JP6576104B2 - Activity analysis method - Google Patents

Description

  The present invention relates to an infant activity analysis method and activity analysis system.

  In addition to actual age such as age and age, weight is often used as a measure of the degree of infant growth. This is because the body weight has a certain degree of correlation with the actual age as a scale indicating the physical growth of the infant. However, as the degree of growth of infants, there are various viewpoints such as the degree of development of intelligence, spirit, and motor ability as well as the degree of physical growth, and these do not necessarily correlate with body weight. Infants move in various ways, such as turning over, sitting and playing, grabbing and walking around, walking up and down, but even if the infant has a large body and heavy weight, Phenomena such as being unable to get caught easily are seen. That is, even if an infant is heavier than his real age and grows physically faster, his / her exercise ability is not always developed to the same extent.

  On the other hand, attempts have been made to analyze infants' movements for purposes other than the evaluation of the degree of development of motor skills. For example, Patent Document 1 describes a method in which a moving image of an infant such as a newborn is photographed with a camera and the movement of the infant's extremities is analyzed from the moving image. According to this method, it is said that it is possible to quantitatively grasp an infant's motor ability and discover a disease or a sign thereof without placing a burden on the infant to wear a measuring instrument.

  Since it is difficult for infants to communicate sufficiently through language, an observer always observes them or takes a moving image with a camera as in the method of Patent Document 1 above. In the past, a method of tracking the movements of infants was adopted.

JP 2005-131150 A

  However, since the method of Patent Document 1 is performed by taking a moving image of an infant with a camera fixed at a specific position such as a newborn bed or a ceiling of a newborn room, the subject moves by itself. It is limited to newborn infants and infants with almost no age. In addition, tracking the movement of an infant by an observer always accompanying the infant or analyzing a long-term moving image is particularly necessary for infants who have grown up to the age of being able to move by themselves. There is a problem that the load is large.

  The present invention has been made in view of the circumstances as described above, and relates to a method capable of easily analyzing an activity even for an infant who can move by himself / herself.

  The present invention relates to an infant activity analysis method, and uses an acceleration sensor attached to an infant to acquire activity information indicating the intensity of exercise of the infant in action in a time series, and the acquired activity information is a plurality of activities. The data is classified into intensity levels and aggregated. Thereby, even if an observer and a camera do not always track an infant, the tendency of the infant's activity can be grasped quantitatively.

  The present invention also relates to an infant activity analysis system, an activity information acquisition unit that acquires, in a time series, activity information indicating exercise intensity of the infant using an acceleration sensor attached to the infant, and an activity information acquisition unit And a data totaling unit that classifies and aggregates the activity information acquired by a plurality of intensity levels. Thereby, the aggregated data for grasping the tendency of the activity of the infant is acquired, and the activity analysis method of the present invention is realized.

  According to the present invention, it is possible to easily analyze an activity even for an infant who is able to move by himself / herself.

It is a block diagram of the activity analysis system of 1st to 4th embodiment. It is a figure which shows the example of the log data containing activity information. (A) And (b) is a figure which shows the example of the total result of the activity information totaled for every infant. It is a relationship figure of the appearance rate (vertical axis) of the activity information whose exercise intensity is 1 and the actual age (horizontal axis) of the infant. It is a figure which shows the example of the activity record for every infant. It is a figure which shows the example of article information. It is a figure which shows the example of risk information.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In addition, in each drawing, the same code | symbol is attached | subjected to the same component and description is abbreviate | omitted suitably.

<First embodiment>
FIG. 1 is a configuration diagram of an activity analysis system 100 according to the embodiment of this invention.

  The activity analysis system 100 according to the present embodiment is a computer system for analyzing the activity of an infant and includes an activity information acquisition unit 10 and a data totaling unit 20. The activity information acquisition unit 10 is means for acquiring, in time series, activity information AI indicating the exercise intensity of the infant B using the acceleration sensor 12 attached to the infant B. The data totaling unit 20 is a means for classifying the activity information AI acquired by the activity information acquisition unit 10 into a plurality of intensity levels and totaling them.

The activity analysis system 100 is configured by combining an activity information acquisition unit 10, an activity analysis device 90, and a display output unit 80. The activity analysis device 90 is a so-called computer, hardware constructed with general-purpose devices such as a CPU (Central Processing Unit), a memory, and an input / output interface, and a dedicated logic circuit constructed to execute a predetermined processing operation. Or a combination thereof. As the memory, a ROM (Read Only Memory), a RAM (Random Access Memory), or a portable storage medium can be used.
In addition to the data totaling unit 20, the activity analysis device 90 includes an activity characteristic calculation unit 30, a storage unit 40, and a determination unit 50.
The data totaling unit 20, the activity characteristic calculation unit 30, and the determination unit 50 are realized by a CPU, and the storage unit 40 is realized by a memory. Note that the storage unit 40 storing information means that the storage unit 40 has a function of storing data, and it is not always necessary to store the data. In addition, the storage unit 40 is configured by a plurality of memories, and may store various information (relation information MR, article information PI, risk information RI, and the like) described later in a distributed manner.

The activity information acquisition unit 10 reads the acceleration signal acquired by the acceleration sensor 12, and calculates exercise intensity by various arithmetic processes as described below. The acceleration sensor 12 may be a uniaxial sensor or a triaxial sensor. The activity information acquisition unit 10 may extract an acceleration component in the gravity direction from the acceleration signal acquired by the acceleration sensor 12. Various arithmetic processes for calculating exercise intensity from acceleration signals have been proposed so far, and are not particularly limited. An example is shown below.
First, an acceleration signal sequentially output from the acceleration sensor 12 is converted into a voltage signal, digital conversion and noise removal are performed, and processing such as synthesis of triaxial acceleration or extraction of a gravity direction component is performed to generate acceleration data. In the present specification, the acceleration means a value obtained by removing the influence of the gravitational acceleration of the earth. That is, an acceleration (for example, 1.2 G) exceeding 1 G (G is the gravitational acceleration) is applied to the subject (infant B) trying to stand up together with the gravitational acceleration. (In this case, 0.2G).
Next, the activity information acquisition unit 10 determines the magnitude of acceleration by comparing the acceleration data with a threshold value, and calculates exercise intensity based on the determination result. The threshold value is stored in advance in the activity information acquisition unit 10. Here, the activity information acquisition unit 10 always measures acceleration from a target person (in this embodiment, an infant B) who is active with the acceleration sensor 12 attached, and a predetermined time interval (hereinafter referred to as an intensity determination interval). The exercise intensity is calculated every moment. The intensity determination interval can be 1 second or more and 10 seconds or less, for example, 4 seconds or 6 seconds. By setting the intensity determination interval to 10 seconds or less, the motion of the subject (infant B) can be analyzed in detail.
The activity information acquisition unit 10 generates one or a plurality of acceleration data based on the acceleration signal from the acceleration sensor 12 at each intensity determination interval and determines the magnitude of the acceleration. Preferably, a plurality of acceleration data is generated at each intensity determination interval, and the magnitude of the acceleration is determined. In this case, the sampling interval at which the acceleration sensor 12 acquires the acceleration signal is set shorter than the intensity determination interval, and more preferably, the acceleration sensor 12 measures the acceleration signal a plurality of times within the time of the intensity determination interval. Set to For this reason, by setting the intensity determination interval to 1 second or more, many acceleration signals can be measured by the acceleration sensor 12 within the time of the intensity determination interval without excessively shortening the sampling interval of the acceleration sensor 12. . The activity information acquisition unit 10 may determine the exercise intensity of the subject (infant B) within the intensity determination interval based on the number of occurrences for each determined magnitude of acceleration. Thereby, it is suppressed that the exercise intensity is determined to be excessive due to instantaneously detecting a large acceleration. Also, as in the case of exercise intensity 2 and exercise intensity 3, which will be described later, even if the detected acceleration is the same, the exercise intensity is distinguished into different intensity levels based on the magnitude of the frequency of the acceleration. be able to.
The activity information acquisition unit 10 may determine the exercise intensity by averaging the accelerations counted within the time of the intensity determination interval. This prevents an excessively high exercise intensity (for example, 4 or more) from being excessively determined by a large acceleration recorded at the moment when the infant B stands up or falls.
As described above, the activity information acquisition unit 10 generates one piece of activity information AI (see FIG. 2) for each intensity determination interval based on a plurality of acceleration data acquired within the time of the intensity determination interval. can do. Hereinafter, individual activity information AI may be referred to as counting, and generation of activity information AI may be referred to as counting exercise intensity.

  As the activity information acquisition unit 10, a so-called activity meter with a built-in acceleration sensor 12 can be used, but is not limited thereto. The activity information acquisition unit 10 may be realized by being distributed between the acceleration sensor 12 and the activity analysis device 90. That is, the activity information acquisition unit 10 may be configured by the acceleration sensor 12 that is detachably attached to the subject (infant B) and the intensity calculation unit (not shown) provided in the activity analysis device 90. In this case, the intensity calculation unit may acquire an analog acceleration signal or digital acceleration data measured by the acceleration sensor 12 via the input unit 60 and perform various arithmetic processes to calculate the exercise intensity.

  FIG. 2 is a diagram illustrating an example of log information LD including activity information AI and activity information AI. The activity information AI refers to individual data that represents the exercise intensity of the subject (infant B) wearing the acceleration sensor 12 or that can be converted into exercise intensity. The activity analysis device 90 aggregates a large number of activity information AI acquired by the activity information acquisition unit 10 by the data aggregation unit 20 and performs various processes by the activity characteristic calculation unit 30 and the determination unit 50.

The activity information acquisition unit 10 may acquire the activity information AI in association with time information as shown in FIG. 2 and generate log data LD including time series information of exercise intensity. The log data LD is a set of a plurality of sets of time information and activity information AI associated with each other. The time information is information indicating the time when the acceleration sensor 12 is acquired or the time when the activity information acquisition unit 10 calculates the individual activity information AI from the acceleration signal. In the present embodiment, the activity information acquisition unit 10 (see FIG. 1) has an intensity determination interval of the activity information AI of 4 seconds, and the log data LD illustrates the case where the activity information AI is acquired at intervals of 4 seconds.
The log data LD is generated by the activity information acquisition unit 10, and the activity analysis device 90 acquires the activity information AI and time information from the activity information acquisition unit 10 in association with each other. May be generated.
A method for utilizing the log data LD will be described later in the second embodiment.

  The activity analysis device 90 includes an input unit 60 that acquires various types of information from the activity information acquisition unit 10, and displays various outputs from the activity characteristic calculation unit 30 and the determination unit 50 and information such as the log data LD described above. Is further provided. The input unit 60 and the output unit 70 are realized by an input / output interface. The input unit 60 is a receiving unit that acquires information such as activity information AI and log data LD from the activity information acquisition unit 10, as well as a keyboard and a mouse for receiving various operation inputs from an operator of the activity analysis system 100. An input device is provided as an input / output interface. When the activity information acquisition unit 10 is realized by the intensity calculation unit (not shown) included in the activity analysis device 90 and the acceleration sensor 12, the input unit 60 receives an analog acceleration signal or digital acceleration data from the acceleration sensor 12. May be. The display output unit 80 is, for example, a display device, and the output unit 70 is an output signal transmission unit.

  By using the activity analysis system 100 of the present embodiment, a method for analyzing the activity of the infant B (activity analysis method: hereinafter referred to as this method) is realized. This method uses the acceleration sensor 12 attached to the infant B to acquire the activity information AI indicating the exercise intensity of the active infant B in time series, and classifies the acquired activity information AI into a plurality of intensity levels. And tabulate.

The number of strength steps is not particularly limited, but is 3 or more, preferably 5 or more. The number of intensity steps may be the same as the number of exercise intensity steps, but is not limited thereto. Two or more levels of exercise intensity counted by the activity information acquisition unit 10 as the activity information AI may be aggregated as one intensity level.

The age of infant B whose activity is analyzed by this method is preferably 90 days or more and 720 days or less, and more preferably 180 days or more and 480 days or less. The growth of infants varies greatly from individual to individual. When the age is about 90 days, the baby's neck gets set up and begins his activities. After that, a turn-over operation is started, and when the age is about 180 days, the number of operations that sit and play and get caught by objects increases. When the age is about 240 days, he / she actively becomes hungry and crawls on all fours. And when the age is about 270 days to about 300 days, he / she can actively stand and walk around, and when about 360 days, he can stand alone. Thereafter, up to about 480 days, a standing motion and a high-high motion appear, and thereafter, the high-high motion gradually disappears, and it becomes possible to walk and run enough.
Infants aged 720 days or less, especially 480 days or less, are difficult to fully communicate through language, so how long they spent their activities such as high and high It is difficult to explain orally. For this reason, the activity information AI is acquired using the acceleration sensor 12 as in the present method, and the activities can be accurately analyzed by collecting the activity information AI.

Table 1 below shows types of movement of infants and exercise intensity for each movement. Table 1 shows typical movements and exercise intensity for infants who still cannot walk enough. This exercise intensity is extracted by calculating the acceleration component in the direction of gravity from the acceleration signal acquired by the acceleration sensor 12 attached to the infant's waist (back side), and determining the magnitude of the acceleration in a plurality of stages . It is classified.

  An infant's exercise intensity of 0 is a state in which acceleration in the direction of gravity is substantially not generated, and corresponds to a state of being awake and resting while sleeping or lying down. Specifically, the acceleration is in the gravitational direction less than 0.1 G, preferably less than 0.06 G. If an adult is a subject, the state in which the same acceleration is detected by the activity information acquisition unit 10 is, for example, during sleep or at rest, and is converted into METs (METs: Metabolic equivalents) representing oxygen consumption and metabolic efficiency. Then, it corresponds to 1.5 METs or less. Hereinafter, an exercise intensity of 0 is referred to as exercise intensity 0. This is also true for exercise intensities of 0.5 or higher.

  In the present specification, a state where the exercise intensity is 0 is treated as being stationary rather than active. This is because, for example, it is difficult to discriminate weak accelerations of less than 0.1 G or less than 0.06 G, for example, to identify whether you are sleeping or awake and rest in a supine position. In addition, as described later, it is preferable to calculate the activity characteristics of the infant based on the number of times or the frequency of acquisition of the activity information AI during the activity, that is, when the exercise intensity is 0.5 or more.

In the following, the exercise intensity of active infants will be described.
The infant's exercise intensity 0.5 is a state in which a slight acceleration occurs in the direction of gravity, which corresponds to a state in which the upper body is moved up and down, that is, a state of sitting and playing. Specifically, it can be set that the acceleration in the gravitational direction of 0.06 G or more and less than 0.2 G, for example, is continuously detected. When an adult is a subject, the same acceleration is detected, for example, in a sitting position and is about 1.6 METs.
The infant's exercise intensity 1 is a state where acceleration in the direction of weak gravitational force is generated, and corresponds to a state where there is a vertical movement, that is, a state where he / she is standing and walking. Specifically, for example, it can be set that the acceleration in the gravity direction of 0.2 G or more and less than 0.5 G is continuously detected. When an adult is a subject, the same acceleration is detected, for example, while walking, which corresponds to about 1.8 METs.
The infant's exercise intensity 2 is a state in which moderate acceleration in the gravitational direction is intermittently generated, and a relatively large up-and-down motion, for example, a high-speed or low-speed walking at a low speed. It hits. This intensity of exercise may also apply when you are turning over during sleep or awakening. Specifically, for example, it can be set that the acceleration in the gravity direction of 0.5 G or more and less than 1.0 G is detected intermittently. When an adult is a subject, the same acceleration is detected, for example, in a standing work state, which is about 2.3 METs.
The infant's exercise intensity 3 is a state in which moderate acceleration in the gravitational direction occurs frequently, for example, a state in which high / high is performed at high speed. Specifically, for example, it can be set that the acceleration in the gravitational direction of 0.5 G or more and less than 1.0 G is continuously detected at a frequency higher than the exercise intensity 2 as in the exercise intensity 2. When an adult is a subject, the same acceleration is detected, for example, when the stairs are moving up and down, which corresponds to about 2.9 METs.
An exercise intensity of 4 or more corresponds to an action of 3.6 METs or more such as jogging in an adult, but is not substantially detected when targeting an infant who still cannot walk enough.

In this method, as shown in Table 1, the exercise intensity is classified into 6 levels such as 0, 0.5, 1, 2, 3 and 4 or more, and the intensity levels are called 1 to 6 in order. When the intensity level is 1, the infant is stationary, and when the intensity level is 2 or more, the infant is active.
The exercise intensity to be obtained from an infant by this method corresponds to the level of relatively mild activity of an adult as shown in Table 1. Therefore, acceleration data or activity information AI can be acquired using an acceleration sensor built in a commercially available activity meter intended for use by adults.

In this method, the activity information AI indicating the exercise intensity of the active infant B is acquired in time series, and is classified into a plurality of intensity levels and tabulated. The aggregated results can be used for various purposes. In this embodiment, the case where the activity characteristic of the infant B is acquired based on the acquisition frequency of the activity information AI classified into at least one intensity level is illustrated.
Here, the activity characteristic of the infant B refers to information indicating a tendency of the activity of the infant B. The activity characteristic may be information indicating an average activity tendency of a plurality of infants B as well as information indicating individual activity trends of the infants B. Examples of activity characteristics include the development level of infant B's athletic ability, a breakdown of activity contents of the infant B within a predetermined time, the amount of energy consumed by the infant B within a predetermined time, and the like. I can't.
By calculating the degree of development of athletic ability as the activity characteristic of the infant B, the athletic ability of the infant B that cannot be grasped by general parameters such as actual age, height, and weight can be evaluated. In addition, by grasping the degree of development of the athletic ability of the infant B, various useful information can be obtained as will be described later in the third to fifth embodiments. In addition, by acquiring a breakdown of the activity contents of the infant B within a predetermined time (for example, one day) as the activity characteristics of the infant B and calculating the amount of energy consumed, the parent and the nursery teacher constantly monitor the infant B Even if it is not doing, the health state and fatigue state of the infant B can be objectively grasped.

  Hereinafter, in the present embodiment, a method for calculating the degree of development of athletic ability as the activity characteristic of the infant B will be described.

According to the study by the present inventors, the infant spends much of the day resting in sleep or supine position with a state of exercise intensity 0, and this length varies greatly between individuals. In other words, the length of the total time during the activity, that is, the state where the exercise intensity is 0.5 or more (hereinafter referred to as the active time) varies greatly among individuals. On the other hand, the ratio of the generation time of specific exercise intensity and the generation time of each exercise intensity occupying the length of the active time is small, and especially the activity of activity intensity 1 that shows grabbing and walking. It was newly found that the ratio of occurrence time significantly correlated with the actual age (age).
The present inventors acquired activity information AI from a large number of infants in time series, and extracted and counted the exercise intensity generated in the active infant without omission. Then, the correlation between exercise intensity and the infant's actual age (age) was statistically analyzed.

  Specifically, with 31 infants whose actual age (days of age) is 121 days or more and 425 days or less as subjects, acceleration in the gravity direction is measured by the activity information acquisition unit 10 attached to the waist (back side) of each subject. Measurements were made in time series and classified into the exercise intensity and intensity levels shown in Table 1 above. From each subject, the activity information AI indicating the exercise intensity during the activity was acquired 450 counts or more, preferably 900 counts or more, and classified into a plurality of intensity levels and tabulated. Then, out of the total counts of the acquired exercise intensity (that is, the exercise intensity of 0.5 or more), the acquisition frequency of the exercise intensity 1 (intensity level = 3) (hereinafter, the acquisition frequency of the activity information AI of a certain exercise intensity appears) (Sometimes referred to as a rate) was calculated for each subject.

FIGS. 3A and 3B are diagrams showing an example of the totaled result of the activity information AI totaled for each infant. The intensity level and exercise intensity correspond to Table 1 above.
In the activity analysis system 100 of the present embodiment, the activity information AI acquired from the activity information acquisition unit 10 is classified into each intensity level by the data aggregation unit 20 based on the exercise intensity, and aggregated, and FIG. The total result shown in (b) can be obtained.

FIG. 3A shows that activity information AI of a total of 2500 counts has been acquired from the infant X, of which, for example, 1000 counts of activity information AI of exercise intensity 0 have been acquired. Moreover, the count number and acquisition frequency (appearance rate) of the activity information AI in each intensity level out of the total 1500 counts of the intensity level = 2 to 6 indicating that the activity is in progress are also shown in FIG. In this example, the appearance rate of the activity information AI at the intensity level = 3 (exercise intensity 1) is 3.0%.
FIG. 3B shows that the activity information AI of 2170 counts in total is acquired from the infant Y different from the infant X, and among them, for example, 800 counts of activity information AI of exercise intensity 0 are acquired. In addition, FIG. 3 also shows the count of activity information AI and the acquisition frequency (appearance rate) of each intensity level out of the total 1370 counts of intensity levels = 2 to 6 indicating that the activity is in progress. In this example, the appearance rate of the activity information AI at the intensity level = 3 (exercise intensity 1) is 8.8%.
As described above, the data totaling unit 20 sets the frequency of acquiring the activity information AI at each intensity level with an exercise intensity of 0.5 or more among the many activity information AI acquired by the activity information acquiring unit 10 in time series. The acquisition frequency (appearance rate) can be calculated by dividing by the count value excluding the activity information AI of exercise intensity 0 corresponding to the stationary state.

  Instead of this embodiment, the activity information acquisition unit 10 does not transmit the activity information AI of exercise intensity 0 when the target person (infant) is stationary to the activity analyzer 90 by deleting the activity information AI. It is good. That is, the data totaling unit 20 of the activity analysis apparatus 90 may select and acquire only the activity information AI with an exercise intensity of 0.5 or more acquired from the active subject (infant).

FIG. 4 is a relationship diagram between the appearance rate of exercise intensity 1 (vertical axis y: unit is [%]) and the infant's actual age (horizontal axis x: unit is [day]). As shown in FIG. 4, the appearance rate of exercise intensity 1 showed a positive correlation with the infant's actual age, and the contribution rate R ² = 0.58 was a high value. From this, it was found that both are well correlated. The regression equation shown in FIG.
Appearance rate of exercise intensity 1 = 0.021 × infant actual age−0.45 (1)
It can be expressed as.
When formula (1) is transformed,
Infant's real age [days] = Appearance rate of exercise intensity 1 [%] × 48 + 23 (2)
It becomes.
That is, the present inventors have found a correlation between the general infant's motor ability and actual age by the single regression equation (2) with the appearance rate of exercise intensity 1 as an explanatory variable. Therefore, according to the present method, the activity level of the infant B is compared with that of a general infant by obtaining the activity information AI and calculating the appearance rate of exercise intensity 1 for the infant B who is the subject. It can be determined whether it is developing or inferior.
That is, according to this method, it is possible to calculate the equivalent actual age evaluated from the viewpoint of athletic ability regarding the infant B serving as the subject. The equivalent real age is one of the activity characteristics indicating the motor ability of the infant and is an index indicating the motor development stage indicating how much the motor ability of the infant is developed in terms of actual age.

In addition, the present inventors have further studied and clarified that the actual age of the infant can be accurately estimated based on the activity information AI by various regression equations other than the single regression equation (2).
For example, the equivalent real age of an infant can be estimated with higher accuracy by the following multiple regression equation (3) using the appearance rates of exercise intensities of 0.5, 1, and 2 as explanatory variables.
Infant's real age [days] = appearance rate of exercise intensity 0.5 [%] × (−952) + appearance rate of exercise intensity 1 [%] × (−912) + appearance rate of exercise intensity 2 [%] × (−961) +95264 (3)
Contribution rate R ² in the above multiple regression equation (3) is reached 0.71, it can be seen that estimate the actual age of the infants at higher accuracy than the single regression equation (2).

Further, according to the study by the present inventors, it is clear that there is also a correlation between the average exercise intensity of all counts excluding exercise intensity 0, that is, the exercise intensity of 0.5 or more, and the infant's actual age. It was. This simple regression equation can be expressed by the following equation (4).
Infant's real age [days] = average exercise intensity × 705-216 (4)
Contribution rate R ² in simple regression (4) above was 0.45, it can be seen that there is a significant correlation also between the actual age of the average exercise intensity and infants.

Based on the above, activities classified into one of the intensity levels are obtained by acquiring activity information indicating the exercise intensity of active infants in time series and classifying them into multiple intensity levels as in this method. It is possible to estimate the actual age of the infant based on the information acquisition frequency and the average value of exercise intensity.
In particular, based on the frequency of acquisition of activity information categorized into intensity level = 3 corresponding to exercise intensity 1 performed at a low acceleration compared to walking high and high, such as standing and walking, single regression equation (2) When the actual age is estimated based on the multiple regression equation (3), it is possible to reduce the influence of physical strength having a large individual difference for each infant and to accurately estimate.
That is, in this method, by performing regression analysis on the acquisition frequency of activity information AI classified into at least one intensity level, for example, by calculating the equivalent age by referring to the above regression equations (2) to (4) As an activity characteristic of the infant B who is the subject, it is possible to acquire the degree of development of the motor ability of the infant B.

  The degree of development of athletic ability, which is an activity characteristic, may be acquired as the equivalent actual age (day age) as described above, or may be acquired as a ratio obtained by dividing the equivalent actual age by the actual age of the infant B. Alternatively, the ratio may be determined by a predetermined threshold value, and may be acquired by dividing into a plurality of levels such as excellent / good / good.

Returning to FIG. 1, the activity analysis system 100 of this embodiment includes a storage unit 40 and an activity characteristic calculation unit 30. The activity characteristic calculation unit 30 is a means for calculating the development level of the infant's athletic ability based on the acquisition frequency of the activity information AI classified into at least one intensity level.
The storage unit 40 stores relationship information MR indicating the correlation between the actual ages of the plurality of sample infants and the acquisition frequency (appearance rate) of the activity information AI acquired from the plurality of sample infants.
Then, the activity characteristic calculation unit 30 refers to the storage unit 40, and based on the acquisition frequency (appearance rate) of the activity information AI acquired from the infant B who is the subject and the relationship information MR, the activity characteristics of the infant B The degree of development (for example, equivalent actual age) is calculated.

  The relationship information MR is information indicating the values of the coefficients and constant terms in the regression equations (2) to (4), for example. As a result, the activity information AI is acquired from the infant B who is the subject, the acquisition count (appearance rate) of one or more exercise intensity counts is calculated by the data totaling unit 20, and transferred to the activity characteristic calculating unit 30. The activity characteristic calculation unit 30 can calculate the equivalent actual age of the infant B by referring to the storage unit 40 and calling the relationship information MR.

  The storage unit 40 may store information indicating the values of the coefficients and constant terms of the single regression equations (2) and (4) as the relationship information MR, or a plurality of coefficients and constants of the multiple regression equation (3). It may be information indicating the value of a term. As described above, since the contribution rate in the multiple regression equation (3) is higher than the contribution rate in the single regression equations (2) and (4), the storage unit 40 stores the relationship information including the coefficient of the multiple regression equation (3). It is preferable to store MR. In other words, it is preferable that the relationship information MR includes a coefficient of a multiple regression equation using the acquisition frequencies of a plurality of activity information AIs, which indicate that they are both active and are classified into different intensity levels, as explanatory variables.

An example of the operation of the activity analysis system 100 will be described.
The activity information acquisition unit 10 acquires the activity information AI from the active and stationary infant B in a time series and transmits it to the activity analyzer 90. The data totaling unit 20 of the activity analyzing apparatus 90 is active together with the first intensity stage (exercise intensity 0) indicating that the activity information AI acquired by the activity information acquiring unit 10 is stationary. And at least second and third intensity stages (for example, exercise intensity 0.5, 1, 2, 3) having different exercise intensity. Here, the second intensity level is exercise intensity 1, and the third intensity level is exercise intensity 0.5 or the like.
Then, the activity characteristic calculation unit 30 excludes the activity information AI (exercise intensity 0) of the stationary infant B, and the activity information AI (exercise intensity 1) classified into the second intensity stage indicating that the activity is active. The acquisition frequency (appearance rate) is calculated based on the acquisition number (count number). Furthermore, the activity characteristic calculation unit 30 calls the relationship information MR with reference to the storage unit 40 as described above, and calculates the equivalent actual age of the infant B. Here, the activity characteristic calculation unit 30 may call the relationship information MR indicating the coefficient of the single regression equation (2) having the exercise intensity 1 as an explanatory variable and the value of the constant term.

Alternatively, the activity characteristic calculation unit 30 includes activities of a plurality of intensity stages including a second intensity stage (exercise intensity 1) and a third intensity stage (exercise intensity 0.5) indicating that the infant B is active. The acquisition frequency of the information AI may be calculated. Preferably, the acquisition frequency of activity information AI at all intensity levels indicating that the infant B is active may be calculated. Then, the activity characteristic calculation unit 30 includes a plurality of coefficients and constant term values in the multiple regression equation (3) having the acquisition frequency of the activity information AI of a plurality of intensity levels including the second and third intensity levels as explanatory variables. It is preferable to call the relationship information MR indicating
Thereby, according to the activity analysis system 100 of the present embodiment, the acquisition of the activity information AI of a plurality of intensity levels such as the second intensity level (exercise intensity 1) and the third intensity level (exercise intensity 0.5). The multiple regression analysis including the frequency as an explanatory variable can be performed to calculate the development level (equivalent actual age) of the infant B's motor ability.

<Second embodiment>
In the second embodiment, a method of using the log data LD shown in FIG. 2 will be described. In the present embodiment, the activity characteristic of the infant is calculated based on the number of acquisitions of the activity information AI classified into at least one intensity level.

  FIG. 5 is a diagram illustrating an example of an activity record for each infant. Here, regarding the infant X shown in FIG. 3A, the count value of the activity information AI for each exercise intensity is converted into action time and displayed as “play time”. Such log data LD is generated by the activity characteristic calculation unit 30, transmitted to the display output unit 80 via the output unit 70, and can be displayed and output by the display output unit 80.

As described above, the activity information AI is calculated at a predetermined intensity determination interval by the activity information acquisition unit 10 and sent to the data totaling unit 20 of the activity analysis apparatus 90. As shown in each figure of FIG. 3, the data totaling unit 20 multiplies the count value totaled for each intensity stage by an intensity determination interval (for example, 4 seconds), so that the infant X is in the measurement period of the activity information AI. The activity time information TI is generated by calculating the activity time of the exercise intensity.
The activity time information TI is time information related to the activity time performed at the exercise intensity with respect to the intensity stage related to one or more exercise intensity. The data totaling unit 20 may generate time information of activity time related to one exercise intensity as the activity time information TI, and may generate time information of activity time related to a plurality of, preferably all, exercise intensity. .

  As shown in FIG. 2, the activity information acquisition unit 10 of the present embodiment acquires the activity information AI in time series in association with the time information. Accordingly, the intensity level of the acquired activity information AI includes a plurality of intensity levels indicating that the infant is active = 2 to 6 and other intensity levels indicating that the infant is sleeping or the like = 1 (exercise intensity 0). The data totaling unit 20 is an activity characteristic (activity time information TI), which is a part or all of the total time aggregated as the activity time for a plurality of intensity levels = 2 to 6 and the other intensity levels = 1. And time.

The activity time information TI generated by the data totaling unit 20 may include the above-described stationary time. Here, based on the total number of counts in which the activity information AI is recorded as exercise intensity 0, the total time during which the user sleeps or awakes and is stationary may be defined as the stationary time. Alternatively, the stationary time may be calculated based on a part of the total count.
The data totaling unit 20 of the present embodiment determines that the infant is sleeping when the exercise intensity 0 indicated by the activity information AI included in the log data LD continues for a predetermined time (for example, 30 minutes) or longer. May be. When the data totaling unit 20 detects that the activity information AI with the exercise intensity of 0 is generated over a predetermined time period as described above, the data totaling unit 20 reads the start time and the end time of the predetermined time from the log data LD, and The stationary time is calculated based on the count number of the activity information AI being performed. As shown in FIG. 5, the data totaling unit 20 may output a stationary time as “nap time” as part of the activity time information TI.

  The activity analysis system 100 may record and save the activity time information TI indicating the activity time concerning each exercise intensity in the storage unit 40 in association with the identification number indicating the corresponding infant (for example, the infant X). Then, when the activity analysis system 100 newly generates the activity time information TI of the infant X and outputs it on the display output unit 80, as shown in FIG. The time information TI may be called from the storage unit 40 and output together.

  As described above, the activity analysis system 100 according to the present embodiment can acquire a breakdown of activity contents of infants as activity characteristics based on the number of acquisitions of activity information AI. Specifically, in this embodiment, the activity time for each of the plurality of intensity stages is calculated as the activity characteristic. Thereby, an infant's parent or childcare worker can easily grasp the health condition and fatigue state of the infant in comparison with the breakdown of the latest activity time.

<Third embodiment>
In the first embodiment, it has been described that the equivalent real age representing the degree of development of motor ability is calculated based on the result of regression analysis as the activity characteristic of the infant as the subject. 3rd embodiment demonstrates the 1st method of utilizing this equivalent real age. The first method is to evaluate the function of an article used by being attached to an infant's body.

In the first method, first, activity information AI is acquired in time series from the infant B wearing the evaluation target article PD (see FIG. 1) on the body, and the activity characteristics of the infant B are calculated. Thereafter or before that, the activity characteristic of the infant B is similarly measured in a state where the evaluation object PD is not worn on the body, or in a state where the infant B is wearing another article instead of the evaluation object PD. calculate.
Next, based on the calculated activity characteristics, the ease of movement of the infant's body in the state where the evaluation target article PD is mounted is evaluated.
Thereby, it is possible to grasp whether or not the activity characteristics of the infant B change when the evaluation target article PD is attached or when the evaluation target article PD is attached instead of another article.

  The evaluation target article PD is not particularly limited as long as the infant wears on the body, and examples thereof include a disposable diaper, clothes, toys, and decorations. In particular, in the case of an article that is worn in close contact with the body, such as a paper diaper or clothes, the infant's B's original motor ability may be hindered. For this reason, by comparing the equivalent actual age of the infant B with the evaluation target article PD attached to the equivalent actual age of the infant B with the other article attached, the evaluation target article PD is placed on the body. It is possible to quantitatively evaluate the ease of movement of the body in the wearing state. In particular, since infants cannot communicate sufficiently in language, it is often impossible to convey the feeling of wearing the evaluation object PD to others in language. According to the present embodiment, the function of the evaluation target article PD can be objectively evaluated from the viewpoint of ease of movement of the body through a change in the equivalent age of the infant B before and after the evaluation target article PD is mounted.

<Fourth embodiment>
4th embodiment demonstrates the 2nd method of utilizing the considerable real age showing the development degree of the infant's athletic ability. The second method proposes a recommended target article based on the equivalent actual age of the infant. FIG. 6 is a diagram illustrating an example of the article information PI.

The activity analysis system 100 according to the fourth embodiment includes a storage unit 40 and a determination unit 50 as shown in FIG. The storage unit 40 of the present embodiment stores article information PI indicating each of a plurality of target articles that the infant B wears on the body in association with the development level of the motor ability of the infant B recommended for use. Keep it.
Further, as described in the first embodiment, the activity characteristic calculation unit 30 calculates the development level of the infant's motor ability based on the acquisition frequency of the activity information AI classified into at least one intensity level. Then, the determination unit 50 refers to the storage unit 40 and extracts and outputs the article information PI based on the degree of development calculated by the activity characteristic calculation unit 30.

  As a result, the target article is output according to the degree of development of the athletic ability of the infant B that cannot be grasped by general parameters such as actual age, height, and weight. For example, in the case of a disposable diaper, the age and weight of an infant recommended as a user are generally displayed (user display) on the product package. However, for example, in the case of an infant who is actively exercising while having a low actual weight (month) and a light weight, the product selected according to the user display may be cramped and uncomfortable. . On the other hand, by extracting and outputting the item information PI based on the development level of the infant's athletic ability as in the present embodiment, the parent of the infant can select the target item that does not cause such discomfort. You can know.

  The target article related to the article information PI is not particularly limited as long as the infant can wear it on the body, and examples thereof include clothes, toys, and decorations in addition to the above-described paper diapers.

  As shown in FIG. 6, the article information PI may be stored in the storage unit 40 in a table format in which the development level (equivalent actual age) of the infant and the infant and the actual age of the infant are associated with each other. However, instead of or in addition to the infant's actual age, height or weight may be associated with the item information PI.

  In the table shown in FIG. 6, the target articles AAA, BBB, CCC, and DDD are for infants 0 to 119 days, 120 to 269 days, 270 to 359 days, and 360 days or more, respectively. This is the name of the standard article. A target article having a name with “+” at the end, such as the target article AAA +, is the same size as the target article AAA that is a standard article, but is designed to be more durable than the standard article. The number of + indicates the strength of the tendency. In addition, the target article with the name “-” added to the end, such as the target article BBB-, has the same size as the target article BBB, which is a standard article, but is more movable and flexible than the standard article. Easy-to-design goods. The number of-indicates the strength of the tendency.

The operator of the activity analysis system 100 inputs real age information indicating the actual age (days) of the infant to be a subject and physical information indicating the height or weight to the activity analysis device 90 through the input unit 60 (FIG. 1). The determination unit 50 acquires the input real age information and physical information. On the other hand, as described in the first embodiment, the activity information AI is acquired from the infant, and the activity characteristic calculation unit 30 calculates the equivalent actual age. The determination unit 50 may extract the item information PI based on the actual age information, the body information, and the calculation result of the equivalent actual age with reference to the table of the item information PI stored in the storage unit 40.
Thereby, the target article concerning the article information PI suitable for the infant can be presented in consideration of both the actual age and the equivalent actual age of the infant who is the subject. Specifically, for example, in the case of an infant having an actual age of 220 days and an equivalent actual age of 280 days, the item information PI relating to the target item BBB + is extracted. Thereby, it is possible to output and recommend the target article BBB + having a size suitable for the actual age of 220 days and excellent in durability for the infant whose athletic ability is developed as compared with the actual age. For example, in the case of an infant whose actual age is 280 days and whose corresponding actual age is 240 days, the item information PI relating to the target item CCC- is extracted. As a result, for the infants whose motor ability is not developed as compared with the actual age, the size is suitable for the actual age of 280 so that the physical activity is not hindered by the target article. The target article CCC- having excellent flexibility can be output and recommended.

<Fifth embodiment>
5th embodiment demonstrates the 3rd method of utilizing the equivalent real age showing the development degree of the infant's athletic ability. The third method presents information indicating the risk of accidents occurring based on the infant's equivalent age. FIG. 7 is a diagram showing an example of the risk information RI.

The activity analysis system 100 of the fifth embodiment includes a storage unit 40 and a determination unit 50 as shown in FIG. The storage unit 40 of the present embodiment stores a plurality of pieces of risk information RI each indicating an accident event caused by an infant's movement in association with the degree of development of the infant's motor ability (equivalent actual age). .
Further, as described in the first embodiment, the activity characteristic calculation unit 30 calculates the development level of the infant's motor ability based on the acquisition frequency of the activity information AI classified into at least one intensity level. Then, the determination unit 50 refers to the storage unit 40 and extracts and outputs the risk information RI based on the degree of athletic ability development (equivalent actual age) calculated by the activity characteristic calculation unit 30.

  Thereby, it is possible to present the risk of occurrence of an accident that cannot be grasped only by the infant's height and actual age. For example, in FIG. 7, one or a plurality of risk information RI is recorded in association with each corresponding real age indicating the exercise ability of the infant. Thus, for example, even if the actual age is 200 days, “falling”, “misingestion”, “burn”, etc. for the parents of infants whose equivalent actual age is calculated to be 210 days or more and 239 days or less It can be notified that there is also an accident risk.

  Further, as shown in FIG. 7, the storage unit 40 stores the prevention information PR indicating the accident event prevention method in association with the risk information RI. When the determination unit 50 refers to the storage unit 40 and extracts the risk information RI based on the equivalent actual age, the determination unit 50 may also extract the corresponding prevention information PR. Then, the associated preventive information PR may be output together with the extracted risk information RI. As a result, it is possible not only to notify the risk of occurrence of an accident, but also to teach the prevention method.

  In addition, in FIG. 7, the risk level called a risk map switches on the boundary of equivalent real age = 119 days, 179 days, 209 days, and 239 days, and there is no overlap in the equivalent real age corresponding to the risk map of each step However, the present invention is not limited to this. The numerical ranges of the equivalent real age corresponding to each stage of the risk map may partially overlap each other. For example, instead of FIG. 7, the equivalent actual age corresponding to the risk map 4 is 210 days or more and 259 days or less, and a part of the numerical range (240 days to 259 days) is adjacent to another risk map (risk map 5). It may be overlapped with a part of the numerical range. As a result, for infants whose equivalent actual age is from 240 to 259 days, both risk information RI and prevention information PR in the risk maps 4 and 5 are extracted and output, so that there is no more leaking attention. May be aroused.

  As mentioned above, although embodiment of this invention was described with reference to drawings, these are the illustrations of this invention, Various structures other than the above are also employable. The present invention is not limited to the above-described embodiment, and includes various modifications and improvements as long as the object of the present invention is achieved.

  The various components of the activity analysis system 100 of the present invention need not be independent of each other. A plurality of components are formed as one member, a component is formed of a plurality of members, one component is a part of another component, and one component is And a part of other components are allowed to overlap.

The present invention discloses the following activity analysis method and activity analysis system regarding the above-described embodiment.
<1> Time-series acquisition of activity information indicating exercise intensity of the infant in action using an acceleration sensor attached to the infant, and classifying the acquired activity information into a plurality of intensity levels Infant activity analysis method characterized by
<2> The activity analysis method according to <1>, wherein the activity characteristic of the infant is acquired based on the acquisition frequency or acquisition frequency of the activity information classified into at least one of the intensity levels.
<3> The activity analysis method according to <2>, wherein an activity time for each of the plurality of intensity stages is calculated as the activity characteristic.
<4> The activity analysis method according to <2>, wherein a degree of development of the motor ability of the infant is calculated as the activity characteristic by performing regression analysis on the frequency of acquisition of the activity information classified into at least one of the intensity levels .
<5> The activity information of the infant is calculated in a time series by acquiring the activity information from the infant wearing the evaluation target article on the body, and the evaluation target article is determined based on the calculated activity characteristic. The activity analysis method according to any one of <2> to <4>, wherein the ease of movement of an infant's body in a worn state is evaluated.
<6> An activity information acquisition unit that acquires, in a time series, activity information indicating exercise intensity of the infant using an acceleration sensor attached to the infant, and the activity information acquired by the activity information acquisition unit includes a plurality of intensity levels. An infant activity analysis system comprising a data totaling unit that classifies and classifies the data.
<7> The activity analysis system according to <6>, further comprising an activity characteristic calculation unit that calculates a development degree of the athletic ability of the infant based on an acquisition frequency of the activity information classified into at least one of the intensity levels.
<8> A storage unit for storing article information indicating a plurality of target articles used by the infant on the body in association with the degree of development of the infant's motor ability recommended for use, The activity analysis system according to <7>, wherein the article information is extracted and output based on the degree of development of the athletic ability calculated by the activity characteristic calculation unit with reference to a storage unit.
<9> A storage unit that stores a plurality of pieces of risk information each indicating an accident event caused by the infant's exercise in association with the degree of development of the infant's motor ability, with reference to the storage unit The activity analysis system according to <7>, wherein the risk information is extracted and output based on the degree of development calculated by the activity characteristic calculation unit.
<10> The activity analysis method according to any one of <1> to <5>, wherein the activity information is acquired in association with time information, and log data including time series information of exercise intensity is generated.
<11> The activity analysis method according to any one of <1> to <5>, wherein an age of the infant is 90 days or more and 720 days or less.
<12> The intensity stage includes the plurality of intensity stages indicating that the infant is active, and another intensity stage indicating that the infant is stationary, and the activity characteristics include a plurality of intensity levels. The activity analysis method according to <3>, wherein the activity time of the intensity level of the current time and the rest time that is a part or all of the total time counted as the other intensity level are calculated.
<13> The activity information is acquired in time series from the active and stationary infants, and the acquired activity information is active together with the first intensity stage indicating that the infant is stationary. The exercise intensity is classified into at least second and third intensity stages different from each other, and is classified into the second intensity stage in the total number of acquisitions excluding the activity information classified in the first intensity stage. The activity analysis method according to <4>, wherein the acquisition frequency is calculated based on the number of acquired activity information.
<14> The activity analysis according to <13>, wherein a degree of development of the athletic ability is calculated by performing a multiple regression analysis using the acquisition frequencies of at least the second and third intensity level activity information as explanatory variables, respectively. Method.
<15> A storage unit that stores correlation information indicating an actual age of a plurality of sample infants and the acquisition frequency of the activity information acquired from the plurality of sample infants, further comprising the activity characteristic The activity analysis system according to <7>, wherein the calculation unit refers to the storage unit and calculates a degree of development of athletic ability related to the infant based on the acquisition frequency acquired from the infant and the relationship information.
<16> The storage unit includes the relation information including a coefficient of a multiple regression equation having an acquisition frequency of a plurality of the activity information classified into a plurality of different intensity levels indicating that both are active. <15> The activity analysis system according to <15>.
<17> The storage unit stores prevention information indicating the accident event prevention method in association with the risk information, and further outputs the prevention information associated with the extracted risk information. The activity analysis system according to <9>.

DESCRIPTION OF SYMBOLS 10 Activity information acquisition part 12 Acceleration sensor 20 Data totaling part 30 Activity characteristic calculation part 40 Memory | storage part 50 Determination part 60 Input part 70 Output part 80 Display output part 90 Activity analyzer 100 Activity analysis system AI Activity information B, X, Y Infant LD Log data MR Relational information PD Evaluation item PI Product information PR Prevention information RI Risk information TI Activity time information

Claims (7)

Activity information indicating the exercise intensity of the infant in action is acquired in time series using an acceleration sensor attached to the infant, and the acquired activity information is a plurality of levels based on the level of the exercise intensity. Categorized and staged ,
An infant activity analysis method comprising: calculating a development level of the infant's motor ability as an activity characteristic of the infant by performing regression analysis on the frequency of acquisition of the activity information classified into at least one of the intensity levels. .   In the regression analysis, the infant is based on a single regression equation with the frequency of acquisition of the activity information classified into the intensity level corresponding to the exercise intensity corresponding to the state where the infant is grabbing and walking around as an explanatory variable. The activity analysis method according to claim 1, wherein the degree of development of athletic ability is calculated.   In the regression analysis, the infant's athletic ability based on multiple regression equations with the frequency of acquisition of the activity information classified into intensity levels corresponding to a plurality of different exercise intensities corresponding to the state of the active infant as an explanatory variable, respectively. Calculate the degree of development, and
  2. The activity analysis according to claim 1, wherein the explanatory variable includes an acquisition frequency of the activity information classified into an intensity level corresponding to an exercise intensity corresponding to a state where the infant is standing and walking. Method. Activity information indicating the exercise intensity of the infant in action is acquired in a time series using an acceleration sensor attached to the infant wearing the evaluation target article on the body, and the acquired activity information is adjusted to the level of the exercise intensity. Classify and classify into the corresponding multiple stages of strength,
The activity characteristic of the infant is calculated based on the acquisition frequency or acquisition frequency of the activity information classified into at least one of the intensity levels, and the evaluation target article is mounted based on the calculated activity characteristic. An activity analysis method characterized by evaluating the ease of movement of an infant's body. An activity information acquisition unit that acquires activity information indicating the exercise intensity of the infant in time series using an acceleration sensor attached to the infant, and the activity information acquired by the activity information acquisition unit based on the level of the exercise intensity a data aggregation unit aggregates classified into intensity stage is a plurality of stages, calculating the acquisition frequency of the activity information classified into at least one of said intensity step Te,
An activity characteristic calculation unit that calculates a development degree of athletic ability as an activity characteristic of the infant by performing regression analysis on the calculated acquisition frequency;
Infant activity analysis system equipped with. A storage unit for storing article information indicating a plurality of target articles used by the infant on the body in association with the development level of the infant's recommended exercise ability;
6. The activity analysis system according to claim 5 , wherein the article information is extracted and output based on the degree of development of the athletic ability calculated by the activity characteristic calculation unit with reference to the storage unit. A storage unit for storing a plurality of risk information each indicating an accident event caused by the infant's movement in association with the development level of the infant's motor ability;
6. The activity analysis system according to claim 5 , wherein the risk information is extracted and output based on the development degree calculated by the activity characteristic calculation unit with reference to the storage unit.

Images