JP2020146435A - Learning method, estimation method, and storage medium - Google Patents

Abstract

To improve estimation accuracy of risk of needing a nursing care.SOLUTION: A learning method of an embodiment causes a computer to perform acquiring processing, creating processing, and discriminating processing. The acquiring processing includes acquiring a posture of an object person and characteristics of external factors of the object person. The creating processing includes creating a time series model relating to the characteristics of the external factors based on the plurality of characteristics of the external factors acquired at different time points. The discrimination processing includes discriminating whether the object person has a disorder of an internal factors based on the acquired posture and the created time series model.SELECTED DRAWING: Figure 12

Description

Embodiments of the present invention relate to learning methods, estimation methods and learning programs.

Conventionally, the risk of elderly people is evaluated by determining locomotive syndrome (hereinafter referred to as "locomotive syndrome"), which indicates a state in which there is a high risk of requiring nursing care due to a disorder of the locomotive organ.

For this locomotive determination, a conventional technique of calculating the acceleration for the movement of each part for each predetermined part of the subject and determining the motor function using a neural network from the characteristic parameters obtained for each part from the calculated acceleration is used. Are known.

JP-A-2016-144598

However, the above-mentioned conventional technique does not evaluate the potential risk of requiring long-term care due to internal factors such as dementia, and has a problem that it may lead to erroneous determination.

In one aspect, it is intended to provide learning methods, estimation methods and learning programs that can improve the estimation accuracy of the risk of requiring long-term care.

In one plan, in the learning method, the computer executes a process of acquiring, a process of creating, and a process of discriminating. The acquisition process acquires the posture of the subject and the characteristics of the external factors of the subject. The process to be created creates a time-series model of the characteristics of the external factors based on the characteristics of the external factors acquired at different times. In the determination process, it is determined whether or not the subject has an internal factor disorder based on the acquired posture and the created model.

According to one embodiment, the accuracy of estimating the risk of requiring long-term care can be improved.

FIG. 1 is a block diagram showing a functional configuration example of the learning device and the estimation device according to the embodiment. FIG. 2 is an explanatory diagram illustrating user information. FIG. 3 is an explanatory diagram illustrating the creation of an aging model. FIG. 4 is an explanatory diagram for explaining the posture determination. FIG. 5 is an explanatory diagram illustrating modeling of an aging model. FIG. 6 is an explanatory diagram illustrating an example of regression analysis. FIG. 7 is an explanatory diagram illustrating dementia estimation. FIG. 8 is an explanatory diagram illustrating dementia estimation. FIG. 9 is a flowchart showing an example of processing related to the creation of an aging model. FIG. 10 is a flowchart showing an example of processing related to the learning phase. FIG. 11 is a flowchart showing an example of processing related to the estimation phase. FIG. 12 is an explanatory diagram showing an example of an output screen. FIG. 13 is an explanatory diagram showing a comparison example with the conventional estimation result. FIG. 14 is a flowchart showing an example of a modified example of dementia determination. FIG. 15 is an explanatory diagram illustrating an example of a dementia determination model. FIG. 16 is an explanatory diagram illustrating a determination example of the dementia determination model. FIG. 17 is a block diagram showing an example of a computer that executes a program.

Hereinafter, the learning method, the estimation method, and the learning program according to the embodiment will be described with reference to the drawings. Configurations having the same function in the embodiment are designated by the same reference numerals, and duplicate description will be omitted. The learning method, estimation method, and learning program described in the following embodiments are merely examples, and the embodiments are not limited. In addition, the following embodiments may be appropriately combined within a consistent range.

FIG. 1 is a block diagram showing a functional configuration example of the learning device and the estimation device according to the embodiment. As shown in FIG. 1, the learning device 1 uses the characteristics of external factors of the estimation target person to perform information processing (supervised learning) of the learning model 31 for estimating the risk of the estimation target person. It is a device. Specifically, the learning device 1 uses information including the characteristics of external factors of the learning target person as a teacher as an explanatory variable at the time of learning (learning phase), and the risk of making the correct answer in the learning target person as an objective variable. The learning model 31 is trained by machine learning.

Further, the estimation device 2 is an information processing device that estimates the risk possessed by the estimation target person by using the learning model 31 learned by the learning device 1 and using the characteristics of the external factors of the estimation target person. Specifically, the estimation device 2 applies information including the characteristics of external factors of the estimation target person to the learned learning model 31 at the time of estimation (estimation phase), so that the estimation target person has a risk. To estimate.

In the present embodiment, as an example, the characteristic of the external factor of the estimation target person is the walking mode (posture, gait, etc.) of the estimation target person, and the risk of requiring nursing care due to a fall or the like (fall risk) is estimated. This will be explained using an example. The characteristic of the external factor is not limited to the mode during walking, and may be, for example, a mode during exercise other than walking. In addition, the estimated risk may be a risk other than the fall risk (for example, the risk of developing dementia).

The learning device 1 has a collecting unit 10, a user information DB 11, a dementia discrimination unit 12, and a model learning unit 13.

The collection unit 10 collects information (age, characteristics of external factors, risk of correct answer determined by an expert, etc.) of the learning target as a teacher of the learning model 31 by operation input or file input by the user. It is a processing unit. The collection unit 10 stores the collected user information of each learning target person in the user information DB 11. Specifically, the collecting unit 10 stores the user information collected at each of a plurality of time points by periodic inspection or the like for each identification information (user ID or the like) that identifies the learning target person in the user information DB 11.

FIG. 2 is an explanatory diagram illustrating user information. As shown in FIG. 2, the user information 11A is information indicating the characteristics of external factors of the subject A, and in the present embodiment, it is information indicating the walking mode (posture, gait, etc.) of the subject A. is there. Specifically, the user information 11A has coordinates of "head", "neck", "waist", and "feet" indicating the posture of the subject A when walking. In addition, the user information 11A has coordinates of "walking speed", "gait", "gait", and "walking angle" indicating the gait of the subject A. Further, in the user information 11A, in addition to the above-mentioned information indicating the characteristics of the external factors of the target person A, the risk determination result (fall risk) by the age of the target person A at the time of collecting the user information 11A, an expert, etc. The correct answer label, etc., indicating the presence or absence of or the degree of risk) is attached.

The dementia determination unit 12 determines whether or not each learning target person has an internal factor disorder (dementia) based on the user information 11A for each learning target person stored in the user information DB 11. It is a department. Specifically, the dementia discrimination unit 12 has an aging model creation unit 12A and 12B.

The aging model creating unit 12A is a processing unit that creates a time-series aging model related to the characteristics of external factors from the user information 11A collected at each of a plurality of time points by a periodical examination or the like for each learning target person. Specifically, the aging model creation unit 12A changes the "walking speed", "step length", "gait", and "walking angle", which indicate the steps in the user information 11A, from the data at a plurality of time points in time series. Create an aging model showing (changes with aging).

The characteristics of external factors such as gait differ greatly depending on the posture of the learning subject. Therefore, the aging model creation unit 12A determines the posture of the learning target person based on the coordinates of the "head", "neck", "waist", and "foot" indicating the posture of the user information 11A.

FIG. 3 is an explanatory diagram illustrating the creation of an aging model. As shown in FIG. 3, when a person walks, there are various postures such as postures C1 to C6 due to, for example, the physiological curvature of the spine. For example, the physiological curvature of the spine includes deformation due to osteoporotic vertebral body fracture and disc degeneration centered on the lumbar spine. In addition, knee pain may cause the posture to collapse in order to balance.

The age-related changes in the characteristics of external factors (for example, walking speed) differ depending on the postures C1 to C6. For example, the posture C1 changes like the aging model M1, and the posture C2 changes like the aging model M2.

Therefore, the aging model creation unit 12A determines which of the postures C1 to C6 is the posture of the learning subject based on the coordinates of the "head", "neck", "waist", and "foot". , Create an aging model based on the determined posture.

FIG. 4 is an explanatory diagram for explaining the posture determination. As shown in FIG. 4, in the aging model creation unit 12A, the coordinates of the "head", "neck", "waist", and "foot" of the subjects A, B ... In the user information 11A are any of the postures C1 to C6. The postures of the subjects A, B ... Are determined depending on whether the conditions of are matched. In the illustrated example, the subject A is determined to be in the posture C2, and the subject B is determined to be in the posture C3.

Next, the aging model creation unit 12A creates an aging model in the determined posture by modeling the characteristics of the external factors of the learning target person in chronological order for each learning target person.

FIG. 5 is an explanatory diagram illustrating modeling of an aging model. As shown in FIG. 5, the aging model creation unit 12A describes the “walking speed”, “gait”, “gait”, and “walking angle”, which indicate the steps in the user information 11A, for the subjects A, B, and so on, respectively. The aging models M1, M2 ... In the above are created. Specifically, the aging model creation unit 12A performs regression analysis of the values plotted in chronological order (age order) for each of "walking speed", "step length", "step distance", and "walking angle". Then, the aging models M1, M2 ... Are obtained.

FIG. 6 is an explanatory diagram illustrating an example of regression analysis. As shown in FIG. 6, the aging model creation unit 12A plots the walking speeds obtained from April to September in chronological order in the user information 11A. Next, the aging model creating unit 12A obtains an aging model M relating to walking speed in the form of Y = ax + b by performing multiple regression based on the plotted values. In the following explanation, if the aging model is not distinguished for each subject or for each "walking speed", "step length", "step distance", and "walking angle", the aging model is used. It shall be called M.

Next, the aging model creating unit 12A adds identification information corresponding to the target person to the obtained aging model M, and then stores it in a memory or the like as the aging model information 30. As a result, in the learning phase and the estimation phase, the age model M of each subject can be acquired by referring to the age model information 30 based on the identification information of the subject.

Returning to FIG. 1, the dementia estimation unit 12B determines whether or not the subject has an internal factor disorder (dementia) based on the acquired posture and the created aging model M for each learning subject. It is a processing unit that determines whether or not.

Disorders of internal factors (dementia in this embodiment) possessed by the subject appear as significant changes in time series in the characteristics of external factors by posture. For example, when the subject has cerebrovascular dementia, the dementia progresses in stages, so the subject is compared with the aging model (hereinafter referred to as the reference model), which is the standard according to the aging of each posture. There is a significant difference in the aging model M of. In addition, in dementia with Lewy bodies, since the condition is repeated when the condition is good and when the condition is bad, fluctuations of a predetermined width with time series occur in the aging model M of the subject. The dementia estimation unit 12B determines whether or not the subject has an internal factor disorder (dementia) based on the time-series characteristics of the aging model M of the subject.

7 and 8 are explanatory views for explaining the dementia estimation. Specifically, as shown in FIG. 7, the dementia estimation unit 12B is included in the reference model MK for each posture preset in the memory or the like based on the posture determined from the user information 11A of the subject A. The reference model MK corresponding to the posture of the subject A is read from. As the reference model MK, data set in advance in a memory or the like may be used, or the average of a large number of subjects stored in the aging model information 30 may be used.

Next, the dementia estimation unit 12B also includes a reference model MK and an aging model M created for the gait of subject A (“walking speed”, “gait”, “gait”, “walking angle”). In addition, a statistical significance test is performed. Next, the dementia estimation unit 12B determines that subject A has dementia when there is a significant difference between the reference model MK and the aging model M in the significance test, and when there is no significant difference, the subject A is the subject. Person A determines that there is no dementia.

Further, as shown in FIG. 8, the dementia estimation unit 12B describes the aging model M created for the gait of the subject A (“walking speed”, “gait”, “gait”, “walking angle”). , Determines the presence or absence of fluctuations of a predetermined width with time series on a rule basis. Next, the dementia estimation unit 12B determines that the subject A has dementia when there is a fluctuation of a predetermined width with time series, and determines that the subject A has no dementia when there is no fluctuation.

Returning to FIG. 1, the model learning unit 13 is a processing unit that learns the learning model 31. Specifically, the model learning unit 13 uses the presence or absence of an internal factor disorder (dementia) determined by the dementia discrimination unit 12 and the user information 11A indicating the characteristics of the external factor of the subject as explanatory variables. , The learning model 31 is trained by machine learning with the risk given as the correct answer label as the objective variable.

As the learning model 31, for example, a neural network in which units imitating neurons in the brain are hierarchically connected from an input layer to an output layer via an intermediate layer can be applied.

For example, the model learning unit 13 inputs the value of the objective variable to the input layer of the learning model 31, and outputs the output value indicating the calculation result from the output layer. Then, the model learning unit 13 learns the parameters at each node of the neural network in the learning model 31 based on the comparison between the risk of the correct label and the output value. More specifically, the model learning unit 13 learns the parameters of the neural network by an error back propagation (BP) method or the like using the comparison result between the output value and the correct label.

The estimation device 2 has an input unit 20, a dementia discrimination unit 21, an estimation unit 22, and an output unit 23.

The collection unit 10 is a processing unit that receives input of information (age, characteristics of external factors, etc.) of the estimation target person by operation input or file input by the user. The collecting unit 10 outputs the information of the estimated target person who has received the input to the dementia discrimination unit 21.

The dementia discrimination unit 21 is a processing unit that determines whether or not the estimation target person has an internal factor disorder (dementia) based on the information of the estimation target person. The processing in the dementia discrimination unit 21 is the same as that in the dementia discrimination unit 12, although there is a difference in whether the processing target is a learning target person or an estimation target person. That is, the dementia discrimination unit 21 has the same configuration as the aging model creation unit 12A and the dementia estimation unit 12B. Therefore, the description of the processing content of the dementia discrimination unit 21 will be omitted.

The estimation unit 22 estimates the risk possessed by the estimation target person by using the learning model 31 that has been trained with the presence or absence of the obstacle of the internal factor and the characteristics of the external factor as explanatory variables and the risk as the objective variable. It is a processing unit. Specifically, the estimation unit 22 applies the presence or absence of a disorder of the internal factor determined by the estimation unit 22 to the learning model 31 and the characteristics of the external factor of the estimation target person, and applies the characteristics of the external factor of the estimation target person to the learning model 31. Estimate the risk. As an example, when the learning model 31 is a neural network, the estimation unit 22 inputs the presence or absence of a failure of an internal factor and a value corresponding to the characteristics of the external factor into the input layer of the learning model 31, and outputs the calculation result. Obtain the indicated output value (estimated risk) from the output layer.

The output unit 23 is a processing unit that outputs the estimation result of the estimation unit 22 by, for example, display output to a display or file output. From the output from the output unit 23, the user can know the risk that the estimation target person has.

FIG. 9 is a flowchart showing an example of processing related to the creation of an aging model. As shown in FIG. 9, when the process is started, the aging model creation unit 12A uses the user information DB 11 to display the characteristics of external factors (“head”, “neck”, “waist”, “feet”) in the learning target person. "Coordinates) is collected (S1).

Next, the aging model creation unit 12A determines the posture of the learning target person based on the collected characteristics of the external factors (coordinates of "head", "neck", "waist", and "foot") ( S2).

Next, the aging model creation unit 12A has a plurality of characteristics of external factors (“walking speed”, “gait”, “gait”, “walking angle”) in the learning target person from the user information DB 11. Data at the time of (S3) is collected.

Next, the aging model creating unit 12A uses data at a plurality of time points as a time-series model (aging model M) for each of the "walking speed", "step length", "gait", and "walking angle" indicating the gait. ) Is created (S4). Next, the aging model creating unit 12A stores the created aging model M in the aging model information 30 after giving the learning target person the corresponding identification information, and ends the process.

FIG. 10 is a flowchart showing an example of processing related to the learning phase. As shown in FIG. 10, when the process related to the learning phase is started, the dementia estimation unit 12B uses the user information DB 11 to display the characteristics of external factors in the learning subject (“head”, “neck”, “waist”, (Coordinates of "foot") are collected (S11).

Next, the dementia estimation unit 12B determines the posture of the learning subject based on the collected characteristics of the external factors (coordinates of "head", "neck", "waist", and "foot") (S12). ). Next, the dementia estimation unit 12B acquires a time-series model (aging model M) to which the identification information corresponding to the learning target is added from the aging model information 30 (S13).

Next, the dementia estimation unit 12B collects data at a plurality of time points regarding the characteristics of external factors (“walking speed”, “gait”, “gait”, “walking angle”) in the learning subject. Collect (S14).

Next, the dementia estimation unit 12B determines whether or not the learning subject has an internal factor disorder (dementia) based on the determined posture and the aging model M of the learning subject (S15). ).

Next, the model learning unit 13 was assigned as a correct answer label with the presence or absence of an internal factor disorder (dementia) determined by the dementia discrimination unit 12 and the characteristics of the external factors of the learning subject as explanatory variables. Learning data with risk as the objective variable is generated (S16). Next, the model learning unit 13 learns the learning model 31 by machine learning (supervised learning) based on the generated learning data (S17).

In the learning device 1, the learning of the learning model 31 is advanced by executing the above-mentioned processes S11 to S17 for each learning target person in the learning phase.

FIG. 11 is a flowchart showing an example of processing related to the estimation phase. As shown in FIG. 11, when the processing related to the estimation phase is started, the dementia discrimination unit 21 determines the characteristics of external factors (“head”, “neck”, “waist”, “feet”) in the estimation target person. Coordinates) are collected (S21).

Next, the dementia discriminating unit 21 discriminates the posture of the estimated target person based on the collected characteristics of the external factors (coordinates of "head", "neck", "waist", and "foot") (S22). ).

Next, the dementia discrimination unit 21 acquires a time-series model (aging model M) to which the identification information corresponding to the estimated target person is added from the aging model information 30 (S23). Next, the dementia discrimination unit 21 collects data at a plurality of time points regarding the characteristics of external factors (“walking speed”, “gait”, “gait”, “walking angle”) indicating the gait in the estimated subject. Collect (S24).

Next, the dementia discrimination unit 21 determines whether or not the presumed subject has an internal factor disorder (dementia) based on the discriminated posture and the aging model M of the presumed subject (S25). ).

Next, the estimation unit 22 generates estimation data including the presence or absence of an internal factor disorder (dementia) determined by the dementia determination unit 21 and the characteristics of the external factors of the estimation target person (S26). Next, the estimation unit 22 applies estimation data to the learning model 31 to estimate the risk of the estimation target person (S27). Next, the output unit 23 outputs the estimation result of the estimation unit 22 to a display screen or the like, and ends the process.

FIG. 12 is an explanatory diagram showing an example of an output screen. As shown in FIG. 12, the output screen 40 output by the output unit 23 has a gait display area 41, a model display area 42, and an estimation result display area 43.

The gait display area 41 is a display area showing the “walking speed”, “step length”, “gait”, “walking angle”, and the overall average of each value for the subject A. From the display content of the gait display area 41, the user can easily know the gait (“walking speed”, “gait”, “gait”, “walking angle”) of the target person A.

The model display area 42 is an area for displaying the aging model M for the subject A and the reference model MK corresponding to the posture of the subject A (posture C2 in the illustrated example). The display content of the model display area 42 allows the user to easily compare the aging model M of the gait of the subject A with the reference model MK corresponding to the posture of the subject A.

The estimation result display area 43 is an area for displaying the determination result of the disorder (dementia) of the internal factor by the dementia determination unit 21 and the estimation result of the estimation unit 22 for the subject A. From the display content of the estimation result display area 43, the user can easily confirm whether or not the subject A has an obstacle of an internal factor and the estimation result of the fall risk.

Next, a modified example of dementia determination will be described. In the above dementia determination, the disorder of internal factors (cognition) is based on gait (“walking speed”, “step length”, “step distance”, “walking angle”) as a characteristic of external factors in the subject. It is determined whether or not the patient has dementia.

However, even if there is an internal factor disorder (dementia), the walking speed may increase. In such a case, the subject may be erroneously determined to have no dementia, and the accuracy of estimating the potential risk of requiring long-term care will decrease due to internal factors such as dementia.

In addition, the characteristics of external factors in the subject may appear not only in gaits but also in meals and conversations. For example, in the case of dementia, choking due to aspiration may occur by simply putting food in the mouth without swallowing it. It is difficult to estimate the potential risk of requiring long-term care due to such aspiration from the characteristics of external factors such as gait.

In a modified example of dementia determination, in the processing of S15 and S25 for determining dementia, dementia determination based on the characteristics of external factors other than gait using a dementia logic different from the above dementia determination. I do. As a result, it is possible to improve the accuracy of estimating the potential risk of requiring long-term care due to internal factors such as dementia.

FIG. 14 is a flowchart showing an example of a modified example of dementia determination. As shown in FIG. 14, when the process of determining dementia is started, the dementia estimation unit 12B first determines dementia by the same gait as S15 described above (S31).

Next, the dementia estimation unit 12B collects data at a plurality of time points from the user information DB 11 regarding the characteristics (other than gait) of external factors in the subject (S32). The characteristics of external factors (other than gaits) include items such as "meal speed", "conversation speed", "repeat the same thing", and "forget to eat". Regarding the characteristics of these external factors (other than gaits), it is assumed that the data collected from each user is registered in the user information DB 11 in advance.

Next, the dementia estimation unit 12B makes a dementia determination using a dementia determination model based on a dementia logic different from S15, etc., based on the characteristics of external factors (other than gaits) collected for the subject. Do (S33).

FIG. 15 is an explanatory diagram illustrating an example of a dementia determination model. As shown in FIG. 15, the dementia determination model M3 for determining the presence or absence of dementia as the probability (appearance probability) is obtained by logistics regression based on the sample data D1 collected from each user.

Specifically, the characteristics of external factors (other than gaits) included in the sample data D1 of each user are "meal speed", "conversation speed", "repeat the same thing", "forget meal", etc. Use the item of as an explanatory variable. Further, the presence or absence of dementia included in the sample data D1 of each user is set as the objective variable. Then, by performing logistics regression analysis on the objective variable and the explanatory variable, the regression coefficient and the constant term related to the dementia determination model M3 are obtained.

FIG. 16 is an explanatory diagram illustrating a determination example of the dementia determination model. As shown in FIG. 16, the dementia estimation unit 12B determines the presence or absence of dementia by applying the user information 11A of the subject to the dementia determination model M3. Specifically, the dementia estimation unit 12B recognizes the items of explanatory variables such as "meal speed", "conversation speed", "repeat the same thing", and "forget the meal" in the user information 11A of the subject. It is applied to the disease determination model M3. As a result, the dementia estimation unit 12B can determine the presence or absence of dementia as the probability (for example, the probability of occurrence). The dementia estimation unit 12B determines whether or not there is dementia based on the probability of occurrence obtained for dementia (for example, by comparing with a predetermined threshold value).

Next, the dementia estimation unit 12B determines that the probability of occurrence is lower than the threshold value in the dementia determination and there is no dementia (S33: none), and the characteristics of the external factors are no dementia (S34). Further, the dementia estimation unit 12B determines that the probability of occurrence is higher than the threshold value in the dementia determination and that there is dementia (S33: yes), and that there is dementia as a characteristic of external factors (S35).

The dementia estimation unit 12B may treat the result of the dementia determination in S31 and the result of the dementia determination in S34 or S35 as separate determinations, or may combine them.

For example, the dementia estimation unit 12B may use S31 with / without dementia as the first dementia determination result and S34 or S35 with / without dementia as the second dementia determination result. .. In addition, the dementia estimation unit 12B considers that there is dementia if either the first dementia determination result or the second dementia determination result has dementia, and the first and second dementia determination results. If both of them have no dementia, they may be without dementia.

As described above, the aging model creation unit 12A of the learning device 1 acquires the posture of the target person and the characteristics of the external factors of the target person, and based on the characteristics of the external factors at different time points acquired in plurality. , Create an aging model M for the characteristics of external factors. The dementia estimation unit 12B of the learning device 1 determines whether or not the subject has an internal factor disorder (dementia) based on the acquired posture and the created aging model M. As a result, the learning device 1 can evaluate the potential risk of requiring long-term care due to internal factors such as dementia.

Further, the model learning unit 13 of the learning device 1 learns the learning model 31 by using the presence or absence of the discriminated internal factor as an explanatory variable and the acquired characteristics of the external factor as explanatory variables and the risk of the subject as the objective variable. I do.

In this way, since the learning device 1 adds the presence or absence of the disorder of the internal factor determined for the subject as an explanatory variable, the risk is taken into consideration after considering the potential factor due to the internal factor such as dementia. The learning model 31 can be trained to make an estimation, and the risk estimation accuracy can be improved.

FIG. 13 is an explanatory diagram showing a comparison example with the conventional estimation result. As shown in FIG. 13, when the subjects A and B are compared, the risk seems to be higher because the subject B, who has a lower eye and a rubbing leg, is in a deformed posture. Therefore, in the conventional estimation result, the risk of the subject B is evaluated higher than that of the subject A because the potential fall risk due to the internal factor is not evaluated.

Actually, the subject A is in an upright posture closer to normal than the subject B, but there is a significant difference between the aging model M and the reference model MK, and the subject has dementia. In the present embodiment, the learning model 31 is trained by adding potential factors due to internal factors such as dementia to the explanatory variables. Therefore, in the estimation result of the present embodiment, the subject A having dementia has a higher risk than the subject B, and the estimation result is more accurate.

Further, the dementia estimation unit 12B is based on a comparison between the reference model corresponding to the acquired posture and the created aging model M in the reference models (M1, M2 ...) According to the aging for each posture. To determine whether or not there is an internal factor failure. In this way, the learning device 1 can determine the presence or absence of an internal factor disorder by comparing the age-appropriate reference model corresponding to the posture of the subject with the age model M of the subject. For example, when there is a significant difference between the reference model and the aging model M, it can be determined that there is an internal factor disorder.

In addition, the dementia estimation unit 12B determines in the aging model M that there is a disorder of the internal factor when the characteristic of the external factor has a fluctuation with time series. For example, the dementia estimation unit 12B has an internal factor disorder (dementia) when there is a fluctuation with a predetermined range such that the walking speed repeats good times and bad times. Can be determined.

Further, the posture of the subject is the posture of the subject during walking, and the characteristic of the external factor may be the gait of the subject. In this way, the learning device 1 may learn the learning model 31 that estimates the risk that the target person has based on the posture of the target person during walking and the gait of the target person.

Also, the disorder of the internal factor may be dementia. As described above, the learning device 1 may add the presence or absence of dementia as an explanatory variable in the learning of the learning model 31 for estimating the risk possessed by the subject.

Further, the dementia discriminating unit 21 of the estimation device 2 acquires the posture of the subject and the characteristics of the external factors of the subject, and based on the characteristics of the external factors at different times obtained, the external factors are obtained. An aging model M is created for the characteristics of. The dementia discrimination unit 21 of the estimation device 2 determines whether or not the subject has an internal factor disorder (dementia) based on the acquired posture and the created aging model M. The estimation unit 22 of the estimation device 2 uses the presence or absence of an internal factor disorder and the characteristics of the external factor as explanatory variables, and the learning model 31 learned with the risk as the objective variable, and determines the presence or absence of an internal factor disorder. And the characteristics of the acquired external factors are applied to estimate the risk of the subject.

In this way, the estimation device 2 estimates the risk using the learning model 31 in consideration of the potential factors due to internal factors such as dementia, so that the risk estimation accuracy can be improved.

Further, the dementia discrimination unit 21 is based on a comparison between the reference model corresponding to the acquired posture and the created aging model M in the reference models (M1, M2 ...) According to the aging for each posture. To determine whether or not there is an internal factor failure. In this way, the estimation device 2 can determine the presence or absence of an internal factor disorder by comparing the age-appropriate reference model corresponding to the posture of the subject with the age model M of the subject. For example, when there is a significant difference between the reference model and the aging model M, it can be determined that there is an internal factor disorder.

In addition, the dementia discrimination unit 21 determines in the aging model M that there is a disorder of the internal factor when the characteristic of the external factor has a fluctuation with time series. For example, the dementia discriminating unit 21 has an internal factor disorder (dementia) when there is a fluctuation with a predetermined range such that the walking speed repeats good times and bad times. Can be determined.

Further, the posture of the subject is the posture of the subject during walking, and the characteristic of the external factor may be the gait of the subject. In this way, the estimation device 2 may estimate the risk that the subject has based on the posture of the subject when walking and the gait of the subject.

Also, the disorder of the internal factor may be dementia. In this way, the estimation device 2 may apply the presence or absence of dementia to the learning model 31 to estimate the risk that the subject has.

It should be noted that each component of each of the illustrated devices does not necessarily have to be physically configured as shown in the figure. That is, the specific form of distribution / integration of each device is not limited to the one shown in the figure, and all or part of the device is functionally or physically distributed / physically in arbitrary units according to various loads and usage conditions. It can be integrated and configured.

The various processing functions performed by the learning device 1 and the estimation device 2 may be executed in whole or in any part on the CPU (or a microcomputer such as an MPU or MCU (Micro Controller Unit)). .. In addition, various processing functions may be executed in whole or in any part on a program analyzed and executed by a CPU (or a microcomputer such as MPU or MCU) or on hardware by wired logic. Needless to say, it's good. Further, various processing functions performed by the matching device 1 may be executed by a plurality of computers in cooperation by cloud computing.

By the way, various processes described in the above-described embodiment can be realized by executing a program prepared in advance on a computer. Therefore, an example of a computer (hardware) that executes a program having the same function as that of the above embodiment will be described below. FIG. 14 is a block diagram showing an example of a computer that executes a program.

As shown in FIG. 14, the computer 3 includes a CPU 101 that executes various arithmetic processes, an input device 102 that receives data input, a monitor 103, and a speaker 104. Further, the computer 3 has a medium reading device 105 for reading a program or the like from a storage medium, an interface device 106 for connecting to various devices, and a communication device 107 for communicating with an external device by wire or wirelessly. Further, the computer 3 has a RAM 108 for temporarily storing various information and a hard disk device 109. Further, each part (101 to 109) in the computer 3 is connected to the bus 110.

The hard disk device 109 stores a program 111 for executing various processes described in the above embodiment. Further, the hard disk device 109 stores various data 112 referred to by the program 111. The input device 102 receives, for example, an input of operation information from an operator of the computer 3. The monitor 103 displays, for example, various screens operated by the operator. For example, a printing device or the like is connected to the interface device 106. The communication device 107 is connected to a communication network such as a LAN (Local Area Network), and exchanges various information with an external device via the communication network.

The CPU 101 reads the program 111 stored in the hard disk device 109, expands it into the RAM 108, and executes it to perform various processes. For example, in the case of the learning device 1, the CPU 101 executes the program 111 to perform processing related to the collecting unit 10, the dementia discrimination unit 12, and the model learning unit 13. Further, in the case of the estimation device 2, the CPU 101 executes the program 111 to perform processing related to the input unit 20, the dementia determination unit 21, the estimation unit 22, and the output unit 23.

The program 111 does not have to be stored in the hard disk device 109. For example, the computer 3 may read and execute the program 111 stored in the storage medium that can be read by the computer 3. The storage medium that can be read by the computer 3 corresponds to, for example, a CD-ROM, a DVD disk, a portable recording medium such as a USB (Universal Serial Bus) memory, a semiconductor memory such as a flash memory, a hard disk drive, or the like. Further, the program 111 may be stored in a device connected to a public line, the Internet, a LAN, or the like, and the computer 3 may read the program 111 from these and execute the program 111.

Regarding the above embodiments, the following additional notes will be further disclosed.

(Appendix 1) Acquire the posture of the subject and the characteristics of the external factors of the subject.
Based on the characteristics of the external factors obtained at different time points, a time-series model for the characteristics of the external factors was created.
Based on the acquired posture and the created model, it is determined whether or not the subject has an internal factor disorder.
A learning method characterized by a computer performing processing.

(Appendix 2) The computer further performs learning using the presence or absence of a disorder of the internal factor determined and the acquired characteristics of the external factor as explanatory variables and the risk of the subject as the objective variable. Execute,
The learning method according to Appendix 1, characterized in that.

(Appendix 3) The discriminating process is performed internally based on a comparison between the acquired reference model corresponding to the posture and the created model among the age-appropriate reference models for each posture. Determine if there is a factor failure,
The learning method according to Appendix 1 or 2, characterized in that.

(Appendix 4) The discriminating process determines that there is an obstacle of the internal factor when there is a significant difference between the reference model and the created model.
The learning method according to Appendix 3, characterized in that.

(Appendix 5) In the created model, the discriminating process determines that there is an obstacle of the internal factor when the characteristics of the external factor have fluctuations with time series.
The learning method according to any one of Supplementary notes 1 to 4, wherein the learning method is characterized by the above.

(Appendix 6) The posture is the posture of the subject during walking.
The characteristic of the external factor is the gait of the subject.
The learning method according to any one of Appendix 1 to 5, wherein the learning method is characterized by the above.

(Appendix 7) The disorder of the internal factor is dementia.
The learning method according to any one of Appendix 1 to 6, wherein the learning method is characterized by the above.

(Appendix 8) Acquire the posture of the subject and the characteristics of the external factors of the subject.
Based on the characteristics of the external factors obtained at different time points, a time-series model for the characteristics of the external factors was created.
Based on the acquired posture and the created model, it is determined whether or not the subject has an internal factor disorder.
The presence or absence of the disorder of the internal factor and the characteristics of the external factor as explanatory variables, and the presence or absence of the disorder of the internal factor determined in the learning model learned with the risk as the objective variable, and the acquired external factor Estimate the subject's risk by applying the characteristics of the factors,
An estimation method characterized by a computer performing processing.

(Appendix 9) The discriminating process is performed internally based on a comparison between the acquired reference model corresponding to the posture and the created model among the age-appropriate reference models for each posture. Determine if there is a factor failure,
The estimation method according to Appendix 8, characterized by the above.

(Appendix 10) The discriminating process determines that there is an obstacle of the internal factor when there is a significant difference between the reference model and the created model.
The estimation method according to Appendix 9, characterized by the above.

(Appendix 11) In the created model, the discriminating process determines that the internal factor is impaired when the characteristics of the external factor have fluctuations with time series.
The estimation method according to any one of Supplementary Provisions 8 to 10, characterized in that.

(Appendix 12) The posture is the posture of the subject during walking.
The characteristic of the external factor is the gait of the subject.
The estimation method according to any one of Supplementary Provisions 8 to 11, characterized in that.

(Appendix 13) The disorder of the internal factor is dementia.
The estimation method according to any one of Supplementary Provisions 8 to 12, characterized in that.

(Appendix 14) Acquire the posture of the subject and the characteristics of the external factors of the subject.
Based on the characteristics of the external factors obtained at different time points, a time-series model for the characteristics of the external factors was created.
Based on the acquired posture and the created model, it is determined whether or not the subject has an internal factor disorder.
A learning program characterized by having a computer perform processing.

(Appendix 15) Learning is performed using the determined presence or absence of a disorder of the internal factor and the acquired characteristics of the external factor as explanatory variables and the risk of the subject as the objective variable. Let it run
The learning program according to Appendix 14, characterized by the above.

(Appendix 16) The discriminating process is performed internally based on a comparison between the acquired reference model corresponding to the posture and the created model among the age-appropriate reference models for each posture. Determine if there is a factor failure,
The learning program according to Appendix 14 or 15, characterized in that.

(Appendix 17) The discriminating process determines that there is an obstacle of the internal factor when there is a significant difference between the reference model and the created model.
The learning program according to Appendix 16, characterized in that.

(Appendix 18) In the created model, the discriminating process determines that there is an obstacle of the internal factor when the characteristics of the external factor have fluctuations with time series.
The learning program according to any one of Supplementary Provisions 14 to 17, characterized in that.

(Appendix 19) The posture is the posture of the subject during walking.
The characteristic of the external factor is the gait of the subject.
The learning program according to any one of Supplementary Provisions 14 to 18, characterized in that.

(Appendix 20) The disorder of the internal factor is dementia.
The learning program according to any one of Supplementary note 14 to 19, characterized in that.

(Appendix 21) Acquire the posture of the subject and the characteristics of the external factors of the subject.
Based on the characteristics of the external factors obtained at different time points, a time-series model for the characteristics of the external factors was created.
Based on the acquired posture and the created model, it is determined whether or not the subject has an internal factor disorder.
The presence or absence of the disorder of the internal factor and the characteristics of the external factor as explanatory variables, and the presence or absence of the disorder of the internal factor determined in the learning model learned with the risk as the objective variable, and the acquired external factor Estimate the subject's risk by applying the characteristics of the factors,
An estimation program characterized by having a computer perform processing.

(Appendix 22) The discriminating process is performed internally based on a comparison between the acquired reference model corresponding to the posture and the created model among the age-appropriate reference models for each posture. Determine if there is a factor failure,
21. The estimation program according to Appendix 21.

(Appendix 23) The discriminating process determines that there is an obstacle of the internal factor when there is a significant difference between the reference model and the created model.
22. The estimation program according to Appendix 22.

(Appendix 24) In the created model, the discriminating process determines that the internal factor is impaired when the characteristics of the external factor have fluctuations with time series.
The estimation program according to any one of Supplementary Provisions 21 to 23.

(Appendix 25) The posture is the posture of the subject during walking.
The characteristic of the external factor is the gait of the subject.
The estimation program according to any one of Supplementary Provisions 21 to 24.

(Appendix 26) The disorder of the internal factor is dementia.
The estimation program according to any one of Supplementary Provisions 21 to 25.

1 ... Learning device 2 ... Estimating device 3 ... Computer 10 ... Collecting unit 11 ... User information DB
11A ... User information 12 ... Dementia discrimination unit 12A ... Aging model creation unit 12B ... Dementia estimation unit 13 ... Model learning unit 20 ... Input unit 21 ... Dementia discrimination unit 22 ... Estimating unit 23 ... Output unit 30 ... Aging Model information 31 ... Learning model 40 ... Output screen 41 ... Gait display area 42 ... Model display area 43 ... Estimated result display area 101 ... CPU
102 ... Input device 103 ... Monitor 104 ... Speaker 105 ... Media reader 106 ... Interface device 107 ... Communication device 108 ... RAM
109 ... Hard disk device 110 ... Bus 111 ... Program 112 ... Various data A, B ... Subjects C1 to C6 ... Posture D1 ... Sample data M, M1, M2 ... Aging model M3 ... Dementia judgment model MK ... Reference model

Claims (9)

Acquire the posture of the subject and the characteristics of the external factors of the subject,
Based on the characteristics of the external factors obtained at different time points, a time-series model for the characteristics of the external factors was created.
Based on the acquired posture and the created model, it is determined whether or not the subject has an internal factor disorder.
A learning method characterized by a computer performing processing. Learning is performed using the determined presence or absence of a failure of the internal factor and the acquired characteristics of the external factor as explanatory variables and the risk of the subject as the objective variable, and the computer further executes the process.
The learning method according to claim 1, wherein the learning method is characterized by the above. The discriminating process is based on a comparison between the acquired reference model corresponding to the posture and the created model among the age-appropriate reference models for each posture, and the disorder of the internal factor is determined. Determine the presence or absence,
The learning method according to claim 1 or 2, characterized in that. When there is a significant difference between the reference model and the created model, the discriminating process determines that there is an obstacle of the internal factor.
The learning method according to claim 3, wherein the learning method is characterized by the above. In the created model, the discriminating process determines that the internal factor is impaired when the characteristics of the external factor have fluctuations with time series.
The learning method according to any one of claims 1 to 4, wherein the learning method is characterized by the above. The posture is the posture of the subject during walking, and is
The characteristic of the external factor is the gait of the subject.
The learning method according to any one of claims 1 to 5, characterized in that. The disorder of the internal factor is dementia,
The learning method according to any one of claims 1 to 6, characterized in that. Acquire the posture of the subject and the characteristics of the external factors of the subject,
Based on the characteristics of the external factors obtained at different time points, a time-series model for the characteristics of the external factors was created.
Based on the acquired posture and the created model, it is determined whether or not the subject has an internal factor disorder.
The presence or absence of the disorder of the internal factor and the characteristics of the external factor as explanatory variables, and the presence or absence of the disorder of the internal factor determined in the learning model learned with the risk as the objective variable, and the acquired external factor Estimate the subject's risk by applying the characteristics of the factors,
An estimation method characterized by a computer performing processing. Acquire the posture of the subject and the characteristics of the external factors of the subject,
Based on the characteristics of the external factors obtained at different time points, a time-series model for the characteristics of the external factors was created.
Based on the acquired posture and the created model, it is determined whether or not the subject has an internal factor disorder.
A learning program characterized by having a computer perform processing.

Images