Health risk estimation system

Abstract

To provide a health risk estimation system with which it is possible to suitably estimate the health risk of a subject.SOLUTION: The health risk estimation system comprises: a subject detection sensor 110 capable of detecting the motion of a subject; an action estimation unit capable of estimating an action of the subject on the basis of the detection result of the subject detection sensor 110; a subject detection sensor 110 capable of detecting walking speed data that is data pertaining to the walking of the subject before and after the action; a walking change data acquisition unit for acquiring walking speed change data indicating a change between the walking speed data of the subject before the action and the walking speed data of the subject after the action, on the basis of the detection result of the subject detection sensor 110; and a health risk estimation unit capable of estimating health risks of the subject on the basis of the result of comparison between the current walking speed change data and the past walking speed change data.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to the technology of a health risk estimation system.

Background Art

[0002] Conventionally, the technology of a system for estimating the health risk (health risk) of a subject has been known. For example, it is as described in Patent Document 1.

[0003] Patent Document 1 discloses a technique for evaluating (estimating) the fall risk of a subject using walking data, which is data related to the walking of the subject. In the invention described in Patent Document 1, the walking data of the subject is acquired, and the fall risk of the subject is evaluated by comparing the walking data with a threshold value.

[0004] However, the walking data varies under various influences. For this reason, it is assumed that it is difficult to suitably estimate the health risk (fall risk) of the subject by simply comparing the walking data.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] The present invention has been made in view of the above circumstances, and the problem to be solved is to provide a health risk estimation system capable of suitably estimating the health risk of a subject.

Means for Solving the Problems

[0007] The problem to be solved by the present invention is as described above. Next, the means for solving this problem will be described.

[0008] That is, in claim 1, an action detection unit capable of detecting the actions of a target person, an action estimation unit capable of estimating the actions of the target person based on the detection result of the action detection unit, a walking data detection unit capable of detecting walking data, which is data related to the walking of the target person, before and after the action, and a walking variability data acquisition unit that acquires walking variability data indicating the variation between the walking data of the target person before the action and the walking data of the target person after the action based on the detection result of the walking data detection unit, a first walking variability data that is the walking variability data at a first time, and a second walking variability data that is the walking variability data of an action similar to the action of the first walking variability data and is from a time earlier than the first time, and a health risk estimation unit capable of estimating the health risk of the target person based on the comparison result between the two.

[0009] In claim 2, the walking data detection unit is capable of detecting the walking data of the target person in a non-contact manner.

[0010] In claim 3, the walking data detection unit detects the walking speed of the target person as the walking data.

[0011] In claim 4, it includes a movement trajectory estimation unit capable of estimating the movement trajectory of the target person's walking based on the detection result of the action detection unit, and the walking variability data acquisition unit acquires the walking variability data using the walking data before and after the action and having similar movement trajectories to each other.

[0012] In claim 5, it includes a time information acquisition unit capable of acquiring information on the time related to the actions of the target person detected by the action detection unit, and the action estimation unit estimates the actions of the target person based on the detection result of the action detection unit and the acquisition result of the time information acquisition unit.

Advantages of the Invention

[0013] As an effect of the present invention, the following effects are achieved.

[0014] In the present invention, the health risk of the target person can be suitably estimated.

Brief Description of the Drawings

[0015]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Mode for Carrying Out the Invention

[0016] Hereinafter, with reference to FIG. 1, a health risk estimation system 100 according to an embodiment of the present invention will be described.

[0017] The health risk estimation system 100 can estimate the health risk of the target person (health risk). The "health risk" in the present embodiment refers to, for example, a health risk that can be estimated from the risk of falling (fall risk) of the target person. First, an example of the target person assumed in the present embodiment and the room 1 of the target person will be described.

[0018] In this embodiment, an elderly person is assumed to be the person (hereinafter referred to as the "subject") for whom the health risk estimation system 100 estimates the risk of falling. Further, in this embodiment, it is assumed that the health risk estimation system 100 is applied to Room 1 of the subject in the elderly care facility.

[0019] FIG. 1 shows an example of Room 1 of the subject. In Room 1, a bed 10, a dining table 20, dining chairs 30, a toilet 40, a bathtub 50, a kitchen 60, an entrance / exit 70, etc. are installed. The entrance / exit 70 is the entrance / exit of Room 1. The space where the toilet 40 and the bathtub 50 are arranged is partitioned from the other spaces by a partition wall 41. Note that a washstand (not shown) is arranged in the space where the bathtub 50 is arranged. The partition wall 41 is provided with an entrance / exit for entering and exiting the toilet 40 and the bathtub 50. Although the wall that partitions the space where the toilet 40 is arranged and the space where the bathtub 50 is arranged is not shown in the illustrated example, a wall for partitioning each of the above spaces may be provided. The configuration of Room 1 is not limited to that described above, and various spaces such as a study and a sofa area (a space where a sofa etc. is arranged) can be arranged.

[0020] Next, the configuration of the health risk estimation system 100 will be described. The health risk estimation system 100 mainly includes a subject detection sensor 110, a staff terminal 120, and a server 130.

[0021] The subject detection sensor 110 is for detecting the position, posture, and walking speed of the subject. The subject detection sensor 110 is a non-contact sensor. In this embodiment, a millimeter-wave sensor is used as the subject detection sensor 110. The subject detection sensor 110 irradiates an object with millimeter waves (radio waves with a wavelength of 1 to 10 mm and a frequency of 30 to 300 GHz) and observes the reflected radio waves, thereby measuring the position, speed, angle, movement, etc. of the object.

[0022] In this embodiment, the subject detection sensor 110 is installed at a position (such as the ceiling or wall) in Room 1 where radio waves can be irradiated over the entire area of Room 1. In this embodiment, an example is shown in which the entire area of Room 1 is detected by a single subject detection sensor 110, but a plurality of subject detection sensors 110 may be used as necessary.

[0023] The staff terminal 120 is for notifying the estimation results and the like by the health risk estimation system 100. As the staff terminal 120, for example, a personal computer can be used. The staff terminal 120 is installed in a place where a person who assists (assists, cares for, etc.) the life of the subject can confirm it. In this embodiment, for example, it can be arranged in the cubicle of the staff of a senior facility. By displaying various information on the monitor of the staff terminal 120, the information can be notified to the staff. Note that various other devices (for example, portable terminals, etc.) can also be used as the staff terminal 120.

[0024] The server 130 is for performing processing for estimating the risk of falling by the health risk estimation system 100. The server 130 is constituted by, for example, a cloud server. The server 130 can exchange information with the subject detection sensor 110 and the staff terminal 120.

[0025] Server 130 can determine the position and posture of the target person based on the data obtained by the target person detection sensor 110. Specifically, Server 130 can determine that among the point cloud data obtained based on the reflected radio waves, the data with movement is a person (in this embodiment, the target person). In particular, in this embodiment, since a millimeter-wave sensor is used, even fine movements of the target person can be accurately detected. Server 130 extracts only the data with movement and can determine the position of the data in Room 1 as the position of the target person (each area described later). Also, Server 130 extracts only the data with movement and can determine the posture of the target person, such as standing (standing position), sitting (sitting position), lying down (lying position), etc., by checking the distribution of the data.

[0026] In addition, Server 130 can measure the walking speed of the target person based on the data obtained by the target person detection sensor 110. Hereinafter, Server 130 acquires data (hereinafter referred to as "walking speed data") measuring the walking speed of the target person.

[0027] Server 130 stores in advance the position information of areas obtained by dividing Room 1 into a plurality of parts. In FIG. 1, an example is shown in which a bed area A, a dining area B, a toilet area C, a bathroom area D, a kitchen area E, another area F, and an outdoor area G are set according to the arrangement of furniture and the like.

[0028] The bed area A is the area (bedroom) where the bed 10 is placed. The bed area A is set to be, for example, a range that is slightly larger than the bed 10 in plan view. Similarly, a dining area B is set in the area where the dining table 20 and the dining chairs 30 are placed. Also, a kitchen area E is set in the area where the kitchen 60 is placed. The toilet area C and the bathroom area D (washroom) are within the range surrounded by the partition wall 41 and are set in the range where the toilet 40 and the bathtub 50 are placed. Among the areas of Room 1, the areas other than the bed area A, the dining area B, the toilet area C, the bathroom area D, and the kitchen area E are set as other area F. Also, the area outside the entrance 70 is set as the outdoor area G. Note that the above areas are just examples, and in addition to the areas A to G described above, it is possible to arbitrarily set areas (such as a study or a sofa area, etc.) according to the furniture and the like placed in Room 1.

[0029] The server 130 estimates the behavior based on information such as the information (location information) of the area where the target person is located, data on the posture of the target person, the time (the time when the target person is located in a predetermined area), and the stay time in the area.

[0030] Figure 2 is a table showing the relationship between the area where the target person is located and the behavior of the target person. In the server 130, several types of behaviors (such as "going to bed", etc.) shown in the above table are set according to the actions assumed from information such as the location information and posture data of the target person (for example, taking a lying posture in the bed area A, etc.). When the server 130 detects the above actions, it estimates that the actions corresponding to the actions have been performed. Hereinafter, an example of the method by which the server 130 estimates the behavior of the target person will be described using the above table.

[0031] When the target person is located in the bed area A (bedroom), the server 130 presumes that the target person's action is either "going to bed" or "sleeping (while in bed)". For example, if the server 130 detects the target person in the bed area A at the preset bedtime in the schedule, and the target person's posture is a lying position, the server 130 presumes that the target person is about to go to bed (the target person's action is "going to bed"). Also, for example, after "going to bed", if the server 130 detects that the target person has been in a lying position for a predetermined period or more, the server 130 presumes that the target person is "sleeping".

[0032] Also, when the target person is located in the toilet area C, the server 130 presumes that the target person's action is either "urination" or "defecation". More specifically, when the detection of the target person by the target person detection sensor 110 becomes impossible at the entrance / exit of the toilet area C, the server 130 determines that the target person is located in the toilet area C (is in the toilet). When the detection of the target person resumes thereafter, the server 130 determines that the target person has exited the toilet area C. For example, if the target person's staying time in the toilet (the period from entering the toilet area C to exiting) is less than a preset period, the server 130 presumes that the target person's action is "urination". If the target person's staying time in the toilet is equal to or more than the above period, the server 130 presumes that the target person's action is "defecation".

[0033] Also, when the target person is located in the study (not shown), the server 130 presumes that the target person's action is either "working" or "hobby". For example, if the target person's staying time in the study (the time spent in the study) is less than a preset period, the server 130 presumes that the target person's action is "hobby". If the target person's staying time in the study is equal to or more than the above period, the server 130 presumes that the target person's action is "working".

[0034] Also, when the target person is located in the kitchen area E, the server 130 presumes that the target person's action is either "cooking" or "washing dishes". For example, when the server 130 detects that the target person is located in the kitchen area E before the preset meal time, it presumes that the target person's action is "cooking", and when it detects that the target person is located in the kitchen area E after the meal time, it presumes that the target person's action is "washing dishes".

[0035] Also, when the target person is located in the bathroom area D, the server 130 presumes that the target person's action is either "taking a bath" or "washing face". More specifically, at the entrance of the bathroom area D, when the detection of the target person by the target person detection sensor 110 becomes impossible, the server 130 determines that the target person is located in the bathroom area D (entering the bathroom (washroom)), and when the detection of the target person resumes thereafter, it determines that the target person has exited the bathroom area D. For example, when the staying time of the target person in the bathroom (washroom) (the period from entering to exiting the bathroom area D) is less than the preset period, the server 130 presumes that the target person's action is "washing face", and when the staying time of the target person in the toilet is equal to or more than the above period, the server 130 presumes that the target person's action is "taking a bath".

[0036] Also, when the target person is located in the dining area B, the server 130 presumes that the target person's action is either "having a meal", "taking a rest", or "watching TV". For example, when the server 130 detects that the target person is located in the dining area B at the preset meal time, it presumes that the target person's action is "having a meal". Also, for example, when the staying time of the target person in the dining area B is less than the preset period, the server 130 presumes that the target person's action is "taking a rest", and when the staying time of the target person in the dining area B is equal to or more than the above period, the server 130 presumes that the target person's action is "watching TV".

[0037] In addition, even when the subject is located in the sofa area (not shown), the server 130 can presume that the subject's action is one of "eating", "resting", and "watching TV". For example, when the server 130 detects that the subject is located in the sofa area at a preset meal time, it presumes that the subject's action is "eating". Also, for example, when the subject's staying time in the sofa area is less than a preset period, the server 130 presumes that the subject's action is "resting", and when the subject's staying time in the sofa area is equal to or more than the above period, the server 130 presumes that the subject's action is "watching TV".

[0038] In addition, when the subject is located in the outdoor area G, the server 130 presumes that the subject's action is "going out" or "recreation". More specifically, when the server 130 no longer detects the subject at the entrance / exit 70, it presumes that the subject is located in the outdoor area G (the subject's action is "going out", etc.).

[0039] Note that the above-described method for presuming the subject's action is just an example, and the method for presuming the subject's action is not limited to the above-described example. Specifically, the subject's action may be presumed using any of the staying time in the area, the current time, and the information of the calendar (schedule). In addition to the above-described example, the server 130 can presume the subject's action by various methods using the data obtained by the subject detection sensor 110.

[0040] Hereinafter, an outline of the process for presuming the subject's fall risk by the health risk presumption system 100 configured as described above will be described.

[0041] The health risk estimation system 100 estimates the fall risk of a subject by observing the variation in the walking speed of the subject using the walking speed data of the subject obtained by the subject detection sensor 110. Here, the "variation in walking speed" is the difference between the walking speed of the subject at a certain time and the subsequent walking speed of the subject. In the present embodiment, the difference between the walking speed of the subject before performing a certain action and the walking speed of the subject after performing the above action is observed as the "variation in walking speed".

[0042] Here, as an influence of fatigue on human body movements, it is known that movements become slow. From this, when the walking speed decreases after a certain action, as a factor, there may be a relationship with temporary physical or mental fatigue associated with the above action. In particular, when the width of the variation in the walking speed of the subject before and after a certain action is large, it is assumed that the fatigue associated with the above action has greatly affected the decrease in the walking speed of the subject. For example, when the width of the variation in the walking speed increases before and after "defecation (large)", it is assumed that the walking speed has decreased due to the burden on the circulatory system caused by the act of defecation. Also, for example, when the width of the variation in the walking speed increases before and after "sleep", it is assumed that the sleep is light and the fatigue recovery is insufficient, or there is a disorder due to the accumulation of sleep deprivation.

[0043] From this, when the width of the variation in the walking speed of the subject is larger than the width of the previous variation in the walking speed, there is a possibility that the physical and mental functions of the subject have temporarily deteriorated compared to before, and as a result, there is a possibility that the temporary fall risk associated with the deterioration of the physical function and the like has increased. Here, since the walking speed varies under various influences, simply estimating the fall risk by looking only at the information on the variation in the walking speed lacks accuracy.

[0044] Therefore, the health risk estimation system 100 according to this embodiment preferably estimates the risk of falling by comparing the values of the fluctuations in the walking speed at present and in the past when similar actions are performed by each other. Hereinafter, with reference to FIGS. 3 to 5, specific descriptions of the processes ( "walking speed variability detection process", "walking speed measurement process", "specific action estimation process") by the health risk estimation system 100 will be given.

[0045] First, the "walking speed variability detection process" shown in the flowchart of FIG. 3 will be described below. The walking speed variability detection process is executed, for example, constantly.

[0046] In step S101 of the walking speed variability detection process, the server 130 executes the "walking speed measurement process" and acquires the walking speed data of the subject obtained in the "walking speed measurement process". Hereinafter, the "walking speed measurement process" will be described with reference to the flowchart of FIG. 4.

[0047] In the walking speed measurement process, the server 130 acquires the position information of the subject based on the data obtained by the subject detection sensor 110 (step S201). Thereby, the server 130 acquires the position information of the subject before walking. Each of the acquired data is stored by the server 130.

[0048] Next, the server 130 determines whether or not the posture of the subject has changed (step S202). When the server 130 detects, based on the data obtained by the subject detection sensor 110, that the subject has changed the posture from a certain posture (for example, sitting position) to another posture (for example, standing position), it determines that "the posture of the subject has changed". When the posture of the subject has changed, the server 130 estimates that the subject has started walking.

[0049] When the server 130 determines that the posture of the target person has not changed (step S202: NO), it proceeds to the process of step S201. On the other hand, when the server 130 determines that the posture of the target person has changed (step S202: YES), it measures the walking speed of the target person based on the data obtained by the target person detection sensor 110 (step S203).

[0050] Next, the server 130 determines whether the walking of the target person has ended (step S204). For example, when the movement of the target person is no longer detected, the server 130 determines that the walking has ended. When the server 130 determines that the walking of the target person has not ended (step S204: NO), it proceeds to the process of step S203.

[0051] On the other hand, when the server 130 determines that the walking of the target person has ended (step S204: YES), it acquires the position information of the target person based on the data obtained by the target person detection sensor 110 (step S205). By executing the process of step S205, the server 130 can acquire the position information of the target person after the walking has ended.

[0052] Next, the server 130 refers to the movement trajectory table stored in the server 130 in advance (step S206). Here, the "movement trajectory table" indicates the movement route (movement trajectory) of the target person from the starting point (position information before walking) to the arrival point (position information after walking has ended). The movement trajectory table includes, for example, the movement trajectories between each specific area. Specifically, the movement trajectory table includes, for example, "the movement trajectory between the bed area A and the toilet area C", "the movement trajectory between the bed area A and the dining area B", "the movement trajectory between the bed area A and the outdoor area G", etc. (see Fig. 6(a)).

[0053] Next, the server 130 determines whether the movement trajectory of the target person can be specified (step S207). Specifically, the server 130 determines whether the movement trajectory based on the position information of the target person before walking acquired in step S201 and the position information of the target person after walking acquired in step S205 can be specified by the movement trajectory table. For example, when the position information in step S201 and step S205 is complete and the combination of the above position information is included in the "movement trajectory table", the server 130 determines that "the movement trajectory of the target person can be specified".

[0054] If the server 130 determines in step S207 that the movement trajectory of the target person can be specified, it outputs the movement trajectory of the target person (step S208) and proceeds to step S102 of the "walking speed variability detection process" (step S209).

[0055] Also, if the server 130 determines in step S207 that the movement trajectory of the target person cannot be specified, since the specification of the movement trajectory is impossible, it processes the movement trajectory as non-existent (steps S210, S211) and proceeds to step S102 of the "walking speed variability detection process" (step S209). When the server 130 proceeds to the walking speed variability detection process, it ends the walking speed measurement process.

[0056] As described above, in step S101 of the "walking speed variability detection process", the server 130 can obtain the walking speed data before the specific action estimated in the "specific action estimation process" described later by executing the "walking speed variability detection process".

[0057] Next, in step S102 of the "walking speed variability detection process" shown in FIG. 3, the server 130 executes the "specific action estimation process" and obtains the estimation result of the action (specific action) of the target person in the "specific action estimation process". Hereinafter, the "specific action estimation process" will be described using the flowchart of FIG. 5.

[0058] In the specific action estimation process, the server 130 detects changes in the movement and posture of the target person based on the data obtained by the target person detection sensor 110, and acquires the position information of the target person (step S301). Each of the acquired data is stored by the server 130.

[0059] Next, the server 130 determines whether the target person has moved to a specific area (specific area) based on the above position information (step S302). As the specific area, each area shown in FIG. 1 (for example, the bed area A, the dining area B, the toilet area C, the bathroom area D, etc.) can be adopted. When the server 130 determines that the target person has not moved to the specific area (step S302: NO), it proceeds to the process of step S301.

[0060] When the server 130 determines that the target person has moved to the specific area (step S302: YES), it acquires the stay time of the area where the target person is located, the current time, and the calendar (schedule) information (steps S303 and S304), and determines whether it is possible to estimate the specific action of the target person (step S305). Here, the "specific action" refers to an action that can be specified in the specific area. As the "specific action", an action that can particularly significantly detect the variation in the walking speed of the target person in the subsequent process (walking speed variability detection process) can be adopted. In the present embodiment, a plurality of types of specific actions are set according to the assumed actions of the target person. More specifically, in the present embodiment, as the above specific action, each action shown in FIG. 2 (for example, "sleep" or "excretion", etc.) is adopted. The server 130 determines that "it is possible to estimate the specific action of the target person" when the information necessary for the estimation of the above specific action (information acquired in steps S303 and S304, etc.) is complete.

[0061] If the server 130 determines in step S305 that the estimation of the specific behavior of the target person is possible, it estimates the specific behavior of the target person based on each piece of information (information such as the position information, posture, time, and residence time in the area of the target person) obtained in steps S301, S303, and S304, outputs the behavior estimation result (step S306), and proceeds to step S103 of the "Walking Speed Fluctuation Detection Process" (step S307).

[0062] Also, if the server 130 determines in step S305 that the estimation of the specific behavior of the target person is impossible, since the estimation of the specific behavior of the target person is impossible, it processes the situation as having no specific behavior (steps S308 and S309), and proceeds to step S103 of the "Walking Speed Fluctuation Detection Process" (step S307). When the server 130 proceeds to the walking speed fluctuation detection process, it ends the specific behavior estimation process.

[0063] Next, in step S103 of the walking speed fluctuation detection process shown in FIG. 3, the server 130 determines whether or not there is a specific behavior of the target person (step S103). That is, the server 130 determines whether or not there is a specific behavior estimated in the specific behavior estimation process (step S305). If the server 130 determines that there is no specific behavior (step S103: NO), it proceeds to the process of step S101.

[0064] If the server 130 determines in step S103 that there is a specific behavior (step S103: YES), it executes the "Walking Speed Measurement Process" shown in FIG. 4 again and acquires the walking speed data of the target person obtained in the "Walking Speed Measurement Process" (step S104). The process performed in the "Walking Speed Measurement Process" is the same as the content described in step S101.

[0065] In step S104 above, the server 130 can acquire the walking speed data after the specific behavior by executing the "Walking Speed Fluctuation Detection Process".

[0066] Next, the server 130 calculates the value of the change in the walking speed of the subject associated with the specific action by comparing the walking speed data before the specific action (the walking speed data acquired in step S101) with the walking speed data after the specific action (the walking speed data acquired in step S104). Specifically, the server 130 extracts the walking speed data with the same movement trajectory from among the walking speed data before and after the specific action, and subtracts the walking speed before the specific action from the walking speed after the specific action.

[0067] In FIG. 6(a), the movement trajectories of the subject before and after the action of "sleep" (specific action) are shown. The server 130 compares the walking speed data with the same movement trajectory (for example, the movement trajectory between the bed area A and the toilet area C) among the walking speed data acquired before and after the specific action. In this case, the walking speed of the subject who moved from the toilet area C to the bed area A immediately before "sleep" (before going to bed) can be compared with the walking speed of the subject who moved from the bed area A to the toilet area C immediately after "sleep" (after waking up). The server 130 acquires the value obtained by the above calculation (hereinafter, the above value is referred to as "walking speed variability data") (step S105). The server 130 stores the walking speed variability data obtained as described above.

[0068] Next, the server 130 compares the current walking speed variability data (the current walking speed variability data) obtained in step S105 with the previous walking speed variability data (step S106). More specifically, the server 130 extracts, from the walking speed variability data stored by the server 130 up to now, the walking speed variability data that is the same as the current walking speed variability data and corresponds to a specific behavior and was previously acquired, and compares the current walking speed variability data with the (extracted) previous walking speed variability data that is the comparison target. As the walking speed variability data (previous walking speed variability data) to be compared, it is possible to adopt the walking speed variability data a predetermined period before (for example, one day ago, one week ago, one month ago, one year ago, etc.) based on the time when the current walking speed variability data was acquired. Also, as the walking speed variability data to be compared, it is also possible to adopt data indicating the average value, median value, etc. of the walking speed variability data within a predetermined period (for example, in the past one year, etc.).

[0069] Next, the server 130 evaluates the fall risk of the subject based on the comparison result between the current walking speed variability data and the previous walking speed variability data in step S106 (step S107). The server 130 evaluates the fall risk of the subject, for example, using the value obtained by subtracting the previous walking speed variability data from the current walking speed variability data (the difference in walking speed variability data).

[0070] The graph shown in FIG. 6(b) shows the relationship between the walking speed variability data (the variability of the walking speed) and the time series (date and time). Note that the walking speed variability data shown in the graph has the same movement trajectory. When the walking speed variability data at a certain point in time (for example, the current time) is smaller compared to the previous walking speed variability data (for example, the average value or median value of the previous walking speed variability data), as shown by the points surrounded by the broken line in FIG. 6(b), it is shown that the width of the current walking speed variability is larger than before.

[0071] If the value obtained by subtracting the previous walking speed variability data from the current walking speed variability data is equal to or less than a predetermined reference value, the server 130 can evaluate (estimate) that the risk of the subject falling is high. Further, the server 130 can display the evaluation result on the staff terminal 120. After executing the process of step S107, the server 130 ends the walking speed variability detection process.

[0072] According to the health risk estimation system according to the present embodiment as described above, by comparing the current and past walking speed variability data when similar specific actions are performed, the risk of temporary falls can be suitably estimated. That is, when the width of the walking speed of the subject before and after a certain specific action (for example, "sleep" etc.) is larger than in the past, it is estimated that the walking speed has decreased (become slower) due to the factors caused by the above specific action, and consequently, it is estimated that the risk of temporary falls is high. Thus, in the present embodiment, by aligning the situations (specific actions) and comparing the current and past walking speed variability data, the risk of falls can be suitably estimated.

[0073] Also, in the present embodiment, in the walking speed variability detection process, among the walking speed data before and after a specific action, the walking speed data based on walking with similar movement trajectories are compared with each other to calculate the walking speed variability data. According to the above configuration, by aligning the movement trajectories and comparing the widths of the walking speeds of the subject before and after a specific action, the accuracy of estimating the risk of falls can be improved.

[0074] As described above, the health risk estimation system 100 according to the present embodiment an action detection unit (subject detection sensor 110) capable of detecting the actions of the subject in room 1, an action estimation unit (server 130) capable of estimating the actions of the subject (steps S301 to S306) based on the detection result of the subject detection sensor 110, Before and after the action, a walking data detection unit (subject detection sensor 110) capable of detecting walking data (walking speed data), which is data related to the walking of the subject (step S101, step S104). Based on the detection result of the walking data detection unit (subject detection sensor 110), a walking variability data acquisition unit (server 130) that acquires walking variability data (walking speed variability data) indicating the variation between the walking speed data of the subject before the action and the walking speed data of the subject after the action (step S105). A health risk estimation unit (server 130) capable of estimating the health risk of the subject based on the comparison result between first walking variability data (current walking speed variability data), which is the walking variability data at a first time, and second walking variability data (past walking speed variability data), which is the walking variability data of an action similar to the action of the first walking variability data and is from a time earlier than the first time (step S108). It is equipped with the above components.

[0075] By configuring in this way, the health risk of the subject can be preferably estimated. That is, by comparing the current and past walking speed variability data when similar specific actions are performed, the risk of temporary falls (health risk) can be preferably estimated.

[0076] Also, the walking data detection unit (server 130) is Capable of detecting the walking speed data of the subject without contact.

[0077] By configuring in this way, the burden on the subject when measuring walking speed data can be reduced.

[0078] Also, the walking data detection unit (subject detection sensor 110) is Detecting the walking speed of the subject as the walking data.

[0079] By configuring in this way, it is possible to estimate the health risk of the target person based on the walking speed of the target person.

[0080] In addition, the health risk estimation system 100 includes a movement trajectory estimation unit (server 130) that can estimate the movement trajectory of the target person's walking (steps S201 to S208) based on the detection result of the movement detection unit (server 130). The walking variability data acquisition unit (server 130) uses the walking speed data before and after the action, which is based on walking with similar movement trajectories to each other, to acquire the walking speed variability data.

[0081] By configuring in this way, it is possible to align the movement trajectories and compare the walking speed data before and after a specific action, thereby improving the accuracy of estimating the health risk.

[0082] In addition, the health risk estimation system 100 includes a time information acquisition unit (server 130) that can acquire information on the time related to the actions of the target person detected by the target person detection sensor 110. The action estimation unit (server 130) estimates the action of the target person based on the detection result of the target person detection sensor 110 and the acquisition result of the time information acquisition unit (server 130) (steps S303 to S306).

[0083] By configuring in this way, it is possible to estimate the actions of the target person in consideration of the time (time of day or stay time) related to the actions of the target person.

[0084] Note that the target person detection sensor 110 according to the present embodiment is an embodiment of the movement detection unit and the walking data detection unit according to the present invention. In addition, the server 130 according to the present embodiment is an embodiment of the action estimation unit, the walking variability data acquisition unit, the movement trajectory estimation unit, the health risk estimation unit, and the time information acquisition unit according to the present invention.

[0085] The above describes one embodiment of the present invention. However, the present invention is not limited to the above configuration, and various modifications are possible within the scope of the invention described in the claims. For example, each process executed by the health risk estimation system 100 is not limited to those described above and can be arbitrarily changed.

[0086] Also, in the present embodiment, an example is shown in which both the movement and walking speed data of the subject are detected using the subject detection sensor 110. However, the present invention is not limited to this, and a device for detecting the movement of the subject and a device for detecting the walking speed of the subject may be provided separately.

[0087] Also, in the present embodiment, an example is shown in which the subject detection sensor 110 that non - contact detects the walking speed of the subject is adopted as the walking data detection unit. However, the present invention is not limited to this. For example, as the walking data detection unit, a device that detects the walking speed of the subject while being worn on the body of the subject, such as a wearable device equipped with an acceleration sensor, may be adopted.

[0088] Also, in the present embodiment, as the specific area, each area shown in FIG. 3 (bed area A, dining area B, toilet area C, bathroom area D, etc.) is exemplified. However, the present invention is not limited to this. For example, among the above - mentioned areas, only the areas (toilet area C and bathroom area D) partitioned by the partition wall 41 may be set as the specific area. According to this, based on whether the subject can be detected by the subject detection sensor 110, it is possible to preferably determine that the subject is located in the above - mentioned area and the staying time in the above - mentioned area.

[0089] Also, in the present embodiment, as the specific behavior, each behavior shown in FIG. 2 ("going to bed", "urination (small)", etc.) is exemplified. However, the present invention is not limited to this. As the specific behavior, various behaviors that can be estimated based on information such as the position information and posture data of the subject can be adopted.

[0090] In addition, in this embodiment, an example in which the health risk estimation system 100 is applied to Room 1 of a facility for the elderly has been shown. However, the present invention is not limited to this, and it can be applied to various other facilities, houses, etc.

[0091] In addition, in this embodiment, the subject detection sensor 110 (millimeter wave sensor) has been exemplified as the detection unit that detects the position, posture, and walking speed of the subject. However, the present invention is not limited to this, and for example, various devices such as cameras and other sensors capable of detecting the posture and walking speed of the subject can also be used.

[0092] In addition, in this embodiment, an example in which a configuration for detecting the walking speed of the subject is adopted as the data related to walking (walking data) has been shown. However, the present invention is not limited to this. For example, as the walking data, data such as the time taken by the subject to walk in a certain movement trajectory (walking time) and the periodicity of the subject's walking (the mode of the stance phase and the swing phase) may be adopted.

[0093] In addition, in this embodiment, an example in which various processes are executed by the server 130 has been shown. However, the present invention is not limited to this, and various processes can also be executed using the staff terminal 120 and various other devices (personal computers, tablet terminals, etc.).

Explanation of Signs

[0094] 1 Room 100 Health Risk Estimation System 110 Subject Detection Sensor 120 Staff Terminal 130 Server

Claims

1. An action detection unit capable of detecting the actions of a subject, An action estimation unit capable of estimating the actions of the subject based on the detection result of the action detection unit, A walking data detection unit capable of detecting walking data, which is data related to the walking of the subject, before and after the action, A walking variability data acquisition unit that acquires walking variability data indicating the variation between the walking data of the subject before the action and the walking data of the subject after the action based on the detection result of the walking data detection unit, A health risk estimation unit capable of estimating the health risk of the subject based on the comparison result between first walking variability data, which is the walking variability data at a first time, and second walking variability data, which is the walking variability data of an action similar to the action of the first walking variability data and is from a time earlier than the first time, A health risk estimation system comprising the above components.

2. The walking data detection unit, is capable of detecting the walking data of the subject in a non-contact manner, The health risk estimation system according to Claim 1.

3. The walking data detection unit, detects the walking speed of the subject as the walking data, The health risk estimation system according to Claim 1.

4. comprises a movement trajectory estimation unit capable of estimating the movement trajectory of the subject's walking based on the detection result of the action detection unit, The walking variability data acquisition unit, acquires the walking variability data using the walking data before and after the action, which is based on walking with similar movement trajectories to each other, The health risk estimation system according to Claim 1.

5. comprises a time information acquisition unit capable of acquiring information about the time related to the action of the subject detected by the action detection unit, The action estimation unit, estimates the action of the subject based on the detection result of the action detection unit and the acquisition result of the time information acquisition unit, The health risk estimation system according to any one of Claims 1 to 4.

Images