JP2017221416A - Cognitive ability change prediction device, method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To predict a change in cognitive ability of a user and perform user management more appropriately.SOLUTION: Learning data is created beforehand, and is stored in learning data accumulation means 4. In this state, heartbeat interval data on an object user is received from data input means 2 in a time-series manner, and an HRV feature amount is calculated from the heartbeat interval data for each predetermined time section. Successively, an HRV change amount based on the HRV feature amount calculated for each time section is calculated. Then, the learning data stored in the learning data accumulation means 4 is referred to, and a future change amount of the cognitive ability of the object user is predicted based on the calculated HRV feature amount and the HRV change amount. Information indicating the predicted change mount of the cognitive ability is output to result output means 3, and presented to the object user, an administrator, etc.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an apparatus, method, and program for predicting a change in a user's cognitive information processing ability (hereinafter referred to as cognitive ability) such as “Working Memory” and “Selective Attention”.

  In recent years, for example, various techniques for estimating a user's cognitive ability during work have been proposed in order to manage employee labor. For example, in Non-Patent Document 1, as in Advanced Trail Making Test (hereinafter referred to as ATMT), the work being executed is interrupted and a task of about several minutes is executed, and the user's recognition from the result of the task A method for estimating capacity is described. In Non-Patent Document 2, an electrocardiographic sensor is attached to a user, an electrocardiographic R wave peak interval (hereinafter referred to as RRI) is measured, and a plurality of feature amounts calculated from the RRI are used at the present time. A method for estimating the cognitive ability of a child is described.

K. Mizuno et al .: Utility of an Advanced Trail Making Test as a Neuropsychological Tool for an Objective Evaluation of Work Efficiency During Mental Fatigue, Fatigue Science for Human Health, pp. 47-54, 2008. Tsunoda et al., “Sequential Estimation Method of Cognitive Ability Using Heart Rate Variability”, IPSJ Groupware and Network Service Workshop Proceedings, pp. 1-10, 2015.

  However, the techniques described in Non-Patent Documents 1 and 2 both estimate the user's cognitive ability at the present time. For this reason, when trying to use the estimation results for labor management, it is possible to take a rest when the cognitive ability declines, but it is impossible to take a rest before the cognitive ability declines. It was. Therefore, productivity reduction cannot be prevented and improvement in labor management has been demanded.

  The present invention has been made paying attention to the above circumstances, and the object of the present invention is to recognize a change in cognitive ability so that a change in cognitive ability can be predicted so that the user can be managed more appropriately. It is intended to provide a program.

  In order to solve the above-described problem, according to a first aspect of the present invention, data representing a heartbeat interval of a prediction target user is acquired, and a feature amount of heartbeat variability is calculated based on the acquired data indicating the heartbeat interval. At the same time, the change amount is calculated based on the calculated feature amount of the heartbeat variability. Then, based on the calculated feature value of heart rate variability and the amount of change thereof, and learning data stored in advance, a change in the cognitive ability of the prediction target user is predicted, and information representing the prediction result is obtained. It is designed to output.

  According to a second aspect of the present invention, as means for calculating the feature quantity of the heartbeat variability, means for calculating a feature quantity in a time domain based on the data representing the acquired heartbeat interval, and the acquired heartbeat Means for calculating a feature quantity in the frequency domain based on data representing the interval. Then, by means of calculating the time domain feature amount, the average value of the heartbeat interval, the standard deviation of the heartbeat interval (SDNN), and the ratio (pNN50) in which the difference between adjacent heartbeat intervals exceeds 50 msec (pNN50) Calculates at least one of the root mean square difference (RMSSD) of the difference in heart rate interval, and the low frequency component (0.04 to 0.15) when the heart rate interval is spectrally analyzed by means of calculating the feature quantity in the frequency domain. Hz) total power (LF), the total power (HF) of the high frequency component (0.15 to 0.40 Hz) when the heartbeat interval is spectrally analyzed, the total power LF of the low frequency component and the total power HF of the high frequency component And at least one of the ratio (LF / HF).

  According to a third aspect of the present invention, the feature amount of the heart rate variability is calculated by the means for calculating the feature amount of the heart rate variability in each of a plurality of different time intervals based on the acquired data representing the heart rate interval. The means for calculating the amount of change in the feature value of heartbeat variability between the change amount or change rate between the feature values of heartbeat variability calculated in the plurality of different time intervals and the feature value of each heartbeat variability Either the change amount or the change rate of the correlation coefficient is calculated.

According to a fourth aspect of the present invention, when calculating the amount of change in the feature value of the heartbeat variability, the principal component analysis is performed on the feature value of each heartbeat variability calculated in the plurality of different time intervals. Based on the analysis result of the main component, the T ² statistic indicating the change of the main component and the Q statistic indicating the residual of dimensional compression are respectively calculated.

  In the fifth aspect of the present invention, when the change in the cognitive ability is predicted, the information measured as the result of the Advanced Trail Making Test is predicted as the change in the cognitive ability.

  According to a sixth aspect of the present invention, as a means for predicting the change in the cognitive ability, a classifier representing the change in the cognitive ability is created based on the learning data, and the calculated feature value of the heart rate variability and its By inputting the amount of change into the classifier, a regression model is created based on the means for predicting the change in the cognitive ability and the learning data, and the calculated feature value of the heart rate variability and the amount of change are calculated. By inputting into the regression model, any one of the means for predicting the amount of change in the cognitive ability is provided.

  According to the first aspect of the present invention, a change in the cognitive ability of the prediction target user is predicted and the result is displayed. For this reason, for example, the prediction target user can take a rest before the cognitive ability declines, or the administrator can take measures such as giving the rest to the prediction target user at an appropriate timing. As a result, it is possible to maintain high work efficiency and quality of the user who is executing the task.

  According to the second aspect of the present invention, the time domain feature quantity and the frequency domain feature quantity are calculated based on the heartbeat interval data as the feature quantity of the heart rate variability. For this reason, it becomes possible to obtain the feature-value which has relevance with a user's cognitive ability reliably.

  According to the third aspect of the present invention, as the amount of change in the feature value of heart rate variability, the change amount or change rate between the heart rate variability feature amounts calculated in a plurality of different time intervals, and between each heart rate variability feature amount One of the change amount or the change rate of the correlation coefficient is calculated. For this reason, it is possible to relatively easily calculate the amount of change in the feature amount of heartbeat variability in units of time intervals.

According to the fourth aspect of the present invention, as the amount of change in the feature value of heart rate variability, the T ² statistic indicating the change in the main component of the feature value of heart rate variability and the Q statistic indicating the residual of dimension compression are Calculated. For this reason, the amount of change in the feature amount of heartbeat variability can be calculated more accurately.

  According to the fifth aspect of the present invention, a change in cognitive ability is predicted using information measured as the result of Advanced Trail Making Test. For this reason, it becomes possible to represent the change of a user's cognitive ability in the form which a user can grasp easily.

  According to the sixth aspect of the present invention, the change in the user's cognitive ability can be predicted as the presence or absence of the change in the cognitive ability, or can be predicted as the amount of change in the cognitive ability.

  That is, according to each aspect of the present invention, it is possible to provide a cognitive ability change prediction apparatus, method, and program that can predict a change in cognitive ability so that the user can be managed more appropriately.

The block diagram which shows the function structure of the cognitive ability change prediction apparatus used for the principle explanation of this invention. The flowchart which shows the process sequence and process content of the cognitive ability change prediction apparatus shown in FIG. The block diagram which shows the structure of the cognitive ability change prediction apparatus which concerns on 1st Embodiment of this invention. The flowchart which shows the process sequence and process content of the learning data creation process by the cognitive ability change prediction apparatus shown in FIG. The flowchart which shows the process sequence and processing content of the cognitive ability change prediction process by the cognitive ability change prediction apparatus shown in FIG. The flowchart which shows the process sequence and processing content of the learning data creation process by the cognitive ability change prediction apparatus which concerns on 2nd Embodiment of this invention. The flowchart which shows the process sequence and processing content of the cognitive ability change prediction process by the cognitive ability change prediction apparatus which concerns on 2nd Embodiment of this invention. The block diagram which shows the structure of the cognitive ability change prediction apparatus which concerns on 3rd Embodiment of this invention.

[principle]
FIG. 1 is a diagram for use in explaining the principle of the present invention, wherein 1 is a prediction processing device, 2 is a data input means, 3 is a result output means, and 4 is a learning data storage means.
The data input means 2 inputs the target user's heartbeat interval data to the prediction processing device 1, and is composed of, for example, a wearable heartbeat sensor, a pulse wave sensor, or a database in which heartbeat interval data is accumulated. The result output means 3 outputs the prediction result of the cognitive ability obtained by the prediction processing device 1, and is composed of, for example, an application program used in a smartphone.

  The learning data storage means 4 has the same time for the feature quantity indicating the heartbeat variability during the work of the plurality of users to be predicted, which is measured in advance, and the information regarding the information processing ability of the user during the work. Learning data associated with each other is stored. More specifically, a value indicating a change in the user's cognitive ability (hereinafter referred to as a cognitive ability change value), an HRV feature amount, and information indicating the change amount of the HRV feature amount are stored as learning data.

As a cognitive ability change value, the grade of ATMT (Advanced Trail Making Test) and its change amount are mentioned, for example. HRV features include, for example, the average value of RRI, the standard deviation of RRI, or the power spectrum of each of the high frequency band (0.15 Hz to 0.40 Hz) and low frequency band (0.04 Hz to 0.15 Hz) The ratio is LF / HF. As the amount of change in HRV feature value, a T ² statistic indicating the change in the principal component score calculated after applying the principal component analysis to a plurality of HRV feature values, and Q indicating the change in correlation between the feature values Statistics are listed.

The prediction processing apparatus 1 includes a feature amount calculating unit 11, a feature amount change calculating unit 12, and a cognitive ability change predicting unit 13 as processing functions for carrying out the present invention.
The feature amount calculating unit 11 calculates an HRV feature amount for each predetermined time interval from the heartbeat interval data input from the data input unit 2. The feature amount change calculation unit 12 calculates a change amount of the HRV feature amount (hereinafter referred to as HRV change amount) based on the HRV feature amount calculated for each time interval by the feature amount calculation unit 11. The cognitive ability change prediction means 13 is stored in the learning data storage means 4 based on the HRV feature quantity calculated by the feature quantity calculation means 11 and the HRV change quantity calculated by the feature quantity change calculation means 12. Predict changes in cognitive ability with reference to learning data.

Since it is the above structures, the prediction processing apparatus 1 operate | moves as follows. FIG. 2 is a flowchart showing the processing procedure and processing contents.
First, in step S11, the prediction processing apparatus 1 receives the target user's heartbeat interval data from the data input means 2 in time series, and calculates an HRV feature amount for each predetermined time interval from the heartbeat interval data. Subsequently, in step S12, an HRV change amount is calculated based on the HRV feature amount calculated for each time interval in step S11. Next, in step S13, based on the HRV feature amount calculated in step S11 and the HRV change amount calculated in step S12, the learning data stored in the learning data storage unit 4 is referred to and the target user is recognized. Predict the amount of capacity change. Then, the information indicating the predicted change amount of the cognitive ability is output to the result output means 3 and presented to the target user or the administrator thereof.

  Therefore, according to the present invention, since the prediction result of the change amount of the cognitive ability of the target user is presented, more appropriate labor management such as giving a rest to the user before the cognitive ability of the user is reduced is performed. It becomes possible.

Embodiments according to the present invention will be described below with reference to the drawings.
[First Embodiment]
(Constitution)
FIG. 3 is a block diagram showing a configuration of a system including the cognitive ability change prediction device according to the first embodiment of the present invention.
This system includes the cognitive ability change prediction device 10A as a core, and further includes pulse wave sensors 21, 22, an ATMT input unit 23, a display unit 30, and a learning database (learning DB) 40.

  The pulse wave sensor 21 is attached to a learning data measurement target user. Then, the pulse wave of the learning data measurement target user is measured, data representing the heartbeat interval is calculated based on the measurement data of the pulse wave, and the calculated data is used as the cognitive ability change prediction device 10A via the wireless interface, for example. It has a function to send to.

  The pulse wave sensor 22 is worn by a user who is predicted to change cognitive ability. Then, the pulse wave of the user who is to be predicted for cognitive ability change is measured, data representing the heartbeat interval is calculated based on the measured data of the pulse wave, and the calculated data is predicted, for example, through the wireless interface. A function of transmitting to the device 10A is provided.

  As a wireless interface used by each of the pulse wave sensors 21 and 22, a wireless interface adopting a short-range wireless data communication standard such as Bluetooth (registered trademark) is used. The pulse wave sensors 21 and 22 may not be prepared individually, and one pulse wave sensor may be shared for learning data creation and for cognitive ability change prediction.

  The ATMT input unit 23 includes a mobile terminal (for example, a notebook personal computer, a tablet terminal, or a smartphone) that includes an interface for inputting data representing the execution result of the ATMT task. And the data showing the implementation result of the said input ATMT task are transmitted toward the cognitive ability change prediction apparatus 10A via a wireless interface, for example.

  The display unit 30 includes a mobile terminal (for example, a wearable terminal, a smartphone, or a tablet terminal) that includes an application program for displaying the recognition ability change prediction result. And the information showing the prediction result of the cognitive ability change sent from 10 A of cognitive ability change prediction apparatuses via the wireless interface is received, and the said information is displayed on a display.

  The learning DB 40 is provided, for example, in a server computer or cloud computer on the Web, and receives and stores learning data created by the cognitive ability change prediction device 10A via a network. Moreover, according to the reference request | requirement sent from 10 A of cognitive ability change prediction apparatuses, applicable learning data are transmitted to 10 A of cognitive ability change prediction apparatuses of request origin. Note that the learning DB 40 may be provided in the cognitive ability change prediction device 10A.

  The cognitive ability change prediction device 10A includes a personal computer or a server computer including a processor such as a CPU (Central Processing Unit), a storage unit, and various external interfaces such as a wireless interface. And as a processing function required in order to implement this embodiment, the feature-value calculation parts 101 and 104, the feature-value change calculation parts 102 and 105, the learning data creation part 103, and the cognitive ability change prediction part 106 are included. Have. All of these processing functions are realized by causing the processor to execute a program stored in the storage unit.

  Among the above processing functions, the feature amount calculation unit 101, the feature amount change calculation unit 102, and the learning data creation unit 103 are used to create learning data.

  The feature amount calculation unit 101 receives heartbeat interval data of the learning data measurement target user transmitted in time series from the pulse wave sensor 21, and based on the received heartbeat interval data for each predetermined time interval. HRV feature value is calculated.

Based on the calculated HRV feature value, the feature value change calculation unit 102 calculates the T ² statistic and the Q statistic of the temporally adjacent sections as data representing the change amount of the HRV feature value. Then, the calculated T ² statistic and Q statistic are transferred to the learning data creation unit 103 together with the HRV feature amount calculated by the feature amount calculation unit 101. Note that the feature value change calculation unit 102 calculates the change amount or change rate between the feature values of each heart rate variability calculated in the plurality of different time intervals. The change amount or change rate of the number of relationships may be calculated.

The learning data creation unit 103 uses the HRV feature value, the T ² statistic value, and the Q statistic value passed from the feature value change calculation unit 102 as explanatory variables, and the ATMT error input by the user to be learned data measurement in the ATMT input unit 23 Learning data is created using the data representing the rate increase / decrease as the explained variable, and the created learning data is transmitted to the learning DB 40 for storage.

  On the other hand, among the processing functions described above, the feature amount calculation unit 104, the feature amount change calculation unit 105, and the cognitive ability change prediction unit 106 are used to predict a change in cognitive ability of the prediction target user.

  The feature amount calculation unit 104 receives the heartbeat interval data of the prediction target user transmitted in time series from the pulse wave sensor 22, and based on the received heartbeat interval data, the HRV feature amount for each predetermined time interval Is calculated.

Based on the calculated HRV feature value, the feature value change calculation unit 105 calculates the T ² statistic value and the Q statistic value of the temporally adjacent sections as the HRV feature value change amount. Then, the calculated T ² statistic and Q statistic are output to the cognitive ability change prediction unit 106 together with the HRV feature amount calculated by the feature amount calculation unit 104. Note that the feature value change calculation unit 105 calculates the change amount or change rate between the feature values of each heart rate variability calculated in the plurality of different time intervals. The change amount or change rate of the number of relationships may be calculated.

The cognitive ability change prediction unit 106 uses the T ² statistic and Q statistic calculated as the HRV change amount by the feature amount change calculation unit 105 and the learning data stored in the learning DB 40 to determine the target user's change. Predict changes in cognitive ability. Then, display data of the prediction result is generated and transmitted to the display unit 30.

(Operation)
Next, operation | movement of 10 A of cognitive ability change prediction apparatuses comprised as mentioned above is demonstrated. Here, the case where the increase / decrease in the predicted change in cognitive ability is predicted with the binary values of “whether increased or decreased” will be described.

(1) Creation of learning data Prior to the prediction process of the change in cognitive ability of the prediction target user, the cognitive ability change prediction device 10A executes the process of creating learning data as follows. FIG. 4 is a flowchart showing the processing procedure and processing contents.

  A pulse wave sensor 21 is attached to a user to be measured for learning data. The pulse wave sensor 21 measures the pulse wave of the measurement target user, and calculates heartbeat interval data based on the measured pulse wave waveform data. Then, the calculated heartbeat interval data is transmitted to the cognitive ability change prediction device 10A via the wireless interface.

  In parallel with the measurement operation of the heartbeat interval data, the user to be measured of the learning data performs the ATMT task at an interval of 10 minutes, and inputs the ATMT error rate and execution time as data representing his / her cognitive ability. This is input in the unit 23. Every time the ATMT error rate and the execution time are input, the ATMT input unit 23 transmits them to the cognitive ability change prediction device 10A via the wireless interface.

  The cognitive ability change prediction device 10A first receives and takes in heartbeat interval data transmitted in time series from the pulse wave sensor 21 in step S21 under the control of the feature amount calculation unit 101, and sequentially stores it in a storage unit (not shown). save. In step S22, the ATMT error rate and execution time transmitted from the ATMT input unit 23 are received. The received ATMT error rate and execution time are received in the learning data creation unit 103, and the execution time is converted into the feature amount calculation unit 101. To enter.

  Subsequently, in step S23, the feature amount calculation unit 101 synchronizes the heart rate interval data stored in the storage unit with a 10-minute interval (section t (t = 1, 2,...) In synchronization with the execution time of the ATMT task. , N), and for each interval t 1, HRV feature amounts are calculated while sliding a time window of 5 minutes with a slide width of 6 seconds, and the calculated HRV feature amounts are converted into a feature amount change calculation unit 102. To pass.

The calculated HRV feature value includes a time domain feature value and a frequency domain feature value.
Examples of time domain features include:
(1) Average heart rate interval
(2) Heartbeat interval standard deviation (SDNN)
(3) Ratio of difference between adjacent heartbeat intervals exceeding 50msec (pNN50)
(4) Root mean square (RMSSD) of the difference between adjacent heartbeat intervals
There is.

On the other hand, the frequency domain feature amount includes, for example,
(5) Total power (LF) of low frequency components (0.04 to 0.15 Hz) when the heartbeat interval is spectrally analyzed
(6) Total power (HF) of high frequency components (0.15 to 0.40Hz) when spectrum analysis of heartbeat interval
(7) Ratio of total power LF of low frequency components to total power HF of high frequency components (LF / HF)
There is.

  The feature quantity calculation unit 101 calculates, for example, SDNN of the above-described time domain feature quantities and LF / HF of the frequency domain feature quantities as the HRV feature quantities. Other feature amounts may be calculated for each feature amount, or a plurality of feature amounts may be calculated simultaneously.

Next, the cognitive ability change prediction device 10A, under the control of the feature amount change calculation unit 102, in step S24, based on the HRV feature amount passed from the feature amount calculation unit 101, the interval t (t> 1). HRV change amount is calculated for each. The amount of change in HRV consists of T ² statistics and Q statistics. Then, the calculated HRV change amount is transferred to the learning data creation unit 103 together with the HRV feature amount calculated by the feature amount calculation unit 101.

The calculation method of the T ² statistic and the Q statistic is as follows.
Step1: Apply principal component analysis as the principal component number R to the vector XεR ^{A × B} (A = number of windows, B = number of feature amounts) of the HRV feature amount in the section t−1 to obtain the principal component score vector SεR ^{A × B} and the principal component load matrix V _R are calculated.

Step 2; The principal component score vector S′∈R ^{A × B} is calculated using the load matrix V _R calculated in Step 1 for the vector X′∈R ^{A × B} of the HRV feature quantity in the section t.

Step 3; From each principal component score S _r in the principal component score vector S ′ and each principal component score S ′ _r in the principal component score vector S ′, T ² statistics in the interval t are
Calculated by Σ _Sr is the standard deviation of the principal component score S _r .

Step4: Reconstructed feature vector X ′ ′ obtained by reconstructing HRV feature vector X ′ from the principal component from principal component score vector S ′ and principal component load matrix V _R ,
X ｀ ′ = S′V ^T _R
Calculated by

Step5: The Q statistic is a value indicating a residual of dimension compression, that is, the sum of square errors between the reconstructed feature vector X ′ ′ and the feature vector X ′.
Q = Σ (X ｀ ′ − X ′) ²
Calculated by

Finally, the cognitive ability change prediction device 10A, under the control of the learning data creation unit 103, in step S25, for each interval t (1 <t <n), the HRV feature amount in the intervals t 1 and t−1, and the interval t HRV change amount (T ² statistic and Q statistic) calculated in section t based on -1 is used as an explanatory variable, and increase / decrease in ATMT error rate in section t + 1 and section t is used as an explained variable Learning data is created, and the learning data is stored in the learning DB 40.

  It should be noted that the learning data creation process described above is performed for each of these users when there are a plurality of users whose prediction ability is to be predicted.

  In the above description, the case where learning data is created for a specific target user has been described as an example. However, learning data is similarly created and stored in the learning DB 40 for a plurality of users that are assumed as prediction targets of cognitive ability change.

(2) Prediction process of user's cognitive ability change When the creation of the learning data is completed, the cognitive ability change prediction device 10A executes a process for predicting a change in cognitive ability for the prediction target user as follows. FIG. 5 is a flowchart showing the processing procedure and processing contents.

  The target user is equipped with a pulse wave sensor 22. The pulse wave sensor 22 measures the pulse wave of the target user, and calculates heartbeat interval data based on the measured pulse wave waveform data. Then, the calculated heartbeat interval data is transmitted to the cognitive ability change prediction device 10A via the wireless interface.

  The cognitive ability change prediction device 10A first receives and takes in heartbeat interval data transmitted in time series from the pulse wave sensor 22 in step S31 under the control of the feature amount calculation unit 104, and temporarily stores it in a storage unit (not shown). save. First, for the heartbeat interval data received from the pulse wave sensor 22 in the latest 10 minutes from the present time, HRV feature amounts are calculated while sliding a time window of 5 minutes with a slide width of 6 seconds. The HRV feature value calculated at this time is the same as the values (1) to (7) calculated by the feature value calculation unit 101 at the time of learning data creation described above. Then, the calculated latest 10 minute HRV feature amount is passed to the feature amount change calculation unit 105. Thereafter, the HRV feature amount calculation process is repeated every 10 minutes.

Next, under the control of the feature amount change calculation unit 105, the cognitive ability change prediction device 10A calculates the HRV change amount as follows in step S32. That is, the HRV feature values transferred from the feature amount calculation unit 104 are sequentially stored in the storage unit. First, the HRV feature amount in the section from 10 minutes before the current time is read from the storage unit, and the HRV change amount is calculated based on the HRV feature amount measured and calculated in the section from 10 minutes to 20 minutes ago. To do. Specifically, from the current time point based on the load matrix V _R and the principal component score vector S ′ when the principal component analysis is applied to the HRV feature values measured and calculated in the interval from 10 minutes to 20 minutes ago. T ² statistics and Q statistics are calculated using HRV features up to 10 minutes ago.

  In addition to calculating the change amount or rate of change between the feature values of each heart rate variability, the change amount or change rate of the correlation coefficient between the feature values of each heart rate variability may be calculated.

The cognitive ability change prediction device 10A finally reads data representing the increase / decrease in the ATMT error rate from the learning DB 40 in step S33 under the control of the cognitive ability change prediction unit 106, and uses SVM (Support Vector Machine) for prediction. Create a regression model classifier. Then, by inputting the HRV feature amount calculated by the feature amount change calculation unit 105 and the change amount (T ² statistic and Q statistic) to the classifier, the ATMT error rate is increased or decreased, that is, “increases”. Predict "or" decrease ". Then, display data indicating the prediction result is generated and transmitted to the display unit 30. As a result, a message such as “cognitive ability is predicted to be reduced in the next 10 minutes compared to the previous 10 minutes” is displayed on the display unit 30.

(effect)
As described above in detail, in the first embodiment, in the learning data creation unit 103 in advance, for each section t (1 <t <n), the section t 1, t−1 calculated by the feature amount calculation unit 101. And the HRV change amount (T ² statistic and Q statistic) in the section t calculated based on the section t−1 calculated by the feature quantity change calculation unit 102 in FIG. Learning data having binary data representing the increase / decrease of the ATMT error rate between the interval t + 1 and the interval t input by the ATMT input unit 23 is created and stored in the learning DB 40. In this state, the feature amount calculation unit 104 calculates the HRV feature amount for each predetermined time based on the heartbeat interval data of the prediction target user, and the feature amount change calculation unit 105 based on the HRV feature amount. A T ² statistic and a Q statistic indicating the HRV change amount are calculated. Then, the cognitive ability change prediction unit 106 reads the data representing the increase / decrease in the ATMT error rate from the learning DB 40 to create a classifier, and calculates the calculated HRV feature amount and its change amount (T ² statistic and Q statistic). (Quantity) is input to the classifier, and an increase or decrease in the ATMT error rate is predicted, and the result is displayed on the display unit 30.

  Therefore, by predicting the change in the cognitive ability of the prediction target user and displaying the result, for example, the prediction target user can take a rest before the cognitive ability declines, or the administrator can It is possible to take measures such as giving a rest at an appropriate time. As a result, it is possible to maintain high work efficiency and quality of a user who is executing a task using, for example, a personal computer.

[Second Embodiment]
In the second embodiment of the present invention, an increase / decrease in cognitive ability change to be predicted is predicted as an increase / decrease amount that is not a binary value but a continuous value.

(1) Creation of Learning Data FIG. 6 is a flowchart showing the procedure and content of learning data creation processing by the cognitive ability change prediction device 10A. In the figure, steps having the same processing contents as those in FIG.

The second embodiment is different from the first embodiment in that the change amount of the ATMT error rate is stored in the learning DB 40 as an explained variable. That is, the learning data creation unit 103 of the cognitive ability change prediction device 10A creates HRV feature values in the sections t 1 and t−1 for each section t (1 <t <n) in step S26 when creating learning data. And HRV change (T ² statistic and Q statistic) calculated in interval t 1 based on interval t-1 as explanatory variables, and change in ATMT error rate in interval t + 1 and interval t, That is, learning data is created using the ATMT error rate itself as the explained variable. Then, the created learning data is stored in the learning DB 40.

(2) Prediction process of user's cognitive ability change When the creation of the learning data is completed, the cognitive ability change prediction device 10A executes a process for predicting a change in cognitive ability for the prediction target user as follows. FIG. 7 is a flowchart showing the processing procedure and processing contents. In the figure, steps having the same processing contents as those in FIG.

In step S34, the cognitive ability change prediction unit 106 of the cognitive ability change prediction device 10A first reads data representing the amount of change in the ATMT error rate from the learning DB 40, and creates an estimation regression model using an SVR (Support Vector Regression) model. . Then, the change amount of the ATMT error rate is predicted by inputting the HRV feature amount calculated by the feature amount change calculation unit 105 and the change amount (T ² statistic and Q statistic) into the estimation regression model. Then, display data indicating the prediction result is generated and transmitted to the display unit 30. The regression model used for estimation is not limited to SVR, and other multiple regression models such as a PLS (Partial Least Square) regression model may be used.

  By executing the processing as described above, it is possible to predict the increase or decrease in addition to the increase or decrease in the ATMT error rate. And the display data which shows the said prediction result are produced | generated and transmitted to the display part 30, For example, "For the next 10 minutes, it is estimated that cognitive ability will reduce 20% compared with the last 10 minutes." A message can be displayed. Therefore, it is possible to instruct the prediction target user not only whether to rest, but how much time should be rested, and further, an appropriate rest type (nap, gymnastics, snacks, etc.). Become.

[Third Embodiment]
In the third embodiment of the present invention, after the target user's cognitive ability change prediction process, the target user's cognitive ability change is estimated based on the HRV feature values actually measured and calculated in the prediction target period. Based on the estimation result of the cognitive ability change, new learning data is created and added to the learning DB.

  FIG. 8 is a block diagram showing a configuration of a cognitive ability change prediction apparatus according to the third embodiment of the present invention. In the figure, the same parts as those in FIG.

  The cognitive ability change prediction device 10 </ b> B of the present embodiment further includes a cognitive ability change estimation unit 107. The cognitive ability change estimation unit 107 reads the HRV feature quantity actually calculated by the feature quantity calculation unit 104 during the prediction target period after the target user's cognitive ability change prediction process. Then, the cognitive ability of the target user in the prediction target period is estimated based on the calculated HRV feature amount, and the estimation result is passed to the learning data generating unit 103 to generate new learning data.

  With this configuration, in the cognitive ability change prediction device 10B, when the cognitive ability change prediction unit 106 finishes the prediction process of the target user's cognitive ability change in the next 10 minutes, for example, the cognitive ability change estimation unit 107 continues. The actual cognitive ability in the next 10 minutes is estimated.

  This cognitive ability estimation process is performed as follows. That is, first, the feature amount calculation unit calculates the HRV feature amounts of the latest 10-minute heartbeat interval data received from the pulse wave sensor 22 while sliding the 5-minute time window with the slide width of 6 seconds. Next, the cognitive ability change estimation unit 107 estimates the actual cognitive ability change of the target user for the 10 minutes based on the calculated HRV feature amount. Specifically, the change amount of the ATMT error rate based on the estimation result of the cognitive ability change is estimated. The method for estimating the cognitive ability change using the HRV feature amount is described in detail in Non-Patent Document 2.

The cognitive ability change estimation unit 107 calculates the estimated result of the change amount of the ATMT error rate calculated above by the feature amount calculation unit 104 and the feature amount change calculation unit 105 in the process of predicting the cognitive ability change, respectively. Along with the amount and HRV change amount (T ² statistic and Q statistic), it is passed to the learning data creation unit 103. The learning data creation unit 103 uses the HRV feature value and the HRV change amount (T ² statistic and Q statistic) passed from the cognitive ability change estimation unit 107 as explanatory variables, and estimates the change amount of the ATMT error rate. New learning data having the result as an explained variable is created, and the created learning data is written in the learning DB 40.

  As described above, in the third embodiment, after the target user's cognitive ability change prediction process, the target user's cognitive ability change is calculated based on the HRV feature values actually measured and calculated in the prediction target period. Based on the estimation result of the cognitive ability change, new learning data is created and added to the learning DB. Therefore, it is possible to store learning data based on the estimation result of the cognitive ability change in the latest period of the target user, and thereby it is possible to predict the cognitive ability change thereafter with higher accuracy.

[Other Embodiments]
As an indicator of cognitive ability, in addition to the error rate of ATMT, the number of correct answers and the number of errors in ATMT may be used. It may be used.

In each of the above embodiments, the case where the cognitive ability change prediction apparatus is provided in a server computer or the like has been described as an example. However, the recognition ability change prediction apparatus may be provided in a mobile terminal such as a smartphone or a tablet terminal possessed by the target user.
In addition, the configuration of the cognitive ability change prediction device, the procedure of the prediction process, the processing content, and the like can be implemented with various modifications without departing from the gist of the present invention.

  In short, the present invention is not limited to the above-described embodiments as they are, and can be embodied by modifying the components without departing from the scope of the invention in the implementation stage. Moreover, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the above embodiments. For example, some components may be deleted from all the components shown in each embodiment. Furthermore, you may combine suitably the component covering different embodiment.

  DESCRIPTION OF SYMBOLS 1 ... Processing apparatus, 2 ... Data input means, 3 ... Result output means, 4 ... Learning data storage means, 11 ... Feature-value calculation means, 12 ... Feature-value change calculation means, 13 ... Cognitive ability change prediction means, 21, 22 ... Pulse wave sensor, 23 ... ATMT input unit, 30 ... Display unit, 40 ... Learning database, 101, 104 ... Feature quantity calculation unit, 102, 105 ... Feature quantity change calculation unit, 103 ... Learning data creation unit, 106: Cognitive ability change prediction unit, 107 ... Cognitive ability change estimation unit.

Claims (8)

Means for obtaining data representing a heartbeat interval of the prediction target user;
Means for calculating a characteristic amount of heartbeat variability based on the acquired data representing the heartbeat interval;
Means for calculating a change amount of the feature value of the heartbeat variability based on the calculated feature value of the heartbeat variability;
Means for predicting a change in the cognitive ability of the prediction target user based on the calculated feature amount of the heart rate variability and the amount of change thereof, and learning data stored in advance;
And a means for outputting information representing a prediction result of a change in the cognitive ability of the prediction target user. The means for calculating the feature value of the heart rate variability is:
Means for calculating a time-domain feature quantity based on the data representing the acquired heartbeat interval;
Means for calculating a characteristic quantity in the frequency domain based on the acquired data representing the heartbeat interval,
Means for calculating the time domain feature amount include an average value of heartbeat intervals, a standard deviation of heartbeat intervals (SDNN), a ratio in which a difference between adjacent heartbeat intervals exceeds 50 msec (pNN50), and an adjacent heartbeat interval. And at least one of the root mean square (RMSSD) of the difference between
The means for calculating the feature quantity of the frequency domain includes the total power (LF) of the low frequency component (0.04 to 0.15 Hz) when the heartbeat interval is spectrum-analyzed and the high frequency component (0.15 to At least one of a total power (HF) of 0.40 Hz) and a ratio (LF / HF) of the total power LF of the low-frequency component and the total power HF of the high-frequency component is calculated. The cognitive ability change prediction device according to claim 1. The means for calculating the feature value of the heart rate variability calculates the feature value of the heart rate variability in each of a plurality of different time intervals based on the data representing the acquired heart beat interval,
The means for calculating the change amount of the feature value of the heart rate variability includes a change amount or change rate between the feature values of the heart rate variability calculated in the different time intervals, and a phase between the feature values of the heart rate variability. The cognitive ability change prediction device according to claim 1, wherein one of a change amount and a change rate of the number of relationships is calculated. The means for calculating the change amount of the feature value of the heart rate variability performs a principal component analysis on the feature value of each heart rate variability calculated in the plurality of different time intervals, and based on the analysis result of the principal component, The cognitive ability change prediction apparatus according to claim 3, wherein a T ² statistic indicating the change of the principal component and a Q statistic indicating a dimensional compression residual are respectively calculated.   The cognitive ability change prediction apparatus according to claim 1, wherein the means for predicting a change in cognitive ability predicts information measured as a result of Advanced Trail Making Test as the change in cognitive ability. The means for predicting the change in cognitive ability is:
A classifier representing a change in cognitive ability is created based on the learning data, and the change in the cognitive ability is predicted by inputting the calculated feature value of the heart rate variability and the change amount into the classifier. Means,
A regression model is created based on the learning data, and the calculated amount of heartbeat variability and its change amount are input to the regression model, thereby predicting the change amount of the cognitive ability. The cognitive ability change prediction device according to claim 1, comprising either one of them. A cognitive ability change prediction method executed by a cognitive ability change prediction apparatus including a computer,
Accepting input of data representing the heart rate interval of the prediction target user;
A process of calculating a feature value of heart rate variability based on the data representing the accepted heartbeat interval;
A process of calculating a change amount of the feature value of the heartbeat variability based on the calculated feature value of the heartbeat variability;
A process of predicting a change in the cognitive ability of the prediction target user based on the calculated feature value of the heart rate variability and its change amount and learning data stored in advance.
And a process of outputting information representing a prediction result of a change in the cognitive ability of the prediction target user.   The program which makes a computer function as said each means of the cognitive ability change prediction apparatus in any one of Claims 1 thru | or 6.