Cognitive Function Judgment System

Abstract

To provide a cognitive function determination system capable of easily detecting abnormality of a cognitive function.SOLUTION: A cognitive function determination system includes: a power sensor 110 capable of detecting actions of a subject P1 by making it possible to detect use situations of a facility provided in a building (a residence 1) the subject P1 uses; and a server 120 determining whether or not there is abnormality to the cognitive function of the subject P1 by comparing actual action data related to actions of the subject P1 detected by the power sensor 110 with learning action data related to actions of the subject P1 who learns in advance.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a technology of a cognitive function determination system capable of detecting an abnormality in cognitive function.

Background Art

[0002] Conventionally, a technology of a cognitive function determination system capable of detecting an abnormality in cognitive function has been known. For example, it is as described in Patent Document 1.

[0003] Patent Document 1 describes a dementia risk determination system (cognitive function determination system) including a biological data detection sensor capable of detecting biological data (cerebral blood flow data, heartbeat data, pulse wave data, respiration data, body movement data, etc.) of a subject, and a dementia risk determination device that compares the biological data of the subject with data related to symptoms of dementia to determine the onset risk of dementia.

[0004] The biological data detection sensor described in Patent Document 1 is lent to the subject for a certain period, and the subject detects biological data using the biological data detection sensor by their own operation. Specifically, the subject irradiates near-infrared light from the biological data detection sensor to the forehead (forehead), etc., and detects biological data by the reflected wave. Based on the biological data thus obtained, the onset risk of dementia is determined.

[0005] However, it is cumbersome for the subject to perform a special examination (detection of biological data) to determine the onset risk of dementia as in Patent Document 1. For this reason, it is expected that there will be a certain number of subjects who avoid such examinations, which is not preferable from the viewpoint of early detection of dementia.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0007] The present invention has been made in view of the above circumstances, and the problem to be solved is to provide a cognitive function determination system capable of easily detecting an abnormality in cognitive function.

Means for Solving the Problems

[0008] The problem to be solved by the present invention is as described above. Next, means for solving this problem will be described.

[0009] That is, in claim 1, by being able to detect the usage status of facilities provided in the building used by the subject, an action detection unit capable of detecting the actions of the subject, execution action data regarding the actions of the subject detected by the action detection unit, and learned action data regarding the actions of the subject learned in advance are compared, and a cognitive function determination unit that determines the presence or absence of an abnormality in the cognitive function of the subject, A plurality of is provided. a notification unit capable of notifying a plurality of notification targets; is equipped with and the cognitive function determination unit stores, as a plurality of the facilities, a first facility belonging to a first classification preliminarily classified according to the usage frequency in a predetermined period and a second facility belonging to a second classification having a lower usage frequency than the first classification, determines the presence or absence of an abnormality in the cognitive function of the target person regarding the first facility in a first period, calculates the number of detections of an abnormality in the cognitive function for each action of the target person regarding the first facility in a second period longer than the first period, and when the number of detections of an abnormality in the cognitive function regarding the first facility in the second period is equal to or greater than a predetermined threshold, issues an abnormality warning for the cognitive function to the notification target using the notification unit, determines the presence or absence of an abnormality in the cognitive function of the target person regarding the second facility in the second period, calculates the number of detections of an abnormality in the cognitive function for each action of the target person regarding the second facility in a third period longer than the second period, and when the number of detections of an abnormality in the cognitive function regarding the second facility in the third period is equal to or greater than a predetermined threshold, issues an abnormality warning for the cognitive function to the notification target using the notification unit is what it is.

[0010] In claim 2, the cognitive function determination unit performs an action time determination process for determining that an abnormality has occurred in the cognitive function of the subject when the time zone in which the subject performs a predetermined action is deviated from the learned time zone by a predetermined time or more, or when the subject does not perform the learned action.

[0011] In claim 3, the cognitive function determination unit performs a sleep time determination process for determining that an abnormality has occurred in the cognitive function of the subject when the time zone in which the subject sleeps is deviated from the learned time zone by a predetermined time or more, or when the subject is not sleeping.

[0012] In claim 4, in the sleep time determination process, the cognitive function determination unit determines the degree of abnormality in cognitive function according to the increase or decrease in the sleep time of the subject with respect to the learned sleep time.

[0013] In claim 5, when the number of times the subject has performed a predetermined action within a predetermined period is different from the learned number of times by a predetermined number of times or more, the cognitive function determination unit performs action count determination processing for determining that an abnormality has occurred in the cognitive function of the subject.

Advantages of the Invention

[0015] As an effect of the present invention, the following effects are achieved.

[0016] In claim 1, an abnormality in the cognitive function can be easily detected. In addition, a warning can be appropriately notified to the notification target.

[0017] In claim 2, an abnormality in the cognitive function can be easily detected.

[0018] In claim 3, an abnormality in the cognitive function can be easily detected.

[0019] In claim 4, a more detailed determination of the cognitive function can be performed.

[0020] In claim 5, an abnormality in the cognitive function can be easily detected.

Brief Description of the Drawings

[0022]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Mode for Carrying Out the Invention

[0023] Hereinafter, with reference to FIG. 1, the configuration of the cognitive function determination system 100 according to an embodiment of the present invention will be described.

[0024] The cognitive function determination system 100 determines whether there is an abnormality in the cognitive function of the subject P1. In this embodiment, as an example, it is assumed that an elderly person is the subject P1 and the presence or absence of an abnormality in the cognitive function of the elderly person is determined. The cognitive function determination system 100 mainly includes a power sensor 110, a server 120, and a terminal 130.

[0025] The power sensor 110 detects the power consumption of various facilities used by the subject P1 (especially facilities used in daily life). The power sensor 110 is provided in the distribution board 2 of the building (residence 1) where the subject P1 lives. The power sensor 110 can detect the power for each branch circuit of the distribution board 2. Thereby, the power sensor 110 can detect the usage status (whether it is being used or not) of the facilities connected to each branch circuit. By detecting the usage status of the facilities, the behavior of the subject P1 (which facilities are being used) can be indirectly detected.

[0026] In this embodiment, as an example of equipment for which power consumption is detected by the power sensor 110, cooking equipment (IH, microwave oven, etc.), refrigerator, television, dryer, vacuum cleaner, washing machine, air conditioner, heating appliance, lighting, etc. are assumed. For example, when the use of cooking equipment is detected, indirectly, it is detected that the subject P1 has used the cooking equipment (and thus has eaten). Also, when the use of the television is detected, indirectly, it is detected that the subject P1 has watched the television.

[0027] Note that the method of detecting the usage status of various equipment is not limited to the method of detecting the power for each branch circuit by the power sensor 110, and various methods can be used. For example, it is also possible to detect the power of the main circuit (main power) of the distribution board 2 and identify and grasp the equipment in use by analyzing the waveform of the power. Also, instead of detecting the power of the distribution board 2, it is also possible to directly detect the operating status of various equipment itself (such as the on / off of the power supply of various equipment, the power of the outlet to which various equipment is connected, etc.). Also, when a system for grasping (managing) the usage status of various equipment (for example, HEMS: Home Energy Management System, etc.) is provided in the house 1, it is also possible to use the information grasped by that system.

[0028] The server 120 performs various processes based on the detection results of the power sensor 110. The server 120 is constituted by, for example, a virtual server (cloud server) provided on the cloud. The server 120 can grasp the usage status of various equipment by acquiring information from the power sensor 110. The server 120 can determine the presence or absence of an abnormality in the cognitive function of the subject P1 based on the usage status of various equipment. Also, the server 120 can transmit and receive various information to and from the terminal 130 described later.

[0029] The terminal 130 is capable of displaying various types of information. The terminal 130 is possessed by a person who should grasp the presence or absence of abnormalities in the cognitive function of the target person P1 (for example, family members, relatives, etc. of the target person P1). In the present embodiment, it is assumed that the family member P2 of the target person P1 possesses the terminal 130. The terminal 130 is constituted by, for example, a device (such as a smartphone or a tablet terminal) that can be carried by the family member P2 of the target person P1. The terminal 130 can notify the family member P2 of the target person P1 of the information from the server 120 by an appropriate method (display on a liquid crystal screen, voice, etc.).

[0030] By using the cognitive function determination system 100 configured as described above, it is possible to detect the signs and onset of dementia in the target person P1.

[0031] For example, as symptoms of dementia, "memory impairment", "disorientation", "executive function disorder", "day-night reversal", etc. can be considered.

[0032] "Memory impairment" is a disorder in which symptoms such as an inability to remember new things and a loss of memories that should have been remembered previously occur. When "memory impairment" develops in the target person P1, it is assumed that changes (abnormalities) in behavior such as forgetting having eaten and eating again, forgetting having cleaned and cleaning again, etc. will occur.

[0033] "Disorientation" is a disorder in which a person cannot grasp their situation, such as "when, where, who". When "disorientation" develops in the target person P1, it is assumed that changes (abnormalities) in behavior such as turning on the heater in summer, not knowing the locations in the house and being unable to go to the toilet or bathroom, and not being able to return home after going out will occur.

[0034] "Executive function disorder" is a disorder in which a person cannot perform tasks in an orderly manner by making arrangements and plans. When "executive function disorder" develops in the target person P1, it is assumed that changes (abnormalities) in behavior such as being unable to prepare meals and not knowing how to use electrical appliances will occur.

[0035] "Circadian reversal" is a disorder in which the sleep-wake rhythm is disrupted and day and night are reversed. When "circadian reversal" develops in subject P1, it is assumed that changes (abnormalities) in behavior such as activity during the time period (midnight) when sleep should normally occur will occur.

[0036] Therefore, the cognitive function determination system 100 of the present embodiment detects changes in the behavior of subject P1 as described above, and determines the presence or absence of abnormalities in the cognitive function of the subject P1 based on these changes in behavior. Then, if necessary, the family P2 of subject P1 is notified of the abnormality in the cognitive function of subject P1. As a result, the family P2 of subject P1 can grasp the signs and onset of dementia in subject P1, and can take appropriate measures (treatment, etc.) at an early stage.

[0037] Hereinafter, an outline of a series of processes by this cognitive function determination system 100 will be described.

[0038] As shown in FIG. 2, the server 120 classifies and stores various facilities provided in the house 1 according to the frequency of use. Specifically, the server 120 classifies and stores various facilities into facilities used daily (classification (1)) and facilities used more than once a week but not daily (classification (2)). For example, in the example shown in FIG. 2, cooking appliances (IH, microwave oven), refrigerator, TV, dryer, etc. are used daily, so they are classified into classification (1). Also, vacuum cleaners, washing machines, etc. are not used daily but are used more than once a week, so they are classified into classification (2).

[0039] Furthermore, the server 120 also classifies and stores facilities that are used only in a specific season into classification (1) or classification (2). For example, in the example shown in FIG. 2, air conditioners and heating appliances are used daily in the necessary seasons (summer and winter), so they are classified into classification (1). Although no example is given for classification (2) in FIG. 2, for example, if a washing and drying machine is used only during the rainy season, this washing and drying machine can be classified into classification (2).

[0040] The server 120 can learn the detection results of the power sensor 110 in advance and classify various facilities based on this learning result. That is, it can determine the frequency of use of various facilities within a predetermined period and classify them into classification (1) or (2) based on this frequency. Also, instead of using the learned results, it is also possible to arbitrarily determine the classification by the manufacturer, seller, user, etc. of the cognitive function determination system 100.

[0041] The server 120 determines the presence or absence of an abnormality in the cognitive function of the subject P1 by performing appropriate processing for each classified facility, and notifies the family P2 of the subject P1 about the abnormality in the cognitive function as necessary.

[0042] Specifically, for the facilities classified into classification (1), the server 120 detects the daily usage status (and thus the actions of the subject P1), and performs processing to determine the presence or absence of an abnormality in the cognitive function of the subject P1 based on this detection result. Hereinafter, this processing is referred to as "daily processing". In the daily processing, the server 120 associates and stores the actions to be determined, the presence or absence of an abnormality, and the reasons for the determination with each other.

[0043] Also, the server 120 calculates, for each action, the number of times an abnormality is detected by the above daily processing within one week. Then, for the actions in which an abnormality is detected a predetermined number of times or more, an alert (warning) is reported using the terminal 130. Hereinafter, this processing is referred to as "weekly result notification". Through the weekly result notification, the family P2 of the subject P1 who holds the terminal 130 can grasp that an abnormality has occurred in a predetermined action of the subject P1, and thus there is a possibility that the cognitive function has declined.

[0044] On the other hand, for the facilities classified into classification (2), the server 120 detects the weekly usage status (and thus the actions of the subject P1), and performs processing to determine the presence or absence of an abnormality in the cognitive function of the subject P1 based on this detection result. Hereinafter, this processing is referred to as "weekly processing". In the weekly processing, the server 120 associates and stores the actions to be determined, the presence or absence of an abnormality, and the reasons for the determination with each other.

[0045] Also, the server 120 calculates, for each action, the number of times an abnormality is detected by the above-mentioned per-week processing in one month. Then, for an action in which an abnormality is detected a predetermined number of times or more, an alert is issued using the terminal 130. Hereinafter, this processing is referred to as "monthly result notification". As a result, the family member P2 of the subject P1 who owns the terminal 130 can grasp that an abnormality has occurred in the action of the subject P1, and thus there is a possibility that the cognitive function has declined.

[0046] In this way, in the cognitive function determination system 100, by performing each process according to facilities with different usage frequencies (classification (1) and classification (2)), it is possible to determine the presence or absence of an abnormality in the cognitive function of the subject P1 and issue an alert to the family member P2 of the subject P1.

[0047] Hereinafter, the processing contents (daily processing, weekly processing, weekly result notification, and monthly result notification) of the above-mentioned cognitive function determination system 100 will be specifically described.

[0048] As a prerequisite for various processes, the server 120 learns the tendency of the actions of the subject P1 by detecting the usage status of various facilities in the house 1 in advance (before performing the above-mentioned daily processing, etc.) over a predetermined period (for example, about two weeks to one month), and stores that information (learning action data). Specifically, based on the detection result of the power sensor 110, the server 120 detects the actions performed by the subject P1 during a day of life and the time zones thereof, and grasps the actions performed by the subject P1 for each time zone. An example is shown in the "learning result" in the table of FIG. 3.

[0049] For example, when the server 120 detects that a cooking appliance (IH, microwave oven) has been used, it learns that time zone. By repeating this for a predetermined period, the server 120 grasps the actions of the subject P1, that is, at which time zone the subject P1 usually uses the cooking appliance. Also, the server 120 indirectly detects that the subject P1 has gone to bed from, for example, the fact that the lighting in the house 1 has been turned off, and learns the sleeping time of the subject P1.

[0050] Note that the actions of the subject P1 change according to the season, such as the air conditioner and heating appliances being used only in the necessary seasons (summer and winter). Therefore, the server 120 learns the tendency of the actions of the subject P1 for each season, and when performing daily processing and the like described below, it uses the learning result according to the season at that time.

[0051] In addition, during the period targeted for the above daily processing and the like, the server 120 constantly detects the usage status of various facilities in the house 1 (and thus the actions of the subject P1), and stores that information (execution behavior data). More specifically, the server 120 stores at what time periods the subject P1 actually performed what actions. An example is shown in the "detection result" in the table of FIG. 3.

[0052] First, the daily processing will be described with reference to FIGS. 4 to 7.

[0053] The server 120 performs daily processing at a predetermined time every day. Since the server 120 needs to determine whether the subject P1 is sleeping soundly at night, it is not preferable to perform daily processing during the time period when the subject P1 is considered to be sleeping. Therefore, the server 120 according to this embodiment executes daily processing during the time period when the subject P1 is considered to have surely woken up (for example, 10:00 am, etc.).

[0054] In step S101, the server 120 extracts the actions of the subject P1 during the most recent 24 hours (for example, from 9:00 am the previous day to 9:00 am the current day when performing daily processing at 10:00 am) from the actions of the subject P1 that it has stored. This 24-hour period becomes the period targeted for determination by the daily processing. Hereinafter, this period will be referred to as the "daily processing target period". After performing the processing of step S101, the server 120 proceeds to step S102.

[0055] In step S102, the server 120 checks the time of each action extracted in step S101, and determines whether a predetermined time (X1 hours) has elapsed since it was last detected that the subject P1 performed an action. In other words, when X1 hours or more have elapsed since the subject P1 last performed an action, it means that the action of the subject P1 has not been detected for X1 hours or more. That is, in this case, there is a possibility that some emergency has occurred, such as the subject P1 having fallen. Note that the value of the predetermined time (X1 hours) can be arbitrarily set, but in particular, it is desirable to set a time (for example, 12 hours, 24 hours, etc.) that allows it to be inferred that an emergency has occurred to the subject P1.

[0056] If the server 120 determines that X1 hours or more have elapsed since the subject P1 last performed an action, it proceeds to step S103. On the other hand, if the server 120 determines that X1 hours or more have not elapsed since the subject P1 last performed an action, it proceeds to step S104.

[0057] In step S103, the server 120 issues a predetermined alert using the terminal 130. The family member P2 who confirms this alert can recognize that an emergency has occurred to the subject P1 and can take actions such as heading to the residence 1 of the subject P1.

[0058] Note that since the processing of steps S101 to S103 can confirm that an emergency has occurred to the subject P1, it may not be executed only once a day (at 10:00 am), but may be executed constantly. This allows for a more rapid response to an emergency involving the subject P1. After performing the processing of step S103, the server 120 proceeds to step S104.

[0059] In step S104, the server 120 determines whether an abnormality has occurred in the behavior of the target person P1 based on the behavior of the target person P1 during the activity time period (the time period when the person is awake and active). Hereinafter, this process is referred to as "activity time period determination". Hereinafter, the activity time period determination will be described with reference to FIG. 5.

[0060] The server 120 repeats the processes from step S201 to step S204 in FIG. 5 for each behavior of the target person P1 classified into classification (1). Also, for behaviors that occur multiple times a day (for example, using cooking appliances (IH, microwave oven), watching TV, etc.), the process is performed for each occurrence. Hereinafter, each process from step S201 to step S204 will be described.

[0061] In step S201, the server 120 determines whether the behavior to be determined has been detected during the daily processing target period. Specifically, the server 120 determines whether a behavior learned as a behavior performed by the target person P1 (for example, using the cooking appliances (IH, microwave oven) shown in FIG. 3) has been detected during the daily processing target period.

[0062] If the server 120 detects the behavior to be determined, it proceeds to step S202. On the other hand, if the server 120 does not detect the behavior to be determined, it proceeds to step S204.

[0063] In step S202, the server 120 determines whether the deviation in the time zone of the detected behavior of the target person P1 with respect to the learning result is within a predetermined time (X2 hours). Note that the method for calculating the deviation in the time zone is not particularly limited, and any method that can determine whether the time zones of both (the learned behavior and the detected behavior) are changing may be used. For example, a method of calculating by summing the non-overlapping times of both, a method of calculating by summing the deviation between the start times of both behaviors and the deviation between the end times of both behaviors, etc. can be considered. Also, the value of the predetermined time (X2 hours) can be arbitrarily set, but in particular, it is desirable to set a time (for example, 1 hour, 2 hours, etc.) to the extent that it can be inferred that the target person P1 has symptoms of dementia.

[0064] If the deviation in the time zone of the behavior of the target person P1 with respect to the learning result is within X2 hours, the server 120 proceeds to step S203. On the other hand, if the deviation in the time zone of the behavior of the target person P1 with respect to the learning result is greater than X2 hours, the server 120 proceeds to step S204.

[0065] In step S203, the server 120 determines that the behavior subject to the determination is normal. The server 120 stores the behavior subject to this determination, the presence or absence of abnormality, and the reason for the determination in association with each other. For example, the server 120 stores information such as "TV", "no abnormality", and "the deviation in the execution timing is within X2 hours" in association with each other.

[0066] On the other hand, in step S204 shifted from step S201 or step S202, the server 120 determines that the behavior subject to the determination is abnormal.

[0067] Here, the case of transitioning from step S201 to step S204 (when the answer in step S201 is NO) indicates that the actions that subject P1 should perform every day have not been performed even once during one day (the daily processing target period). In this case, it is presumed that subject P1 has developed some symptoms of dementia, such as memory impairment, disorientation, and executive function impairment. Server 120 associates and stores the actions subject to this determination, the presence or absence of abnormalities, and the reasons for the determination with each other. For example, server 120 associates and stores information such as "cooking equipment (IH, microwave oven)", "abnormal", and "not performed today" with each other.

[0068] Also, the case of transitioning from step S202 to step S204 (when the answer in step S202 is NO) indicates that subject P1 is performing a predetermined action at a time different from normal. In this case, it is presumed that subject P1 has developed some symptoms of dementia, such as memory impairment, disorientation, executive function impairment, and reversal of day and night. Server 120 associates and stores the actions subject to this determination, the presence or absence of abnormalities, and the reasons for the determination with each other. For example, server 120 associates and stores information such as "washing machine", "abnormal", and "the deviation in the execution timing is greater than X2 hours" with each other.

[0069] Server 120 repeats the processing from step S201 to step S204 for each action of subject P1 and for the number of times thereof. As a result, server 120 can determine whether an abnormality has occurred for each action and for each time of subject P1. Thereafter, server 120 proceeds to step S105 in FIG. 4.

[0070] In step S105 of FIG. 4, server 120 determines whether an abnormality has occurred in the actions of subject P1 based on the actions during the time period from when subject P1 goes to bed until getting up. Hereinafter, this processing is referred to as "bedtime - wake - up time period determination". Hereinafter, the bedtime - wake - up time period determination will be described with reference to FIGS. 6 and 7.

[0071] In step S301 of FIG. 6, the server 120 determines the time period from the actions of the target person P1 stored in it until the target person P1 goes to bed and wakes up. After performing the process of step S301, the server 120 proceeds to step S302.

[0072] In step S302, the server 120 determines whether the deviation of the detected time period (sleep time period) from when the target person P1 goes to bed and wakes up with respect to the learning result is within a predetermined time (X3 hours). The method for calculating the deviation of the time period is not particularly limited and can be calculated by the method exemplified in step S202. Note that the value of the predetermined time (X3 hours) can be arbitrarily set, but in particular, it is desirable to set a time (for example, 1 hour, 2 hours, etc.) to the extent that it can be inferred that the target person P1 has symptoms of dementia.

[0073] If the server 120 determines that the deviation of the sleep time period of the target person P1 with respect to the learning result is within X3 hours, it proceeds to step S308 of FIG. 7. On the other hand, if the server 120 determines that the deviation of the sleep time period of the target person P1 with respect to the learning result is greater than X3 hours, it proceeds to step S303.

[0074] Note that in the example shown in FIG. 6, the deviation of the sleep time period of the target person P1 with respect to the learning result is determined, but not only this. For example, it is also possible to determine whether the target person P1 sleeps during the daily processing target period. For example, if the target person P1 does not sleep at all, it may be configured to proceed to step S303.

[0075] In step S303, the server 120 determines whether the sleep time of the target person P1 has become longer than the pre-learned sleep time. If the server 120 determines that the sleep time of the target person P1 has not become longer, it proceeds to step S304. On the other hand, if the server 120 determines that the sleep time of the target person P1 has become longer, it proceeds to step S305.

[0076] In step S304, the server 120 determines that the behavior (sleep) of the subject P1 during the sleep time zone is abnormal. That is, when the deviation in that time zone has increased significantly even though the sleep time has not become longer (NO in step S302 and NO in step S303), it is presumed that the subject P1 has developed some symptoms of dementia such as executive dysfunction, agnosia, or circadian rhythm reversal. The server 120 stores the behavior (sleep) to be judged, the presence or absence of abnormalities, and the reasons for the judgment in association with each other. For example, the server 120 stores the information "sleep", "abnormal", and "the deviation in the execution timing is greater than X3 hours" in association with each other.

[0077] After performing the process of step S304, the server 120 proceeds to step S308 in FIG. 7.

[0078] On the other hand, in step S305 shifted from step S303, the server 120 determines that attention is required for the behavior of the subject P1 during the sleep time zone. Here, the determination of "attention (required)" is a determination that, although it cannot be said to be "normal", the degree of abnormality is lower (minor) than the determination of "abnormal" in step S304.

[0079] When shifting from step S303 to step S305 (YES in step S303), since the sleep time of the subject P1 has become longer, it is possible that the subject P1 is simply tired rather than having a decline in cognitive function. Therefore, in this embodiment, a determination of "attention", which is less severe than "abnormal", is made in step S305. The server 120 stores the behavior (sleep) to be judged, the presence or absence of abnormalities, and the reasons for the judgment in association with each other. For example, the server 120 stores the information "sleep", "attention required", "the deviation in the execution timing is greater than X3 hours and the sleep time has become longer" in association with each other. After performing the process of step S305, the server 120 proceeds to step S306.

[0080] In step S306, the server 120 determines whether the deviation of the sleep time zone of the subject P1 has reduced the activity time zone (time zone other than the original sleep time zone), thereby affecting the detected behavior of the subject P1. For example, if the behavior determined to be abnormal in the activity time zone determination (see FIG. 5) (see step S204) overlaps with the time zone when the behavior should originally be performed and the detected sleep time of the subject P1, it can be determined that the deviation of the sleep time zone of the subject P1 has affected the behavior (the behavior has been determined to be abnormal due to the deviation of the sleep time zone).

[0081] If the server 120 determines that the deviation of the sleep time zone of the subject P1 has affected the detected behavior of the subject P1, it proceeds to step S307. On the other hand, if the server 120 determines that the deviation of the sleep time zone of the subject P1 has not affected the detected behavior of the subject P1, it proceeds to step S308 in FIG. 7.

[0082] In step S307, the server 120 changes the determination result in step S305 from "Attention" to "Abnormal". In this way, when the sleep time has become longer (YES in step S303) and the behavior that the subject P1 should originally perform is affected (YES in step S306), it is assumed that the situation is not favorable for the subject P1. Therefore, the degree of abnormality of the determination result is raised from "Attention" to "Abnormal". After performing the process in step S307, the server 120 proceeds to step S308 in FIG. 7.

[0083] In step S308 of FIG. 7, the server 120 determines whether it has detected the behavior (behavior other than sleep) of the subject P1 in the time zone from when the subject P1 goes to bed until getting up (sleep time zone, see step S301). If the server 120 determines that it has not detected the behavior of the subject P1 in the sleep time zone, it proceeds to step S309. On the other hand, if the server 120 determines that it has detected the behavior of the subject P1 in the sleep time zone, it proceeds to step S310.

[0084] In step S309, the server 120 determines that the actions of the subject P1 during the sleep time period are normal. The server 120 associates and stores the actions to be determined, the presence or absence of abnormalities, and the reasons for the determination with each other. For example, the server 120 associates and stores the information of "sleep", "no abnormality", and "no action detected during the sleep time period" with each other. After performing the process of step S309, the server 120 ends the daily process (see FIG. 4).

[0085] On the other hand, in step S310 migrated from step S308, the server 120 determines whether there are actions other than going to the toilet and related actions among the actions of the subject P1 detected during the sleep time period.

[0086] Specifically, when going to the toilet during the sleep time period, it is assumed that some actions will be performed in relation to this. For example, when the subject P1 goes to the toilet, it is assumed that not only the lighting in the toilet is turned on, but also the lighting in the bedroom, corridor, etc. is turned on to ensure the visibility when going to the toilet. The server 120 determines whether there are any actions other than the actions related to going to the toilet (necessary actions). The type of actions related to the toilet can be learned by the server 120 or determined by the manufacturer, seller, user, etc. of the cognitive function determination system 100.

[0087] If the server 120 determines that there are actions other than going to the toilet and related actions, it proceeds to step S311. On the other hand, if the server 120 determines that there are no actions other than going to the toilet and related actions, it proceeds to step S312.

[0088] In step S311, the server 120 determines that the behavior of the subject P1 during the sleep time zone is abnormal. That is, when there is behavior other than toilet-related behavior during the sleep time zone (YES in step S310), it is presumed that the subject P1 has developed some symptoms of dementia, such as executive dysfunction, disorientation, or reversed day-night rhythm. The server 120 stores the behavior (sleep) to be judged, the presence or absence of abnormality, and the reason for the judgment in association with each other. For example, the server 120 stores the information of "sleep", "abnormal", and "behavior detected during sleep time zone" in association with each other. After performing the process of step S311, the server 120 ends the daily process (see FIG. 4).

[0089] On the other hand, in step S312 shifted from step S310, the server 120 determines that the behavior of the subject P1 during the sleep time zone is normal. The server 120 stores the behavior (sleep) to be judged, the presence or absence of abnormality, and the reason for the judgment in association with each other. For example, the server 120 stores the information of "sleep", "no abnormality", and "no behavior detected during sleep time zone" in association with each other.

[0090] In this embodiment, it is determined to be normal in step S312. However, for example, it is also possible to compare the learned number of toilet visits of the subject P1 with the detected number of toilet visits and determine "attention" when the number has increased. By this, it is also possible to prompt attention to the increase in the number of toilet visits.

[0091] After performing the process of step S312, the server 120 ends the daily process (see FIG. 4).

[0092] Next, the weekly process will be described with reference to FIG. 8.

[0093] The server 120 performs the weekly process at a predetermined date and time once a week. For example, the weekly process is performed once a week at the same time zone as the daily process (for example, 10:00 am, etc.).

[0094] Server 120 repeats the processes from step S401 to step S403 in FIG. 8 for each action of the subject P1 classified into classification (2). Hereinafter, each process from step S401 to step S403 will be described.

[0095] In step S401, server 120 determines whether the number of actions of the detected subject P1 has increased or decreased by a predetermined number of times (Y1 times) or more with respect to the learning result. Note that the value of the predetermined number of times (Y1 times) can be arbitrarily set. If server 120 determines that the number of actions of the detected subject P1 has increased or decreased by Y1 times or more, it proceeds to step S402. On the other hand, if server 120 determines that the number of actions of the detected subject P1 has not increased or decreased by Y1 times or more, it proceeds to step S403.

[0096] In step S402, server 120 determines that the action to be judged is abnormal. That is, when the number of actions has increased or decreased by Y1 times or more, it is presumed that the subject P1 has developed some symptoms of dementia, such as memory impairment, disorientation, and executive function disorder. Server 120 stores the action to be judged, the presence or absence of abnormality, and the reason for the judgment in association with each other. For example, server 120 stores the information "vacuum cleaner", "abnormal", and "the action has decreased" in association with each other.

[0097] On the other hand, in step S403 that has shifted from step S401, server 120 determines that the action to be judged is normal. That is, when the number of actions has not increased or decreased by Y1 times or more, it is presumed that there is no particular change in the actions of the subject P1 and that no symptoms of dementia have developed. Server 120 stores the action to be judged, the presence or absence of abnormality, and the reason for the judgment in association with each other. For example, server 120 stores the information "vacuum cleaner", "no abnormality", and "the action has not increased or decreased" in association with each other.

[0098] Server 120 repeats the processes from step S401 to step S403 for each action of the target person P1. As a result, Server 120 can determine whether an abnormality has occurred for each action of the target person P1. After that, Server 120 ends the weekly process.

[0099] Next, the weekly result notification will be described with reference to Fig. 9(a).

[0100] Server 120 performs a weekly result notification at a predetermined date and time once a week. For example, the weekly result notification is performed once a week at the same time zone as the daily process (e.g., 10:00 am, etc.).

[0101] In step S501, Server 120 determines whether there is an action that has been repeatedly cautioned or abnormally determined a predetermined number of times (Y2 times) or more among the actions targeted for the daily process during one week. Note that the value of the predetermined number of times (Y2 times) can be arbitrarily set, but in order to determine that the determination of abnormality, etc. has been repeated, it is preferably set to a value of 2 or more. If Server 120 determines that there is an action that has been repeatedly cautioned or abnormally determined a predetermined number of times (Y2 times) or more, it proceeds to step S502. On the other hand, if Server 120 determines that there is no action that has been repeatedly cautioned or abnormally determined a predetermined number of times (Y2 times) or more, it ends the weekly result notification.

[0102] In step S502, Server 120 issues an alert (abnormality warning of the recognition function) using the terminal 130. Specifically, the action that has been cautioned or abnormally determined a predetermined number of times or more and the reason for the determination are notified to the family member P2 using the terminal 130. As a method of notification, for example, there are a method of displaying on the liquid crystal screen of the terminal 130 and a method of notifying by emitting sound from the terminal 130.

[0103] In this way, for actions where anomalies have been detected repeatedly (Y2 times or more), since it is highly likely that an anomaly has occurred in the cognitive function, the server 120 notifies the family member P2. As a result, the family member P2 can take appropriate measures (such as treatment). In other words, for actions where the number of times determined to be abnormal or the like is less than a predetermined number (Y2 times), the server 120 does not notify the family member P2. This can prevent the family member P2 from being notified of actions (false detection of anomalies) that are determined to be abnormal or the like when the actions of the subject P1 happen to be different from normal, independent of the cognitive function of the subject P1.

[0104] After performing the process of step S502, the server 120 ends the weekly result notification.

[0105] Next, the monthly result notification will be described with reference to FIG. 9(b).

[0106] The server 120 performs a monthly result notification at a predetermined date and time once a month. For example, the monthly result notification is performed once a month at the same time zone as the weekly result notification (e.g., 10:00 am, etc.).

[0107] In step S601, the server 120 determines whether there are any actions that have been repeatedly cautioned or determined to be abnormal a predetermined number (Y3 times) or more among the actions targeted for weekly processing during a month. Note that the value of the predetermined number (Y3 times) can be arbitrarily set, but in order to determine that the determination of anomalies or the like has been repeated, it is preferably set to a value of 2 or more. If the server 120 determines that there are actions that have been repeatedly cautioned or determined to be abnormal a predetermined number (Y3 times) or more, it proceeds to step S602. On the other hand, if the server 120 determines that there are no actions that have been repeatedly cautioned or determined to be abnormal a predetermined number (Y3 times) or more, it ends the monthly result notification.

[0108] In step S602, the server 120 issues an alert (abnormality warning of the cognitive function) using the terminal 130. Specifically, the server 120 notifies family member P2 of the actions that have been determined to be a caution or an abnormality a predetermined number of times or more, and the reasons for the determination, using the terminal 130. In this way, for actions where the number of times determined to be an abnormality or the like is less than the predetermined number (Y3 times), the server 120 does not notify family member P2.

[0109] After performing the process of step S602, the server 120 ends the monthly result notification.

[0110] By performing the above processes (daily process, weekly process, weekly result notification, and monthly result notification), the cognitive function determination system 100 can detect an abnormality in the cognitive function of the subject P1 and notify family member P2. For example, FIG. 3 shows the determination results of the daily process. In the example shown in FIG. 3, according to the learning result, subject P1 should use the dryer at 21:00, but according to the actual detection result, it is not being used (see step S201), so the server 120 determines it as "abnormal". Also, after 22:00, although subject P1 is sleeping, for some reason (such as the sleep time becoming longer (see step S303)), the server 120 determines it as "caution". When the server 120 detects an abnormality or the like in this way, if it seems to be repeated several times, it determines that there is a high possibility that an abnormality has occurred in the cognitive function of subject P1, and notifies family member P2 (see FIG. 9).

[0111] In this way, by using the cognitive function determination system 100, it is possible to detect an abnormality in the cognitive function based on the actions of subject P1 in daily life. As a result, it is possible to simply detect an abnormality in the cognitive function without subjecting subject P1 to a special examination or the like.

[0112] As described above, the cognitive function determination system 100 according to the present embodiment is capable of detecting the usage status of facilities provided in the building (residence 1) used by the subject P1, and thus can detect the actions of the subject P1, and a power sensor 110 (action detection unit) A server 120 (cognitive function determination unit) that compares the execution behavior data regarding the actions of the subject P1 detected by the power sensor 110 with the learned behavior data regarding the actions of the subject P1, and determines the presence or absence of an abnormality in the cognitive function of the subject P1. It is provided with.

[0113] By configuring in this way, an abnormality in the cognitive function can be easily detected. That is, just by the subject P1 living as usual, the actions of the subject P1 can be detected, and the presence or absence of an abnormality in the cognitive function can be determined. As a result, an abnormality in the cognitive function can be detected without having the subject P1 undergo a special examination.

[0114] Also, the server 120 When the time zone in which the subject P1 performs a predetermined action is deviated from the learned time zone by a predetermined time or more, or when the subject does not perform the learned action (NO in step S201 or step S202), it is determined that an abnormality has occurred in the cognitive function of the subject (step S204), and it performs an action time determination process (activity time zone determination).

[0115] By configuring in this way, an abnormality in the cognitive function can be easily detected. That is, when there is a deviation between the time zone in which the subject P1 acts and the learned time zone, or when the action itself is not performed, it is presumed that some symptom (such as executive function disorder) of dementia has occurred in the subject P1. Therefore, by detecting the deviation between the two time zones and the like, an abnormality in the cognitive function can be easily detected.

[0116] Also, the server 120 When the time zone in which the subject P1 sleeps is deviated from the learned time zone by a predetermined time or more, or when the subject is not sleeping (NO in step S302), it is determined that an abnormality has occurred in the cognitive function of the subject P1 (step S304 or step S305), and it performs a sleep time determination process (bedtime to wake-up time zone determination).

[0117] By configuring in this way, it is possible to easily detect abnormalities in cognitive functions. That is, when there is a deviation between the sleep time zone of the subject P1 and the learned sleep time zone, or when sleep itself is not being carried out, it is presumed that some symptoms (such as executive function disorder) of dementia have occurred in the subject P1. Therefore, by detecting the deviation between the two time zones, etc., abnormalities in cognitive functions can be easily detected.

[0118] Also, in the determination of the bedtime to wake-up time zone, the server 120 determines the degree of abnormality in cognitive functions according to the increase or decrease in the sleep time of the subject P1 with respect to the learned sleep time (from step S303 to step S305).

[0119] By configuring in this way, a more detailed determination of cognitive functions can be made. That is, as illustrated in this embodiment, when the sleep time zone is long, it is possible that the subject P1 is simply tired and there is no abnormality in cognitive functions. Therefore, in such a case, by reducing the degree (level) of abnormality in cognitive functions and determining that "attention is required", a difference can be provided in the degree of abnormality determination to make a detailed determination.

[0120] Also, the server 120 When the number of times the subject P1 has performed a predetermined action within a predetermined period is different from the learned number of times by a predetermined number of times or more (step S401), it determines that an abnormality has occurred in the cognitive function of the subject (step S402) and performs an action count determination process (weekly process).

[0121] By configuring in this way, it is possible to easily detect abnormalities in cognitive functions. That is, when there is a difference between the number of times the subject P1 has acted and the learned number of times, it is presumed that some symptoms (such as executive function disorder) of dementia have occurred in the subject P1. Therefore, by detecting the difference between the two numbers of times, abnormalities in cognitive functions can be easily detected.

[0122] In addition, the cognitive function determination system 100 further includes a terminal 130 (notification unit) capable of performing notification to family member P2 (notification target). The server 120 determines the presence or absence of an abnormality in the cognitive function of the target person during a first period (daily processing, weekly processing), calculates the number of times an abnormality in the cognitive function is detected for each action of the target person during a second period longer than the first period (step S501, step S601), and when the number of times an abnormality in the cognitive function is detected during the second period is equal to or greater than a predetermined threshold, issues an abnormality warning for the cognitive function to the notification target using the notification unit (step S502, step S602).

[0123] With this configuration, it is possible to appropriately notify a warning to family member P2. That is, by issuing a warning only when the number of times an abnormality in the cognitive function is detected reaches a certain number of times, it is possible to prevent an abnormality warning for the cognitive function from being issued even when an abnormality in the cognitive function is occasionally detected (such as when a false detection occurs).

[0124] Note that the house 1 according to this embodiment is an embodiment of a building according to the present invention. In addition, the power sensor 110 according to this embodiment is an embodiment of an action detection unit according to the present invention. In addition, the server 120 according to this embodiment is an embodiment of a cognitive function determination unit according to the present invention. In addition, the activity time zone determination according to this embodiment is an embodiment of an action time determination process according to the present invention. In addition, the bedtime to wake-up time zone determination according to this embodiment is an embodiment of a sleep time determination process according to the present invention. In addition, the weekly processing according to this embodiment is an embodiment of an action frequency determination process according to the present invention. In addition, the terminal 130 according to this embodiment is an embodiment of a notification unit according to the present invention. In addition, the family member P2 according to this embodiment is an embodiment of a notification target according to the present invention.

[0125] The embodiments of the present invention have been described above. However, the present invention is not limited to the above embodiments, and appropriate modifications can be made within the scope of the technical idea of the invention described in the claims.

[0126] For example, although the cognitive function determination system 100 according to this embodiment shows an example with an elderly person as the target person P1, the present invention is not limited to the elderly, and various people can be used as the target person P1.

[0127] Also, in this embodiment, the power sensor 110 (behavior detection unit) is provided in the house 1. However, the present invention is not limited to this, and it can also be provided in various buildings used by the target person P1. That is, not only in the house 1, but also in various other buildings, facilities, etc., it is possible to detect an abnormality in the cognitive function.

[0128] Also, in this embodiment, an example is shown in which the server 120 (a virtual server provided on the cloud, etc.) performs various processes. However, the present invention is not limited to this, and the entity that executes various processes can be arbitrarily changed. For example, it can also be executed by a home server, a personal computer, a portable terminal, etc. provided in the house 1.

[0129] Also, in this embodiment, an example is shown in which when an abnormality in the cognitive function of the target person P1 is detected, the terminal 130 is notified of that fact. Furthermore, when a family member P2, etc. confirms the abnormality and determines that there is no problem (not an abnormality), it is also possible to configure to transmit that fact to the server 120 using the terminal 130 or the like. In this way, by the family member P2, etc. confirming the presence or absence of an abnormality and feeding back to the server 120, the server 120 can learn the behavior of the target person P1 more accurately. As a result, the detection of an abnormality in the cognitive function can be performed with higher accuracy.

[0130] In addition, in this embodiment, an example was shown in which when the actions of the target person P1 were repeatedly determined to be abnormal or the like, the family member P2 was notified to that effect (see FIG. 9). However, the present invention is not limited to this. For example, when the actions of the target person P1 are determined to be abnormal or the like even once, it is also possible to immediately notify the family member P2.

[0131] Also, the periods to which each of the processes (daily process, weekly process, weekly result notification, and monthly result notification) exemplified in this embodiment apply can be arbitrarily changed.

Explanation of Signs

[0132] 100 Cognitive Function Judgment System 110 Power Sensor 120 Server 130 Terminal

Claims

1. An action detection unit capable of detecting the actions of the target person by detecting the usage status of a plurality of facilities provided in the building used by the target person, A cognitive function determination unit that compares the execution action data regarding the actions of the target person detected by the action detection unit with the learned action data regarding the actions of the target person, and determines the presence or absence of an abnormality in the cognitive function of the target person, A notification unit capable of notifying the notification target person, Comprising, The cognitive function determination unit, As the plurality of facilities, a first facility belonging to a first classification, which is pre-classified according to the usage frequency in a predetermined period, and a second facility belonging to a second classification having a lower usage frequency than the first classification are stored, In a first period, determine the presence or absence of an abnormality in the cognitive function of the target person regarding the first facility, In a second period longer than the first period, calculate the number of times of detecting an abnormality in the cognitive function for each action of the target person regarding the first facility, When the number of times of detecting an abnormality in the cognitive function regarding the first facility in the second period is equal to or greater than a predetermined threshold, issue an abnormality warning for the cognitive function to the notification target person using the notification unit, In the second period, determine the presence or absence of an abnormality in the cognitive function of the target person regarding the second facility, In a third period longer than the second period, calculate the number of times of detecting an abnormality in the cognitive function for each action of the target person regarding the second facility, When the number of times of detecting an abnormality in the cognitive function regarding the second facility in the third period is equal to or greater than a predetermined threshold, issue an abnormality warning for the cognitive function to the notification target person using the notification unit, Cognitive function determination system.

2. The cognitive function determination unit, When the time zone in which the target person performs a predetermined action deviates from the learned time zone by a predetermined time or more, or when the target person does not perform the learned action, perform an action time determination process for determining that an abnormality has occurred in the cognitive function of the target person, The cognitive function determination system according to claim 1.

3. The cognitive function determination unit, When the time zone in which the target person sleeps deviates from the learned time zone by a predetermined time or more, or when the target person is not sleeping, perform a sleep time determination process for determining that an abnormality has occurred in the cognitive function of the target person, The cognitive function determination system according to claim 1 or claim 2.

4. The cognitive function determination unit, in the sleep time determination process, Determining the degree of abnormality in cognitive function according to the increase or decrease in the sleep time of the subject with respect to the learned sleep time The cognitive function determination system according to claim 3

5. The cognitive function determination unit When the number of times the subject has performed a predetermined action during a predetermined period is different from the learned number of times by a predetermined number of times or more, performs an action count determination process for determining that an abnormality has occurred in the cognitive function of the subject The cognitive function determination system according to any one of claims 1 to 4

Images