Cause inference system and cause inference method

Abstract

A cause inference system (1) is a cause inference system that infers a cause of a fall risk indicating a likelihood of a fall of a subject (50), including: a calculation unit (21) that acquires body movement data indicating a body movement of the subject (50) while walking, and calculates two or more walking parameters of the subject (50) based on the acquired body movement data; and a cause analysis unit (23) that infers one or more principal components included in the cause of the fall risk of the subject (50) based on the two or more walking parameters based on the two or more walking parameters, and outputs an inference result.

Description

TECHNICAL FIELD

[0001] The present application relates to a cause inference system and a cause inference method that infer a cause of a fall risk indicating a likelihood of a fall of a subject. BACKGROUND

[0002] In the past, a method of evaluating or determining a fall risk and the like has been proposed (for example, refer to Patent Literature 1). In Patent Literature 1, a method of evaluating a fall risk based on a single leg jump number that is an index indicating a motor function of a subject and a TUG (Timed Up to Go) test value that is one index of a muscle-skeletal instability syndrome is disclosed.

[0003] PRIOR ART DOCUMENTS

[0004] PATENT LITERATURE

[0005] Patent Literature 1: Japanese Patent Application Publication No. 2017-042618 SUMMARY

[0006] PROBLEMS TO BE SOLVED BY THE INVENTION

[0007] However, in the method described in Patent Literature 1, although it is possible to evaluate a fall risk, in the case where there is a fall risk, the cause thereof is not known. Therefore, there is a case where a monitor (for example, a caregiver or the like) of a subject cannot appropriately make a proposal to a person having a fall risk for reducing a fall risk.

[0008] Therefore, an object of the present application is to provide a cause inference system and a cause inference method that can infer a cause of a fall risk.

[0009] MEANS FOR SOLVING THE PROBLEMS

[0010] A cause inference system according to one aspect of the present application is a cause inference system that infers a cause of a fall risk indicating a likelihood of a fall of a subject, and includes: a calculation unit that acquires body movement data indicating a body movement at the time of walking of the subject, and calculates two or more walking parameters of the subject based on the acquired body movement data; and an inference unit that infers one or more principal components based on the two or more walking parameters, which are included in the cause of the fall risk of the subject, based on the two or more walking parameters, and outputs an inference result.

[0011] Further, a cause inference method relating to a technical solution of the present application is a cause inference method of inferring a cause of a fall risk indicating a possibility of a fall of a subject, acquiring body movement data indicating a body movement of the subject at the time of walking; calculating two or more walking parameters of the subject on the basis of the acquired body movement data; inferring one or more principal components included in the cause of the fall risk of the subject on the basis of the two or more walking parameters, the one or more principal components being based on the two or more walking parameters, and outputting an inference result.

[0012] Effects of Invention

[0013] According to the cause inference system and the like relating to a technical solution of the present application, it is possible to infer a cause of a fall risk. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 is a diagram showing an outline configuration of the cause inference system relating to Embodiment 1.

[0015] Figure 2 is a block diagram showing a functional configuration of the cause inference system relating to Embodiment 1.

[0016] Figure 3 is a diagram showing an example of an expression in which the risk analysis section calculates a fall risk value relating to Embodiment 1.

[0017] Figure 4 is a flowchart showing an action performed before an inference action of the cause inference system relating to Embodiment 1.

[0018] Figure 5 is a diagram showing an example of the first correspondence information.

[0019] Figure 6 is a diagram showing an example of the second correspondence information.

[0020] Figure 7 is a flowchart showing an inference action of the cause inference system relating to Embodiment 1, the inference action inferring a cause of a fall risk.

[0021] Figure 8 is a block diagram showing a functional configuration of the cause inference system relating to Embodiment 2.

[0022] Figure 9 is a flowchart showing an action of the cause inference system relating to Embodiment 2.

[0023] Figure 10 is a diagram showing an example of a correspondence relationship between a cause and an intervention method.

[0024] Figure 11 is a diagram showing a displacement in a vertical direction of a body of a subject at the time of walking.

[0025] Figure 12 is a graph showing the result of frequency analysis in the case where the cognitive function of the subject is normal.

[0026] Figure 13 is a graph showing the result of frequency analysis in the case where the cognitive function of the subject is decreased.

[0027] Figure 14 is a block diagram showing the functional configuration of the cause inference system relating to Embodiment 3.

[0028] Figure 15 is a flowchart showing the operation of the cause inference system relating to Embodiment 3.

[0029] Figure 16A is a flowchart showing an example of the operation of the risk determination section relating to Embodiment 3.

[0030] Figure 16B is a flowchart showing another example of the operation of the risk determination section relating to Embodiment 3. DETAILED DESCRIPTION

[0031] Hereinafter, the embodiments will be described with reference to the drawings. Note that the embodiments described below each show an example or a specific embodiment. Thus, numerical values, shapes, materials, components, arrangement positions of components, connection modes of components, steps, orders of steps, and the like shown in the following embodiments are examples, and do not limit the present application. Furthermore, regarding components of the embodiments described below, components not recited in independent claims are arbitrary components.

[0032] In addition, each drawing is a schematic view, and is not necessarily strictly illustrated. Furthermore, in each drawing, the same reference signs are given to substantially identical components, and there are cases where the same components are omitted or simplified.

[0033] Furthermore, in the present specification, relational terms not only indicate strict relationships, but also mean relationships within a range of substantial equivalence, for example, including a difference of several percent points.

[0034] (Embodiment 1)

[0035] [1-1. Outline configuration of cause inference system]

[0036] Figure 1 is a graph showing the outline configuration of the cause inference system 1 relating to the present embodiment. As shown in Figure 1 , the cause inference system 1 is provided with a measurement device 10, an inference device 20, an input device 30, and a display device 40.

[0037] The reason inference system 1 generates motion image data by measuring the body movement of the subject 50 while walking (in the middle of walking) by the measuring device 10 such as a camera. The measuring device 10 is provided, for example, at the ceiling or wall of a nursing home or care facility, and photographs the room regularly. The inference device 20 analyzes the walking pattern of the subject 50 based on the motion image data photographed (generated) by the measuring device 10, and infers the reason for the fall risk of the subject 50. The inference result is displayed on the display device 40. In addition, the motion image data is an example of body movement data. Furthermore, the subject 50 is an example of a subject.

[0038] The reason inference system 1 using such a measuring device 10 can evaluate the past inference result and the current inference result of the subject 50 by storing the motion image data photographed regularly by the measuring device 10. Furthermore, the reason inference system 1 can infer the reason for the fall risk of the subject 50 without the subject 50 noticing. In addition, the measuring device 10 is not limited to photographing the subject 50 regularly.

[0039] [1-2. Functional configuration of reason inference system]

[0040] Referring to Figure 2 The functional configuration of the reason inference system 1 according to the present embodiment will be described. Figure 2 is a block diagram showing the functional configuration of the reason inference system 1 according to the present embodiment. The reason inference system 1 is a system that rapidly infers the reason for the fall risk of the subject 50 by measuring the body movement of the subject 50 while walking.

[0041] As shown in Figure 2 , the reason inference system 1 includes the measuring device 10, the inference device 20, the input device 30, and the display device 40.

[0042] The measuring device 10 is a device for measuring the body movement of the subject 50 while walking. In the present embodiment, the measuring device 10 is a camera for photographing the motion image data of the subject 50 while walking. The measuring device 10 can be either a camera using a CMOS (Complementary Metal Oxide Semiconductor) image sensor or a camera using a CCD (Charge Coupled Device) image sensor.

[0043] In addition, the frame rate (the number of image data per 1 second included in the moving image data) is not particularly limited, and can be, for example, 40 fps (frames per second) or 60 fps.

[0044] The inference device 20 infers the cause of the fall risk of the subject 50 on the basis of the analysis of the gait pattern of the subject 50 by the moving image data captured by the measurement device 10, and outputs the inference result of the cause of the fall risk of the subject 50 to the display device 40. By so doing, the inference device 20 can, for example, notify the caregiver who cares for the subject 50 of the inference result of the cause of the fall risk of the subject 50, so that the caregiver can make a more appropriate proposal (intervention) to the subject 50 for reducing the fall risk. In addition, the cause inference system 1 can make the caregiver aware of the fall risk of the subject 50 by notifying the subject 50 of the cause of the fall risk even in a case where, for example, the caregiver does not notice that the subject 50 has the fall risk. In addition, the cause inference system 1 can make the subject 50 aware of the fall risk by notifying the subject 50 of the cause of the fall risk in a case where the subject 50 is not aware of the fall risk.

[0045] The inference device 20 has a calculation section 21, a risk analysis section 22, a cause analysis section 23, and a storage section 24.

[0046] The calculation section 21 acquires the measurement result (for example, the moving image data) from the measurement device 10, and calculates the gait parameter on the basis of the acquired measurement result. The calculation section 21, for example, acquires the moving image data captured by the measurement device 10 as the body movement data indicating the body movement of the subject 50 at the time of walking. In addition, the calculation method of the gait parameter on the basis of the moving image data is not particularly limited, and can be performed, for example, by image analysis of the moving image data.

[0047] In addition, it is known that a person in whom at least one of muscle strength, muscle mass, balance sense, and cognitive function has decreased is different in body movement at the time of walking from a person (a healthy person) in whom at least one of muscle strength, muscle mass, balance sense, and cognitive function has not decreased. Therefore, the gait parameter includes at least one of walking speed, stride, joint angle, displacement of the waist or the head, which has a correlation with at least one of muscle strength, muscle mass, balance sense, and cognitive function. The gait parameter includes at least two of walking speed, stride, joint angle, displacement of the waist or the head. In addition, the joint angle is, for example, the angle of the knee joint.

[0048] The risk analysis section 22 analyzes the fall risk of the subject 50 on the basis of the gait parameter. The risk analysis section 22, for example, analyzes the fall risk of the subject 50 by calculating the fall risk value on the basis of the calculation formula shown below. Figure 3 The risk analysis section 22 is an example of the 2nd determination section.

[0049] Figure 3 This is a diagram illustrating an example of the formula used by the risk analysis unit 22 in this embodiment to calculate the fall risk value. Figure 3 The scores X1, X2, and X3 shown are based on walking parameter values. For example, score X1 could be based on stride length, score X2 on walking speed, and score X3 on waist position. Furthermore, scores can be based on two or more walking parameters; for example, score X1 could be based on both stride length and walking speed. Additionally, in Figure 3 The diagram only shows the principal components described later (see reference). Figure 6 The fall risk value can be calculated based on two or more principal components, as described later. For example, the fall risk value can also be calculated based on each of the principal components described later.

[0050] like Figure 3 As shown, the risk analysis unit 22 calculates a fall risk value by adding scores X1, X2, X3 and a score related to fall history. Scores X1 and X2 are, for example, based on walking parameters corresponding to muscle strength. Alternatively, score X1 may be based on walking speed, and score X2 on stride length. Furthermore, muscle strength is correlated with walking speed and stride length (see below). Figure 6 Furthermore, the score X3 can be based, for example, on gait parameters corresponding to the balance system (e.g., sense of balance). The score X3 can also be based on lumbar displacement. Additionally, muscle mass is correlated with lumbar displacement (see below). Figure 6 ).

[0051] Furthermore, the score regarding fall history can be based on factors such as the presence or absence of falls or the number of falls. By including a score regarding fall history in the fall risk value, a proper assessment of fall risk can be made even when muscle strength and mass are normal. Additionally, the risk analysis unit 22 can obtain information about fall history, for example, via the input device 30, but it can also obtain such information by reading it from the storage unit 24.

[0052] Risk Analysis Department 22 Outputs and Passes Figure 3 The formula shown corresponds to the analysis result of the fall risk value. The risk analysis unit 22 can output either the presence or absence of fall risk, or the level of fall risk (e.g., "high," "medium," "low," etc.). There are no special restrictions as long as the fall risk level has three or more levels. Furthermore, the risk analysis unit 22 can also output the fall risk value.

[0053] in addition, Figure 3 The formula shown is an example; the fall risk value can also be calculated based on walking parameters.Figure 3 The fall risk value is calculated by a formula other than the formula shown. The fall risk value can be calculated, for example, by assigning a predetermined weight to the scores X1 to X3 and the like. Further, the fall risk value can be calculated, for example, using at least one of addition, subtraction, multiplication, division, and the like. Further, the fall risk value can be calculated using a numerical value regarding cognitive function as well. That is, the fall risk can be analyzed also taking into account the cognitive function of the subject 50.

[0054] Referring back to Figure 2 , the cause analysis section 23 analyzes the cause of the fall risk indicating the likelihood of the fall of the subject 50 based on the walking parameters. The cause analysis section 23 analyzes the cause of the fall risk of the subject 50 based on, for example, the walking parameters and correspondence information indicating the correspondence relation of the physical strength index of a person and the walking parameters. The physical strength index indicates the physical strength or the motor ability of a person, and includes, for example, items measured in physical strength measurement and the like. The physical strength index includes, for example, grip strength, leg muscle strength, single-leg standing with eyes open, stepping (e.g., repeated lateral jumps), and the like. Further, in the physical strength index, body composition inferred from the measurement result of a body composition meter can be included. The body composition is, for example, the measurement result of a body composition meter using the BIA method (Bioelectrical Impedance Analysis). Note that the walking parameters are not included in the physical strength index. Further, the cause includes a main cause (component) that has an influence on the fall risk of a person. The cause analysis section 23 is an example of an inference section.

[0055] The storage section 24 is a storage device that stores various data acquired or calculated by each processing section. The storage section 24 can store, for example, the moving image data acquired from the measurement device 10, and can store the walking parameters calculated by the calculation section 21. Further, the storage section 24 can store, for example, the analysis results of the risk analysis section 22 and the cause analysis section 23.

[0056] For example, in a case where the temporal change in the long-term of the body movement of the subject 50 at the time of walking is intended to be analyzed, the calculation section 21 can cause the storage section 24 to store the moving image data of the subject 50 or the calculated walking parameters. Further, for example, in a case where the temporal change in the long-term of the analysis result of the subject 50 is intended to be analyzed, the risk analysis section 22 or the cause analysis section 23 can cause the storage section 24 to store the analysis result. Note that the period of the long-term is not particularly limited, and can be, for example, one week, one month, or one year. Further, the moving image data, the walking parameters, and the analysis result will be also referred to as information based on body movement hereinafter.

[0057] By so doing, the inference device 20 is able to make a determination, for example, regarding the current fall risk, based on information (e.g., walking parameters, etc.) based on the body movement of the subject 50 at the time of the past walking. The inference device 20 is able to make a determination, for example, regarding the presence or absence of the current fall risk.

[0058] Further, in the storage section 24, for example, a program for each processing section to execute the cause inference method of the embodiment and information data used at the time of performing the cause analysis are also stored. The storage section 24 is realized by a semiconductor memory or an HDD (Hard Disk Drive), etc.

[0059] In addition, the inference device 20 can not have the risk analysis section 22. The inference device 20 is only required to be configured to be able to infer the cause of the fall risk of the subject 50.

[0060] In addition, each processing section of the inference device 20 can be realized by one processor, microcomputer, or dedicated circuit having each function, or can be realized by a combination of two or more of the processor, microcomputer, or dedicated circuit. Further, the calculation section 21 and the cause analysis section 23 can be configured to include a communication module (communication circuit) to perform wired or wireless communication. In this case, the calculation section 21 is only required to be able to communicate with the measurement device 10, and the communication method (communication specification, communication protocol) of the calculation section 21 is not particularly limited. Further, the cause analysis section 23 is only required to be able to communicate with the display device 40, and the communication method (communication specification, communication protocol) of the cause analysis section 23 is not particularly limited. In this way, the calculation section 21 can have a function as a retrieval section, and the cause analysis section 23 can have a function as an output section.

[0061] In addition, the inference device 20 is, for example, a personal computer, but can be a server device. Further, the inference device 20 can be provided in a building in which the measurement device 10 is provided, or can be provided outside the building.

[0062] The input device 30 is a user interface to accept input of prescribed information from the subject 50. The input device 30, for example, accepts input of information regarding the fall history of the subject. The input device 30 is realized by a hardware key (hardware button), a slide switch, a touch panel, etc.

[0063] The display device 40 displays an image based on the analysis result of the reason for the fall risk output from the inference device 20. The display device 40 is specifically a monitor device constituted by a liquid crystal panel or an organic EL panel, or the like. As the display device 40, an information terminal such as a television, a smartphone or a tablet terminal, a wearable terminal, or the like can also be used. The communication between the inference device 20 and the display device 40 is, for example, wired communication, but can also be wireless communication in the case where the display device 40 is a smartphone, a tablet terminal, or a wearable terminal.

[0064] [1-3. Action of the reason inference system]

[0065] Next, the action of the reason inference system 1 according to the present embodiment will be described with reference to Figure 4 and Figure 5 Figure 4 is a flowchart showing the action of the reason inference system 1 according to the present embodiment, which is executed before the inference action. Specifically, Figure 4 shows the action executed before the reason analysis for the fall risk of the subject 50 is performed.

[0066] As shown in Figure 4 , the calculation section 21 acquires the first correspondence information based on the measurement result with respect to the physical index of the person (S11). The calculation section 21 acquires, for example, the first correspondence information showing the correspondence relationship of the fall risk for each physical index and the measurement result with respect to the physical index. Figure 5 is a diagram showing an example of the first correspondence information D1. In addition, the fall risk for each physical index will be also referred to as a sub-index fall risk hereinafter. In addition, the first correspondence information is an example of the information showing the relationship of the physical index of the person and the fall risk.

[0067] As shown in Figure 5 , the first correspondence information D1 is a diagram showing the correspondence relationship of the physical index including "grip strength" "single-leg standing with eyes open" "fall history" and the like and the sub-index fall risk including "high" "medium" "low". With respect to the "grip strength", an example is shown in which the sub-index fall risk is "high" if the grip strength is less than 10 kgw, the sub-index fall risk is "medium" if it is around 15 kgw, and the sub-index fall risk is "low" if it is 20 kgw or more. In addition, Figure 5 the items and values shown in Figure 5 are an example and are not limited thereto. In addition,

[0068] ​Further, scores are assigned to the sub-index fall risk "high", "medium", and "low", respectively. For example, 2 points are assigned to the sub-index fall risk "high", 1 point is assigned to "medium", and 0 point is assigned to "low", but the assignment of scores is not limited thereto. Further, the calculation section 21 can also acquire a threshold value for determining whether there is a fall risk based on the scores of the sub-index fall risk. The calculation section 21 can also acquire, for example, a threshold value for the operation value of each score of the physical strength index that is operated. The operation is, for example, addition, but can also be at least one of subtraction, multiplication, and division. Further, the operation can also be weighted addition or the like. Hereinafter, an example in which the operation is addition and the operation value is the total value of the scores of each physical strength index will be described. Assume that the calculation section 21 has acquired, for example, 6 points as a first threshold value for determining that the fall risk of the subject 50 is "high" and 2 points as a second threshold value for determining that the fall risk of the user is "medium". The first threshold value and the second threshold value can also be stored in the storage section 24, for example.

[0069] Next, the calculation section 21 acquires second correspondence information D2 that indicates the correspondence relationship between the physical strength index and the walking parameter (S12). The calculation section 21 can also acquire the second correspondence information D2 via the input device 30, for example. Figure 6 is a diagram indicating an example of the second correspondence information D2. In addition, the second correspondence information is an example of information indicating the relationship between the physical strength index of a person and two or more walking parameters.

[0070] As shown in Figure 6 , the second correspondence information D2 is information in which the principal component corresponding to each of the components 1 to 4 included in the cause of the fall risk, the physical strength index, and the walking parameter are established in correspondence. The principal component indicates an element of the body that is related to the fall risk of a person and is set in advance. The principal component includes, for example, "muscle strength", "balance", "agility", and "muscle mass". If an example is described with respect to the component 1, the physical strength index corresponding to the principal component "muscle strength" is "grip strength" and "leg muscle strength", and the walking parameter corresponding to "muscle strength" is "walking speed" and "stride length". In other words, the second correspondence information D2 indicates that "walking speed" and "stride length" can be used instead of "grip strength" and "leg muscle strength" in the estimation of the cause of the fall risk.

[0071] The principal component "muscle strength" means that one of the causes of the fall of a person is the muscle strength of the person. The physical strength indexes "grip strength" and "leg muscle strength" are indexes indicating the state of the principal component "muscle strength". The walking parameters "walking speed" and "stride length" are walking parameters that have a correlation with the physical strength indexes "grip strength" and "leg muscle strength".

[0072] In addition, the correlation here can include a correlation of the value of the grip strength and the value of the walking parameter in the case where the physical strength index is "grip strength" and the walking parameter is "walking speed". The correlation can include, for example, a correlation in which the walking speed is 2 km / h and the grip strength is 10 kgw.

[0073] In addition, the correlation of the physical strength index and the walking parameter can be obtained by regression analysis of the measurement results of the physical strength index and the walking parameter of a plurality of persons, but the method of obtaining the correlation is not limited thereto.

[0074] In addition, the "joint angle" in the component 2 includes, for example, a difference in joint angle of the left and right legs. The joint angle here is an angle of a joint related to walking, for example, an angle of a knee joint. The difference in the left and right legs is, for example, a difference in the angle of the knee joint of the left leg and the right leg.

[0075] In addition, the "joint angle" in the component 3 includes, for example, a magnitude of the joint angle. The joint angle here is an angle of a joint related to walking, for example, a magnitude of the angle of a knee joint.

[0076] In addition, the "shift of the waist" in the component 4 includes a shift in the position of the waist. The walking parameter in the component 4 can include, for example, a "shift of the head" in addition to or simultaneously with the "shift of the waist".

[0077] Referring again to Figure 4 , the calculation section 21 stores the first correspondence information D1 and the second correspondence information D2 in the storage section 24 (S13).

[0078] Next, the operation of the reason inference system 1 to infer the reason for the fall risk will be described with reference to Figure 7 Figure 7 is a flowchart showing the operation of the reason inference system 1 to infer the reason for the fall risk.

[0079] As shown in Figure 7 , the calculation section 21 obtains the motion image data of the subject 50 when walking from the measurement device 10 (S21). The motion image data can be data photographed when the subject 50 walks normally, or data photographed when the subject 50 walks in a prescribed place for the purpose of inferring the reason for the fall risk. The prescribed place can be, for example, a passage including a walking surface with a marker. In addition, the motion image data can be motion image data obtained by photographing the subject 50 from a plurality of viewpoints.

[0080] ​Next, the calculation section 21 calculates the walking parameters of the subject 50 based on the moving image data (S22). The method of calculating the walking parameters by the calculation section 21 is not particularly limited, and for example, the calculation can be performed by image analysis of the moving image data. The calculation section 21 can calculate the feature points of the subject 50 from the image data, for example, and calculate the walking parameters based on the movement trajectories of the feature points. The calculation section 21 can calculate the feature points by the background subtraction method when the moving image data of the subject 50 walking in the passage described above is acquired. The calculation section 21 outputs the walking parameters to the risk analysis section 22.

[0081] Next, the risk analysis section 22 determines whether the subject 50 has a fall risk based on the walking parameters (S23). The risk analysis section 22, for example, calculates scores from the walking parameters, and determines whether the subject 50 has a fall risk based on the plurality of calculated scores. The risk analysis section 22 calculates the score of each walking parameter based on the first correspondence information D1 and the second correspondence information D2 stored in the storage section 24, for example. The risk analysis section 22, for example, acquires that the walking speed 2 km / h corresponds to the grip strength 10 kgw when the walking parameter is the walking speed and the walking speed is 2 km / h. Further, the risk analysis section 22 acquires that the score of the walking speed 2 km / h is 2 points based on the first correspondence information.

[0082] The risk analysis section 22, for example, calculates the fall risk value by calculating the scores described above from the walking parameters and adding the plurality of calculated scores as shown in the following equation. Figure 3 Further, the risk analysis section 22, for example, determines that there is a fall risk when the total value of the plurality of scores, that is, the fall risk value is equal to or greater than a threshold value. The threshold value in this case is a numerical value for determining the presence or absence of a fall risk. The threshold value can be a fixed value, or can be set for each subject 50.

[0083] Further, the risk analysis section 22 can determine the degree of fall risk when the first threshold value (for example, 6 points) and the second threshold value (for example, 2 points) are set as the threshold values, for example. The risk analysis section 22 can determine that there is a fall risk when the degree of fall risk is equal to or greater than a predetermined degree (for example, "medium" or more), for example.

[0084] Note that the method of determining the presence or absence of a fall risk by the risk analysis section 22 is not limited to the above. The risk analysis section 22 can determine that there is a fall risk when the walking speed is equal to or less than a threshold value, for example. That is, the risk analysis section 22 can determine whether there is a fall risk based on the numerical value of the walking parameter.

[0085] The risk analysis unit 22 outputs the determination result to the cause analysis unit 23. Furthermore, the risk analysis unit 22 can also cause the storage unit 24 to store the determination result. The determination result output by the risk analysis unit 22 is an example of the second determination result.

[0086] If the causal analysis unit 23 obtains a judgment result indicating a risk of fall from the risk analysis unit 22 (Yes in S23), then it calculates the degree of influence on the fall risk for each principal component based on the physical strength indicators related to the walking parameters (S24). For example, based on the second corresponding information D2, the causal analysis unit 23 obtains the correlation between the walking parameters "walking speed" and "stride length" and the principal component "muscle strength". The causal analysis unit 23 calculates the degree of influence of the principal component "muscle strength" on the fall risk based on walking speed and stride length. The causal analysis unit 23 may also calculate the degree of influence on the fall risk based on the scores of walking speed and stride length. For example, the causal analysis unit 23 calculates the sum of the scores of walking speed and stride length as the degree of influence of the principal component "muscle strength" on the fall risk. The causal analysis unit 23 can also be said to infer the principal components included in the causes of fall risk by performing principal component analysis based on walking parameters.

[0087] Cause Analysis Department 23 according to Figure 6 The components 1 to 4 shown above calculate the degree of influence for each principal component. The degree of influence can be based on the absolute value of the score (e.g., 6 points) or the relative value of the score (e.g., 50%). When the degree of influence is based on the score value, the cause analysis unit 23 can also be described as performing a process of summing up each score included in the fall risk value calculated by the risk analysis unit 22 according to each principal component.

[0088] Next, the cause analysis unit 23 infers the cause of the fall risk of the measured person 50 based, for example, on the degree of influence of each principal component (S25). That is, the cause analysis unit 23 infers the cause of the fall risk based on two or more walking parameters. Based on two or more walking parameters, the cause analysis unit 23 infers one or more principal components included in the cause of the fall risk of the measured person 50 from multiple principal components. For example, the cause analysis unit 23 can infer the principal component with the highest degree of influence as the cause of the fall risk of the measured person 50, or it can infer the principal component with an influence degree of a predetermined degree or higher as the cause of the fall risk of the measured person 50.

[0089] Next, the cause analysis unit 23 outputs information indicating the deduction result to the display device 40 (S26). That is, the cause analysis unit 23 causes the display device 40 to display the deduction result.

[0090] Next, the inference device 20 causes the storage unit 24 to store at least one of the motion image data, walking parameters, and inference results (S27).

[0091] Furthermore, if the cause analysis unit 23 obtains a determination result from the risk analysis unit 22 indicating that there is no risk of falling (in S23, it is "no"), then the process of deducing the cause of the fall risk ends.

[0092] [1-4. Effects, etc.]

[0093] As described above, the cause inference system 1 of this embodiment is a cause inference system for inferring the cause of fall risk indicating the possibility of a fall by the subject 50. It includes: a calculation unit 21, which acquires motion image data (an example of body motion data) representing the body movement of the subject 50 during walking, and calculates two or more walking parameters of the subject 50 based on the acquired body motion data; and a cause analysis unit 23 (an example of an inference unit), which infers one or more principal components based on the two or more walking parameters that are included in the cause of the fall risk of the subject 50, and outputs the inference result.

[0094] Therefore, the cause analysis unit 23 can infer the cause of the fall risk of the subject 50 based on two or more walking parameters. Specifically, the cause analysis unit 23 can infer one or more principal components based on two or more walking parameters. Thus, the cause inference system 1 of this embodiment can infer the cause of the fall risk.

[0095] Furthermore, the cause analysis unit 23 infers two or more principal components based on information indicating the relationship between physical fitness indicators and fall risk, and information indicating the relationship between physical fitness indicators and two or more walking parameters.

[0096] Therefore, by using the aforementioned information, the cause analysis unit 23 enables the subject 50 to infer one or more principal components based on two or more walking parameters without requiring physical strength measurements. Thus, the cause inference system 1 can more easily infer the cause of the fall risk. In other words, the cause analysis unit 23 infers the capabilities possessed by the subject 50 by using the aforementioned information.

[0097] Furthermore, the cause inference system 1 also includes a risk analysis unit 22 (an example of the second determination unit) that determines the fall risk of the subject 50 based on two or more walking parameters. Moreover, the cause analysis unit 23 infers two or more principal components when the risk analysis unit 22 determines that the subject 50 has a fall risk.

[0098] Therefore, the cause inference system 1 can determine the presence or absence of fall risk. By outputting the determination result, the result can be notified to the subject 50 and the caregiver. In addition, since the processing load of the cause analysis unit 23 can be reduced, the cause inference system 1 is energy-efficient.

[0099] In addition, one or more principal components include at least one of muscle strength, muscle mass, balance, and cognitive function.

[0100] Therefore, when there is a risk of falling for the subject 50, the causal analysis unit 23 can infer whether the cause is physical decline or cognitive decline.

[0101] In addition, at least two of the following walking parameters are required: walking speed, stride length, joint angle, and lumbar displacement.

[0102] Therefore, the cause analysis unit 23 can infer the cause of the subject 50's fall risk based on at least two of the walking speed, stride length, joint angle, and waist displacement that can be obtained from motion image data. That is, the cause inference system 1 can infer the cause of the subject 50's fall risk based on motion image data captured from the subject 50's daily walking pattern, without performing measurements (e.g., physical fitness measurements) used to infer the cause of the fall risk. Thus, the cause inference system 1 can more easily infer the cause of the fall risk.

[0103] Furthermore, as described above, the reasoning method of the reasoning system 1 of this embodiment is a reasoning method for inferring the cause of fall risk that represents the possibility of a fall for the subject 50. It obtains body motion data representing the body movement of the subject 50 during walking (S21), calculates two or more walking parameters of the subject 50 based on the obtained body motion data (S22), infers one or more principal components based on the two or more walking parameters that are included in the cause of fall risk of the subject 50 (S25), and outputs the reasoning result (S26).

[0104] Thus, it achieves the same effect as the cause inference system 1 mentioned above.

[0105] (Implementation Method 2)

[0106] The following is a reference. Figures 8-13 The cause inference system 1a of this embodiment will be explained. In addition to inferring the causes of fall risk, the cause inference system 1a of this embodiment is also characterized by proposing intervention methods to reduce fall risk based on the inference results.

[0107] Furthermore, in the following description, the focus is on the differences from Embodiment 1, and the same reference numerals are assigned to the same configurations as in Embodiment 1. There are cases where the description is omitted or simplified.

[0108] [2-1. Functional Structure of a Cause Inference System]

[0109] While referring to Figure 8The functional configuration of the cause inference system 1a in this embodiment will be explained. Figure 8 This is a block diagram illustrating the functional configuration of the cause inference system 1a in this embodiment.

[0110] like Figure 8 As shown, the cause inference system 1a replaces the inference device 20 included in the cause inference system 1 of embodiment 1 with an inference device 20a. The inference device 20a further includes a suggestion determination unit 25 in addition to the inference device 20 of embodiment 1.

[0111] The recommendation determination unit 25, based on the inference result of the cause of the fall risk of the subject 50, performs processing for caregivers and others to provide interventions to the subject 50 corresponding to the inference result. For example, the recommendation determination unit 25 performs processing to suggest methods with higher intervention efficiency to caregivers. For example, the recommendation determination unit 25 performs a determination process to suggest methods (improvement menus) with higher intervention efficiency to caregivers based on two or more principal components included in the inference result. A method with higher intervention efficiency refers to a method (improvement menu) that can provide interventions to the subject 50 suitable for the cause of the subject 50's fall risk. That is, a method with higher intervention efficiency is a method that can effectively reduce the fall risk of the subject 50. In other words, the recommendation determination unit 25 determines (decides) a method to reduce the fall risk of the subject 50. The recommendation determination unit 25 performs the above determination, for example, based on the degree of influence of the two or more principal components on the fall risk. The recommendation determination unit 25 is an example of the first determination unit.

[0112] Furthermore, it is suggested that the determination unit 25 may, for example, store the determination result in the storage unit 24 when it is desired to analyze the long-term changes in the determination result. Additionally, the determination result may be based on information about body movement.

[0113] Furthermore, the recommendation determination unit 25 may also be configured to include a communication module (communication circuit) for wired or wireless communication. In this case, the recommendation determination unit 25 only needs to be able to communicate with the display device 40, and the communication method (communication specification, communication protocol) of the recommendation determination unit 25 is not particularly limited.

[0114] [2-2. Actions of the Cause Inference System]

[0115] Then, while referring to Figure 9 and Figure 10 The operation of the cause inference system 1a in this embodiment will be explained. Figure 9 This is a flowchart illustrating the operation of the cause inference system 1a in this embodiment. Specifically, Figure 9This indicates actions proposed to reduce the risk of falls based on inferences about the causes of the fall risk in the 50% of the subjects. Additionally, Figure 9 The processing of S21 to S25 shown is consistent with the implementation method 1. Figure 7 The same applies; further explanation omitted.

[0116] like Figure 9 As shown, if the cause analysis unit 23 deduces the cause of the fall risk of the measured person 50 (S25), it outputs the deduction result to the suggestion determination unit 25.

[0117] If the recommendation determination unit 25 obtains the inference result from the cause analysis unit 23, it determines the intervention method to be recommended to the caregiver or other caregiver of the subject 50 based on the inference result (S31). For example, the recommendation determination unit 25 determines the intervention method corresponding to the inference result from a plurality of intervention methods stored in the storage unit 24. Figure 10 This is a diagram illustrating an example of the correspondence between causes and intervention methods.

[0118] like Figure 10 As shown, the recommendation judgment unit 25 determines the intervention method corresponding to the degree (proportion) of the influence of "muscle strength," "muscle mass," "balance," and "cognition" on fall risk. For example, if the proportion of "muscle strength" among "muscle strength," "muscle mass," "balance," and "cognition" is the highest, the recommendation judgment unit 25 determines that "exercise improvement menu (slow-twitch muscle fibers)" is the recommended intervention method. Thus, by effectively improving muscle strength through slow-twitch muscle fiber training, the risk of falls can be easily reduced.

[0119] Furthermore, the recommendation section 25 suggests that, for example, if the proportion of "muscle mass" is the highest among "muscle strength," "muscle mass," "balance," and "cognition," then the "Exercise Improvement Menu (Fast-Twitch Muscles)" is the recommended intervention method. Thus, by exercising fast-twitch muscle fibers, muscle mass can be effectively increased, thereby easily reducing the risk of falls.

[0120] Thus, the recommendation and judgment section 25 suggests, for example, an exercise program for fall prevention and motor function improvement if the proportion of either "muscle strength" or "muscle mass" is the highest.

[0121] Furthermore, the recommendation judgment unit 25 may determine that a "meal improvement menu" is a recommended intervention method if, for example, the proportions of "muscle strength" and "muscle mass" in "muscle strength," "muscle mass," "balance," and "cognition" are similar (e.g., consistent). Thus, the recommendation judgment unit 25 may suggest a meal improvement method instead of muscle training if, for example, the proportions of the main muscle components such as "muscle strength" and "muscle mass" are similar. Additionally, the similarity may also refer to the difference between the two proportions being within a predetermined value. This predetermined value could be, for example, 10%, 20%, or other values.

[0122] in addition, Figure 10 The proportions shown are calculated, for example, based on the sum of the scores of each principal component.

[0123] in addition, Figure 10 The term "cognition" indicates the degree to which a decline in cognitive function affects walking. It is known that individuals exhibiting cognitive decline or signs of cognitive decline differ from those without cognitive decline (healthy individuals) in their walking movements. Therefore, the degree of "cognition's" impact can be calculated, for example, based on walking parameters. Furthermore, cognitive function represents the ability to recognize, store, or judge information.

[0124] The following is an example of a method for analyzing (evaluating) cognitive function. However, the methods for analyzing cognitive function are not limited to the following.

[0125] When the gait parameters include head position, the cause analysis unit 23 performs frequency analysis on head displacement, for example. The cause analysis unit 23 analyzes head displacement (e.g., indicating...) Figure 11 The signal representing the time-varying position of the head is subjected to a Discrete Fourier Transform. That is, the cause analysis unit 23 performs a frequency transform process to convert the signal representing the body's displacement from the time domain to the frequency domain. Furthermore, Figure 11 This is a graph showing the vertical displacement of the body of the person being measured 50 while walking. Furthermore, the displacement of the head is an example of the position of the center of gravity, calculated, for example, by the calculation unit 21.

[0126] If the cognitive function of the subject is normal at 50%, then the result is... Figure 12 The analysis results shown indicate that, in cases where 50% of the subjects experienced a decline in cognitive function, the following results were obtained: Figure 13 The analysis results are as shown. Figure 12 This is a graph representing the frequency analysis results when the cognitive function of 50% of the subjects is normal. Figure 13 This is a graph representing the frequency analysis results when 50% of the subjects experienced a decline in cognitive function.

[0127] exist Figure 12 and Figure 13 In the analysis results shown, the lowest frequency peak (the highest level peak) represents the frequency peak of the walking cycle. In other words, the lowest frequency peak is the dominant frequency component. If the cognitive function of the subject 50 is normal, then the subject 50 is able to walk at a certain cycle. Therefore, in Figure 12 In, with Figure 13 Compared to the frequency peaks that represent the walking cycle, the peak levels are sharper and higher.

[0128] On the other hand, when the cognitive function of the subject 50 declines, the deviation in the walking cycle increases because the subject 50 becomes more difficult to walk at a certain cycle. Therefore, in Figure 13 In, with Figure 12 Compared to the lower peak level of the frequency peak representing the walking cycle, the lower swing of this frequency peak is wider.

[0129] Therefore, the cause analysis unit 23 analyzes the cognitive function of the subject 50 based on the frequency peaks representing the walking cycle of the subject 50 obtained through frequency analysis using discrete Fourier transform. For example, the cause analysis unit 23 analyzes the cognitive function of the subject 50 based on the peak level of the frequency peaks. The lower the peak level, the more the cause analysis unit 23 analyzes that the cognitive function of the subject 50 is declining. For example, the cause analysis unit 23 sets a threshold value for the peak level (…). Figure 12 and Figure 13 In the above cases (as shown in the diagram), cognitive function is judged to be normal, and a score representing normal cognitive function is assigned. The score representing normal cognitive function is preset, for example, it can be 0 points.

[0130] On the other hand, the cause analysis unit 23 determines that cognitive function has declined, for example, if the peak level is less than a threshold, and assigns a score indicating a decline in cognitive function. The score indicating a decline in cognitive function is preset, for example, it could be 2 points. In addition, the threshold could be stored in the storage unit 24, for example.

[0131] Refer again Figure 9 The recommendation determination unit 25 outputs information indicating the determination result to the display device 40 (S32). That is, the recommendation determination unit 25 causes the display device 40 to display the determination result. The determination result output by the recommendation determination unit 25 is an example of the first determination result.

[0132] Furthermore, the inference device 20 causes the storage unit 24 to store at least one of the motion image data, walking parameters, inference result, and determination result (S33). It is suggested that the determination unit 25 may also cause the storage unit 24 to store the determination result.

[0133] [2-3. Effects, etc.]

[0134] As described above, the cause analysis unit 23 of the cause inference system 1a of this embodiment infers two or more principal components. Furthermore, the cause inference system 1a also includes a suggestion determination unit 25 (an example of the first determination unit) that determines an intervention method for reducing the risk of falls for the subject 50 based on two or more principal components and outputs the determination results.

[0135] Therefore, the recommendation determination unit 25 can notify caregivers and others of appropriate intervention methods for the causes of fall risk. Furthermore, the cause inference system 1a can urge the subject 50 to reduce fall risk through appropriate intervention methods even when caregivers and others lack knowledge about reducing fall risk.

[0136] In addition, it is recommended that the decision-making unit 25 determine an intervention method to reduce the influence of the principal component with the greatest influence on the risk of falling among two or more principal components.

[0137] Therefore, it is recommended that the determination unit 25 be able to output an intervention method that effectively reduces the fall risk of the subject 50.

[0138] (Implementation Method 3)

[0139] The following is a reference. Figures 14-16B The cause inference system 1b of this embodiment will be explained. The cause inference system 1b of this embodiment is characterized by processing related to fall risk based on past time series data. Past time series data refers to time series data obtained earlier than the present, such as time series data of the most recent week, the most recent month, the most recent year, or other types.

[0140] Furthermore, in the following description, the focus is on the differences from Embodiment 2, and the same reference numerals are assigned to the same configurations as in Embodiment 2. There are cases where the description is omitted or simplified.

[0141] [3-1. Functional Components of a Cause Inference System]

[0142] While referring to Figure 14 The functional configuration of the cause inference system 1b in this embodiment will be explained. Figure 14 This is a block diagram illustrating the functional configuration of the cause inference system 1b related to this embodiment.

[0143] like Figure 14 As shown, the cause inference system 1b replaces the inference device 20a included in the cause inference system 1a of embodiment 2 with an inference device 20b. The inference device 20b, in addition to the inference device 20a of embodiment 2, also includes an analysis unit 26 and a risk determination unit 27. Alternatively, the inference device 20b may not include a suggestion determination unit 25.

[0144] The analysis unit 26 analyzes information based on past walking movements. For example, the analysis unit 26 obtains the temporal changes in time-series data related to fall risk through statistical processing. Furthermore, by analyzing the temporal changes in the time-series data, the analysis unit 26 can both identify the trend of the time-series data and calculate a threshold for determining the fall risk at the current time. For example, the analysis unit 26 can also calculate a threshold for the fall risk value of a user by analyzing the temporal changes in the user's past fall risk values.

[0145] Furthermore, the following describes an example of the analysis unit 26 analyzing time-series data related to fall risk, but it is not limited to this. It can also analyze at least one time-series data point, such as walking parameters, principal components of the cause (e.g., muscle strength), and the determination result. The risk determination unit 27 analyzes the time-series data of the principal components of the cause using the analysis unit 26. For example, it can determine whether the influence of muscle strength on fall risk has decreased, i.e., whether the fall risk has been reduced through the intervention method-based menu.

[0146] The risk assessment unit 27 determines the fall risk of the subject 50 based on the analysis results of the analysis unit 26. The risk assessment unit 27 determines the fall risk of the subject 50 based on time-series data containing at least one piece of information from past walking motion information. For example, if the analysis unit 26 calculates a threshold for the fall risk value of the subject 50, the risk assessment unit 27 can also determine whether there is a fall risk for the user based on whether the fall risk at the current time exceeds the threshold. That is, the analysis unit 26 can also use past time-series data to set a threshold for the walking motion information at the current time.

[0147] Therefore, the risk assessment unit 27 can make a judgment corresponding to the increase, for example, when the risk value of the measured person 50 falls increases sharply.

[0148] Furthermore, the risk assessment unit 27 may also be configured to include a communication module (communication circuit) for wired or wireless communication. In this case, the risk assessment unit 27 only needs to be able to communicate with the display device 40, and the communication method (communication specification, communication protocol) of the risk assessment unit 27 is not particularly limited.

[0149] [3-2. Actions of the Cause Inference System]

[0150] Then, while referring to Figures 15-16B The operation of the cause inference system 1b in this embodiment will be explained. Figure 15 This is a flowchart illustrating the operation of the cause inference system 1b in this embodiment.

[0151] like Figure 15As shown, the analysis unit 26 acquires time-series data of at least one of the following: walking parameters, fall risk, inference result, and judgment result (S41). The analysis unit 26 acquires this time-series data, for example, by reading it from the storage unit 24.

[0152] Next, the analysis unit 26 analyzes the time series data (S42). For example, if the analysis unit 26 has obtained walking parameters including walking speed in step S41, it can also calculate information representing the degree of change in walking speed based on the walking speed at a specified time point. The degree of change can be, for example, the difference between the walking speed at the specified time point and the walking speed at other times, or a ratio.

[0153] Furthermore, if the analysis unit 26 has obtained information including the fall risk value in step S41, it can also calculate information representing the degree of change of the fall risk value based on the fall risk value at a predetermined time point. The degree of change can be, for example, the difference between the fall risk value at the predetermined time point and the fall risk value at other times, or it can be a ratio.

[0154] Furthermore, the analysis unit 26, for example, obtains the degree of influence of the principal components in step S41 (e.g. Figure 10 In the case of the proportions shown, the trend of proportion changes can also be calculated for each principal component. The analysis unit 26 can, for example, generate a line graph representing this trend.

[0155] Furthermore, if the analysis unit 26 obtains a determination result including the intervention method in step S41, it can also calculate the trend of change in the intervention method. The analysis unit 26 can also, for example, calculate the number of times each of the multiple intervention methods is proposed during a specified period.

[0156] Furthermore, the analysis unit 26 can also perform statistical processing on the aforementioned numerical values ​​(e.g., degree of change, fall risk value, degree of impact, frequency). The statistical values ​​calculated in the statistical processing are, for example, the average, but can also be the maximum, minimum, median, or values ​​representing deviations (e.g., standard deviation).

[0157] The analysis unit 26 outputs the analysis results to the risk assessment unit 27.

[0158] Based on the analysis results, the risk assessment unit 27 performs a fall risk assessment process (S43). Alternatively, the risk assessment unit 27 can be said to perform a fall risk assessment process based on time series data. For example, the risk assessment unit 27 performs... Figure 16A and Figure 16B The determination process involves at least one of the parties shown. Figure 16A This is a flowchart illustrating an example of the operation of the risk assessment unit 27 in this embodiment. Figure 16AThis is a flowchart illustrating the case where time-series data of the walking parameters were obtained in step S41.

[0159] like Figure 16A As shown, the risk assessment unit 27 determines whether the change in walking parameters is above a predetermined value (S101). For example, the risk assessment unit 27 determines whether the change in walking speed is above a predetermined value. For example, the risk assessment unit 27 determines whether the walking speed has decreased by above a predetermined value.

[0160] If the change in the walking parameters is greater than or equal to a predetermined value (Yes in S101), the risk assessment unit 27 determines that the risk of falling has increased (S102). Furthermore, if the change in the walking parameters is less than the predetermined value (No in S101), the risk assessment unit 27 determines that the change in the risk of falling is small (S103). Additionally, the risk assessment unit 27 may, for example, determine that the risk of falling has decreased if the change in the walking parameters is a predetermined change. Moreover, the predetermined change may be, for example, a change that approaches the optimal value of the walking parameters.

[0161] Therefore, the cause inference system 1b is able to inform caregivers and others of the changing trend of the fall risk of the subject 50.

[0162] Then, while referring to Figure 16B Here is another example of the actions of the risk assessment department 27. Figure 16B A flowchart illustrating another example of the operation of the risk assessment unit 27 in this embodiment. Figure 16B This is a flowchart illustrating the case where time series data for inference results were obtained in step S41.

[0163] like Figure 16B As shown, the risk assessment unit 27 determines whether the proportion of a specified principal component in the inference result has decreased (S201). For example, the risk assessment unit 27 determines whether the proportion of "muscle strength" in the inference result has decreased. The specified principal component may be, for example, one of multiple principal components whose influence on the risk of falling is the highest at least once within a specified period.

[0164] If the proportion of the specified principal component decreases (yes in S201), the risk assessment unit 27 determines that an improvement effect brought about by the intervention method can be seen (S202). Furthermore, if the proportion of the specified principal component does not decrease (no in S201), the risk assessment unit 27 determines that no improvement effect brought about by the intervention method can be seen (S203). Alternatively, the risk assessment unit 27 may also determine "yes" in step S201 if the specified principal component decreases by a specified proportion or more.

[0165] Refer again Figure 15Next, the risk assessment unit 27 generates information indicating the assessment result (S44) and outputs the generated assessment result information to the display device 40 (S45). That is, the risk assessment unit 27 causes the display device 40 to display the assessment result.

[0166] Therefore, the causal inference system 1b can inform caregivers and others of the improvement effects brought about by the intervention method.

[0167] Furthermore, the timing of the aforementioned actions performed by the cause inference system 1 is not specifically limited, and they can also be performed periodically.

[0168] [3-3. Effects, etc.]

[0169] As described above, the cause inference system 1b of this embodiment also includes a risk determination unit 27 (an example of a third determination unit) that determines the risk of falling based on time series data of at least one of two or more walking parameters, inference results, and determination results of the risk analysis unit 22.

[0170] Therefore, the risk assessment unit 27 can determine the risk of falling based on the changes in two or more walking parameters, inference results, and the assessment results of the risk analysis unit 22 over time, thus enabling earlier detection of the risk of falling.

[0171] (Other implementation methods)

[0172] The above describes various embodiments (hereinafter referred to as embodiments, etc.), but the present invention is not limited to the above embodiments, etc.

[0173] For example, in the embodiments described above, the inference device does not have a measuring device, an input device, and a display device; that is, an example where the computing device and the measuring device, input device, and display device are separate has been described, but this is not a limitation. The inference device may also have the function of at least one of the measuring device, the input device, and the display device. In this case, the measuring device functions as a measuring unit that is part of the inference device, the input device functions as an input unit that is part of the inference device, and the display device functions as a display unit that is part of the inference device. For example, the cause inference system may also be composed of a single device.

[0174] Furthermore, in the above embodiments, an example of the inference device in the cause inference system being implemented by a single device has been described, but it can also be implemented by multiple devices. For example, the inference device can be implemented by one server device or by three or more server devices. When the cause inference system is implemented by multiple server devices, the components of the inference device can be allocated to the multiple server devices in any way.

[0175] Furthermore, in the above-described embodiments, each component may be constructed using dedicated hardware, or implemented by executing software programs suitable for each component. Each component may also be implemented by reading and executing software programs recorded on recording media such as hard disks or semiconductor memories using a program execution unit such as a CPU or processor.

[0176] Furthermore, in the embodiments described above, the processing to be performed by a specific processing unit may be performed by other processing units. Also, the processing order described in the flowcharts of the above embodiments is just one example. The order of multiple processes may be changed, or multiple processes may be executed in parallel.

[0177] Furthermore, in the above-described embodiments, each component can also be implemented by executing a software program suitable for each component. Each component can also be implemented by reading and executing a software program recorded on a recording medium such as a hard disk or semiconductor memory using a program execution unit such as a CPU or processor.

[0178] Furthermore, each component can also be implemented in hardware. For example, each component can be a circuit (or integrated circuit). These circuits can either form a single circuit as a whole, or they can be separate circuits. Moreover, these circuits can be general-purpose circuits or special-purpose circuits.

[0179] Furthermore, the segmentation of functional blocks in a block diagram is one example. Multiple functional blocks can also be implemented as a single functional block, or a single functional block can be divided into multiple functional blocks, or a portion of the functionality can be transferred to other functional blocks. Additionally, the functionality of multiple functional blocks with similar functions can be processed in parallel or time-divisionally by a single piece of hardware or software.

[0180] Furthermore, the inclusive or specific technical solutions of the present invention can also be implemented by a system, method, integrated circuit, computer program or a recording medium such as a computer-readable CD-ROM, or by any combination of the system, method, integrated circuit, computer program and recording medium.

[0181] In addition, the present invention also includes forms obtained by applying various modifications to the embodiments that can be conceived by those skilled in the art, or forms achieved by arbitrarily combining the constituent elements and functions of each embodiment without departing from the spirit of the present invention.

[0182] Label Explanation

[0183] 1. Cause Inference System (1a, 1b)

[0184] 21. Computing Department

[0185] 22 Risk Analysis Department (Second Judgment Department)

[0186] 23. Root Cause Analysis Department (Inference Department)

[0187] 25. Recommendation for Judgment Department (First Judgment Department)

[0188] 27 Risk Assessment Department (Third Assessment Department)

[0189] 50 subjects

[0190] D1 First Corresponding Information

[0191] D2 Second Corresponding Information

Claims

1. A causal inference system that infers the causes of fall risk, representing the likelihood of a subject falling. The above-mentioned reasoning system possesses the following characteristics: The calculation unit acquires body motion data representing the body movements of the subject during walking, and calculates two or more walking parameters of the subject based on the acquired body motion data; and The inference unit, based on the two or more walking parameters mentioned above, infers at least one principal component representing a bodily element related to the aforementioned fall risk of the subject, including at least one of muscle strength, muscle mass, balance, and cognitive function, and outputs the inference result. The inference unit infers one or more principal components based on information representing the relationship between the two or more walking parameters and the principal components. The above two or more walking parameters include at least two of the subject's walking speed, stride length, joint angle, and lumbar displacement.

2. The cause inference system as described in claim 1, The above inference part infers two or more principal components; The aforementioned cause inference system also includes a first determination unit, which uses the aforementioned two or more principal component determinations to determine intervention methods for reducing the risk of falls for the aforementioned subjects, and outputs determination results.

3. The cause inference system as described in claim 2, The first determination unit determines the intervention method used to reduce the influence of the principal component that has the greatest impact on the risk of falling among the two or more principal components.

4. The cause inference system as described in claim 2, It also has a second determination unit that assesses the fall risk of the subject based on two or more of the aforementioned walking parameters; When the second determination unit determines that the measured subject is at risk of falling, the inference unit infers two or more principal components.

5. The cause inference system as described in claim 4, It also has a third determination unit that determines the risk of falling based on time series data of at least one of the above two or more walking parameters, the above inference results, and the determination results of the second determination unit.

6. A causal inference method for inferring the causes of fall risk that indicate the likelihood of a subject falling. In the above methods of inferring causes, Obtain body motion data representing the body movements of the aforementioned subjects during walking; Based on the obtained body movement data, calculate two or more walking parameters for the subject. Based on the above two or more walking parameters, infer at least one principal component representing a bodily element related to the aforementioned fall risk of the subject, including at least one of muscle strength, muscle mass, balance, and cognitive function, based on the above two or more walking parameters, and output the inference result. In the above inference, based on information representing the relationship between the above two or more walking parameters and the above principal components, the above one or more principal components are inferred. The above two or more walking parameters include at least two of the subject's walking speed, stride length, joint angle, and lumbar displacement.

Images