (sleep information) regarding the user's sleep on a predetermined day and a second state during the user's activity on the next day of the predetermined day. As a result, by using the trained model, it is possible to predict the condition of the person on the day after a given day based on the sleep information of that day.

[0120] Moreover, the parameters inputted into the learning model are the user information, the first state information, and the biological information during sleep (the biological information acquired by the first detection device 20). Specific processing will be described below.

[0121] First, the control unit 100 (i.e., the learning unit 107) acquires user information from the user information storage area 111 (step S152). The control unit 100 also acquires biometric information from the biometric information storage area 112 (step S154). The control unit 100 also acquires first status information from the first status information storage area 113 (step S156). The control unit 100 also acquires second status information from the second status information storage area 114 (step S158).

[0122] Next, the control unit 100 performs data pre-processing on the acquired information (step S160). The data pre-processing is a process of weighting the information (parameters) acquired in steps S152 to S158 and identifying necessary parameters. Here, the control unit 100 does not necessarily need to acquire all of the information that can be acquired in steps S152 to S158. The control unit 100 acquires one or more pieces of information as parameters as necessary.

[0123] Here, the control unit 100 judges whether prediction information is stored in the prediction information storage area 115 (step S162). That is, the presence of prediction information in the prediction information storage area 115 means that the control unit 100 is outputting prediction information using a learning model, and a valid learning model is already stored in the learning model storage area 116.

[0124] Therefore, the control unit 100 acquires prediction information from the prediction information storage area 115 (step S164), and regenerates (re-learns) the learning model using the prediction information (step S166). At this time, the control unit 100 may re-learn the learning model using the user's condition as teacher data. The control unit 100 may acquire the user's condition determined from bio-information (e.g., heart rate, etc.) as teacher data as the user's condition. The control unit 100 may also acquire the user's condition inputted as teacher data as the user's condition.

[0125] In addition, when there is no prediction information in the prediction information memory area 115, the control unit 100 generates a learning model using the user information, biometric information, first status information, and second status information as parameters, since a valid learning model has not yet been generated (step S168).

[0126] For example, the control unit 100 uses one or more pieces of information among the user information, the biological information, and the first state information, and groups (clusters) and displays the second state information. The user provides information such as the user's state (e.g., "sleepy," "concentrated," "hyperalert," and "normal") for the grouped second states, and the control unit 100 generates a learning model. That is, the learning model is a model that has learned the correlation between the first state and the second state. In addition, the control unit 100 acquires the user's state by having the user take a test by an application during the time of the grouped second state. Then, the control unit 100 stores the generated learning model in the learning model storage area 116.

[0127] [1.4.5 Prediction Processing] FIG. 17 is an example of an operational flow showing the flow of prediction processing executed by the control unit 100 (prediction unit 108) of the terminal device 10.

[0128] The control unit 100 acquires user information from the user information storage area 111 (step S182). The control unit 100 also acquires biometric information from the biometric information storage area 112 (step S184). The control unit 100 also acquires first status information from the first status information storage area 113 (step S186).

[0129] Next, the control unit 100 predicts the second state information for the specified day by inputting the information acquired in steps S182 to S186 into the learning model, and stores the predicted information in the predicted information storage area 115 (step S188). Then, the control unit 100 outputs a recommendation based on the predicted information (step S190).

[0130] Here, the control unit 100 may output the recommendation to, for example, the display unit 140 or another device (for example, the display unit 340 of the second detection device 30). The control unit 100 may also output the recommendation by printing a report from a printer or the like.

[0131] [1.5 Example of operation] Next, an example of operation in this embodiment will be described. Fig. 18 is a diagram showing a comparison between the prediction information and the second state information.

[0132] The top graph is a graph based on the prediction information output by the terminal device 10 (prediction unit 108). Here, diagonal lines indicate "normal," black indicates "concentration," white indicates "drowsiness," and shading indicates "hyperarousal."

[0133] In the prediction information output by the terminal device 10, the state is predicted as "concentration" at time t1 and "drowsiness" at time t2, and the state is predicted as "normal" at the other times.

[0134] The graph below the prediction information shows the second state information based on the second state actually determined on the day predicted as the prediction information. In the second state determined by biological information measured during the user's activity, the predicted time t1 of concentration is consistent, but the drowsiness at time t2 is actually shorter at time t3.

[0135] Moreover, at time t4, which the terminal device 10 predicted as "normal," the state was determined as "concentration" in the actual second state. During time t4, the state of "hyperarousal" was determined at time t5. At time t6, the terminal device 10 predicted as "normal" with no drowsiness, but the state was determined as "drowsiness" in the actual second state.

[0136] By using such predicted values ​​and actual measured values, the terminal device 10 can perform re-learning. That is, the terminal device 10 performs re-learning using the second state information based on the actual measured values ​​as teacher data.

[0137] The terminal device 10 may output these prediction information as something that can be specifically determined, such as some biological information value. For example, the line graph at the bottom of Fig. 18 shows a state in which the heart rate is used to determine wakefulness and drowsiness. The line graph at the bottom of Fig. 18 is a graph that represents the rate of variation from the average heart rate in the positive and negative directions, with the average heart rate at the center.

[0138] That is, the terminal device 10 outputs the predicted heart rate as prediction information. The terminal device 10 predicts the time when the predicted heart rate is between 5% and 10% lower than the average heart rate as "concentration". The terminal device 10 predicts the time when the predicted heart rate is likely to be 8% or more lower than the average heart rate as "drowsiness". If the predicted heart rate is within ±5% of the average heart rate, the terminal device 10 predicts the time as "normal". The terminal device 10 may also predict the time when the predicted heart rate is likely to be 10% or more higher than the average heart rate as "hyperarousal".

[0139] Moreover, the terminal device 10 can determine the second state of the user by comparing the heart rate acquired by the second detection device 30 from the biological information acquisition unit 302.

[0140] The terminal device 10 (learning unit 107) can improve the accuracy of prediction by re-learning the learning model using the second status information, which is the actual measured value, as teaching data for the biological information and first status information during sleep and the prediction information output as a predicted value.

[0141] FIG. 19 is an example of a display screen W100 that displays the learning state displayed on the display unit 140 of the terminal device 10. As shown in FIG.

[0142] For example, the result of classifying the user's state during a day is displayed in region R100. Region R100 displays the classified time frames in regions R102 and R104.

[0143] In each of the classified regions R102 and R104, pull-down menus that allow the user to select and input the user's state are displayed as items K102 and K104. For example, as shown in item K104, the user can select the user's state for that time period from the pull-down menu. The displayed items may also display the state predicted by the control unit 100 in advance. Even if the state predicted in advance is displayed in the item, the user may reselect the correct state. Furthermore, the user does not need to make a selection (for example, "-") unless he or she is particularly able to concentrate or is sleepy. In this case, the time period may be determined to be a normal state.

[0144] 20 is an example of a display screen W110 that displays an operation screen of a reaction test executed on the terminal device 10. For example, the control unit 100 may present it to the user in order to determine the user's state. The display screen W110 may also be displayed when a function is called by the user. In other words, the reaction test is a test used to determine the user's state, and may be read and displayed by the user at any timing, or may be read and displayed by the control unit 100 for each predetermined condition.

[0145] FIG. 20 shows an example of a reaction test in which numbers are displayed in an area R110. The user is made to select numbers in ascending order, and the state of the user is estimated from the reaction time involved in the selection. For example, if the reaction time, which indicates the time it takes for the user to finish selecting all the numbers, is less than 5 seconds, the control unit 100 may determine that the user is in a "concentration" state. If the reaction time of the user is 10 seconds or more, the control unit 100 may determine that the user is in a "drowsy" state. If the reaction time of the user is within a predetermined time (e.g., less than 5 seconds) but the user makes a predetermined number of mistakes or more (e.g., three or more), the control unit 100 may determine that the user is in a "hypervigilance" state.

[0146] 21 is an example of a display screen W120 that displays recommendations based on predicted information displayed on the display unit 140 of the terminal device 10. The display screen W120 displays a second state of the user based on the predicted information in a display region R120 that indicates a time period.

[0147] For example, users can confirm that the time between 9:00 and 12:00 is the time when they can concentrate. Therefore, by allocating study, work, or other tasks that require concentration to this time period, users can spend their day more efficiently.

[0148] In this way, according to this embodiment, it is possible to predict the user's active state on the next day (today) based on the user's sleeping state on the previous day.

[0149] [2. Second embodiment] A second embodiment will be described. In the second embodiment, main processing is realized by a server device 40. Note that, in this embodiment, differences from the first embodiment will be mainly described, and descriptions of configurations and processing common to the first embodiment will be omitted.

[0150] [2.1 Overall structure] 22 is a diagram showing an entire prediction system 1B. The terminal device 10, the first detection device 20, and the second detection device 30 are capable of communicating with a server device 40 via a network NW.

[0151] [2.2 Functional configuration] 23 shows a hardware configuration of the server device 40. The server device 40 has a control unit 400, a storage 410 as a memory unit, a ROM 420 and a RAM 430, and a communication unit 470.

[0152] The explanation of each component is as explained in [1.2.1 Common Components] and is described in the relevant section.

[0153] FIG. 24 is a diagram for explaining the software configuration of the terminal device 10, the first detection device 20, the second detection device 30, and the server device 40. As shown in FIG.

[0154] The server device 40 functions as an information receiving unit 402, a sleep score calculating unit 406, a learning unit 407, and a predicting unit 408 by the control unit 400 executing a program stored in the storage unit.

[0155] The information receiving unit 402 receives user information from the terminal device 10, first status information from the first detecting device 20, and second status information from the second detecting device 30. In addition, the information receiving unit 402 receives biological information during sleep from the first detecting device 20 and biological information during activity from the second detecting device 30.

[0156] Then, the information receiving unit 402 stores the user information in a user information storage area 411, the biometric information in a biometric information storage area 412, the first status information in a first status information storage area 413, and the second status information in a second status information storage area 414.

[0157] In server device 40, sleep score calculation unit 406 calculates the sleep score. Furthermore, learning unit 407 generates learning data and stores it in learning data storage area 416. Furthermore, prediction unit 408 uses the learning data based on the parameters to output prediction information and stores it in prediction information storage area 415.

[0158] Note that the sleep score calculation unit 406 has the same configuration as the sleep score calculation unit 106 of the first embodiment, the learning unit 107 has the same configuration as the learning unit 107 of the first embodiment, and the prediction unit 408 has the same configuration as the prediction unit 108 of the first embodiment.

[0159] [2.3 Processing flow] The flow of processing in this embodiment will be described. Note that processing similar to the processing in Fig. 13 to Fig. 17 described in the first embodiment is executed by each device as necessary. Each processing has been described in the first embodiment, so detailed processing description will be omitted.

[0160] 25 is a sequence diagram showing an example of the flow of information between the devices. The terminal device 10 acquires user information (step S1002) and transmits the information to the server device 40 (S1004). The server device 40 acquires the user information from the terminal device 10 (S1006).

[0161] The first detection device 20 acquires biometric information (S1008) and acquires first status information (S1010). Then, the first detection device 20 transmits the biometric information and the first status information to the server device 40 (S1012). The server device 40 receives the biometric information and the first status information from the first detection device 20 (S1014).

[0162] The second detecting device 30 acquires the biometric information (S1020) and acquires the second status information (S1022). Then, the second detecting device 30 transmits the biometric information and the second status information to the server device 40 (S1024). The server device 40 receives the biometric information and the second status information from the second detecting device 30 (S1026).

[0163] The server device 40 generates a learning model based on the user information, the biometric information, the first state information, and the second state information. When the learning model has already been stored, the server device 40 further uses the prediction information and the second state information to regenerate the learning model (S1030).

[0164] Then, the server device 40 outputs prediction information using the received information and the learning model (S1032).Then, the server device 40 transmits the prediction information to the terminal device 10 (S1034).The terminal device 10 receives the prediction information from the server device 40 (S1036) and outputs a recommendation based on the prediction information (S1038).

[0165] [2.4 Example of operation] 26 is an example of a display screen W200 displaying recommendations on the terminal device 10. The terminal device 10 may generate a recommendation screen based on prediction information received from the server device 40 and display it.

[0166] Furthermore, the server device 40 may transmit information of the display screen W200 to the terminal device 10 without directly transmitting the prediction information. For example, the terminal device 10 accesses the server device 40 by using a browser or a dedicated application. The terminal device 10 displays the display screen W200 based on the screen information (for example, HTML data or XML data) transmitted from the server device 40.

[0167] The display screen W200 includes and displays several recommendations. For example, the display screen W200 displays in an area R202 the time periods when the user can concentrate and the time periods when the user feels sleepy. By referring to the area R202, the user can efficiently consider what time periods of the day to work on.

[0168] Area R204 of display screen W200 recommends a bedtime. For example, prediction unit 408 may predict an optimal bedtime using learning data to increase the amount of concentration time. The bedtime predicted by prediction unit 408 to maximize the concentration time is then displayed in area R204.

[0169] [3. Third embodiment] A third embodiment will be described. The third embodiment is an embodiment in which the terminal device 10 or the like communicates with a plurality of server devices 40, which is different from the second embodiment. Note that the present embodiment will be described mainly with respect to the differences from the first and second embodiments, and descriptions of configurations and processes common to the first and second embodiments will be omitted.

[0170] 27 is a diagram illustrating an entire prediction system 1C in the third embodiment. In the prediction system 1C, a plurality of terminal devices 10 and the like are connected to a network NW and are capable of communicating with a server device 40.

[0171] That is, in the first and second embodiments, the training data is generated based on information about the same user, but in this embodiment, training data is generated from multiple users.

[0172] As a result, for example, predictive information can be generated from learning data not only of the user using the terminal device 10, but also of user information that is the same or similar to that of the user (e.g., age, gender, medical history, lifestyle pattern, etc.).

[0173] Therefore, even if not much information about a user has been accumulated and prediction cannot be made from the learning data of that user, appropriate measurement information can be output. For example, in the case of a user who goes to bed after 1:00 a.m. every day, it is difficult to predict what the second state will be if the user goes to bed at 10:00 p.m.

[0174] However, by using learning data of users of the same age and sex, it is possible to output prediction information on what the second status information will be.

[0175] [4. Modifications] The present disclosure is not limited to the above-described embodiments, and various modifications are possible. In other words, embodiments obtained by combining technical means that are appropriately modified within the scope of the present disclosure are also included in the technical scope.

[0176] In addition, although the above-mentioned embodiments are described separately for convenience of explanation, they can be combined to the extent possible. In addition, the present invention intends to obtain rights to any of the technologies described in the specification through amendments or divisional applications, etc.

[0177] In addition, the programs that run on each device in each embodiment are programs that control the CPU and the like (programs that make a computer function) so as to realize the functions of the above-mentioned embodiments. Information handled by these devices is temporarily stored in a temporary storage device (e.g., RAM) during processing, and is then stored in various ROMs and HDDs, and is read, modified, and written by the CPU as necessary.

[0178] Here, the recording medium for storing the program may be any of semiconductor media (e.g., ROM, non-volatile memory cards, etc.), optical recording media, magneto-optical recording media (e.g., DVD (Digital Versatile Disc), CD (Compact Disc), BD (Blu-ray (registered trademark) Disc), etc.), magnetic recording media (e.g., magnetic tape, flexible disk, etc.), etc.

[0179] In addition, when distributing the program on the market, the program can be stored in a portable recording medium and distributed, or transferred to a server computer connected via a network such as the Internet. In this case, the storage device of the server device is of course included in the present invention.

[0180] Furthermore, the above-mentioned data may be stored in an external device and called up as needed, rather than being stored within the device. For example, the data may be stored in a network attached storage (NAS) or on the cloud.

[0181] The scope of the present disclosure is not limited to the configurations explicitly described in the specification, but includes combinations of the technologies disclosed in the present specification. The configurations of the present disclosure that are to be patented are described in the attached claims, but it is not intended to exclude them from the technical scope because they are not described in the claims.

[0182] In addition, in the above-mentioned specification, the statements "in the case of" and "when" are merely examples, and are not intended to limit the configuration to the described contents. Configurations other than these cases and times are also disclosed to the extent that would be obvious to a person skilled in the art, and the company intends to obtain rights to them.

[0183] In addition, the processes and data flows described in the specification are not limited to the order in which they are described. For example, the present application discloses configurations in which some parts of the processes are deleted or the order is changed, and the applicant intends to obtain the rights to such configurations.

[0184] Furthermore, although the functions described in the embodiments are executed by each device, they may be realized by one device, or an external server may be used.

[0185] In addition, each functional block or feature of the device used in the above-mentioned embodiment may be implemented or performed by an electric circuit, for example, an integrated circuit or a plurality of integrated circuits. The electric circuit designed to perform the functions described herein may include a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or a combination thereof. The general-purpose processor may be a microprocessor, or a conventional processor, controller, microcontroller, or state machine. The electric circuit described above may be composed of a digital circuit or an analog circuit. In addition, when a technology for integrated circuitization that replaces current integrated circuits appears due to the progress of semiconductor technology, one or more aspects of the present disclosure may use a new integrated circuit according to that technology. [Explanation of symbols]

[0186] 1. Prediction System 3 Bed equipment 5. Mattress 10 Terminal Equipment 100 control section; 110 storage; 120 ROM; 130 RAM 140 display unit; 150 operation unit; 170 communication unit 20 First detection device 200 control section; 210 storage; 220 ROM; 230 RAM 240 display unit; 250 operation unit; 260 sensor unit; 270 communication unit 30 Second detection device 300 control unit; 310 storage; 320 ROM; 330 RAM 340 display unit; 350 operation unit; 360 sensor unit; 370 communication unit 40 Server device 400 control unit; 410 storage; 420 ROM; 430 RAM 470 Communications Department

Claims

1. a first acquisition unit that acquires biological information of a user while sleeping and a first state of the user while sleeping; a second acquisition unit that acquires a second state of the user during the activity based on biometric information of the user during the activity; a learning model that has been machine-learned to output the second state when the biometric information and the first state are input; a control unit that predicts the second state by inputting the biological information and the first state into the learning model and outputs the second state as prediction information; A prediction device comprising:

2. the first state includes sleep information related to sleep; The second state includes a state of drowsiness. The prediction device according to claim 1 .

3. The sleep information includes at least one of bedtime, wake-up time, sleep duration, and sleep evaluation; The biological information includes information about the heart rate. The prediction device according to claim 2 .

4. The control unit inputs the biometric information and the first state into the learning model, and outputs the prediction information including the second state on a predetermined day different from the day used for the prediction information. The prediction device according to claim 1 .

5. The control unit, based on the prediction information, (1) The time when the user can concentrate and the time when it is difficult for the user to concentrate (2) The amount of sleep required by the user and the ideal time to fall asleep (3) A report comparing the time the user can concentrate and the measured time he or she can concentrate Output recommendations that include at least one of The prediction device according to claim 1 .

6. Further comprising a communication unit capable of communicating with a terminal device, The control unit notifies the terminal device via the communication unit when a time comes when the user can concentrate based on the prediction information. The prediction device according to claim 1 .

7. The control unit The learning model is retrained using the second state included in the forecast information on a predetermined day and the second state acquired by the second acquisition unit on the predetermined day as training data. The prediction device according to claim 1 .

8. The learning model is machine-learned using user information including attributes from a plurality of the users, biometric information of the users, and the first state; The control unit The second state is predicted by inputting the user's attributes, the biometric information, and the first state into the learning model, and the second state is output as predicted information. The prediction device according to claim 1 .

Images