JP2024018876A - Information processing system, server, information processing method, program, and learning model - Google Patents

Abstract

To provide an information processing system, a server, an information processing method, a program, and a learning model that generate especially at least one of a user's BMI information or weight information and body fat percentage information based on measurement data via a biometric data generation unit.SOLUTION: An information processing system generates biological data from measurement data of a measurement device worn by a user. The measurement device includes an LED that emits light of at least one of infrared or red, green, and blue, and a photodiode. The measurement device also includes a biometric data generation unit that generates at least one of BMI information or weight information and body fat percentage information by inputting received light amount data obtained by radiating the light from the LED onto the user's skin and receiving light by the photodiode into a corresponding learning model. The received light amount data is data of any of only a DC component of the received light, or only a high frequency band component of the received light, or only the DC component and the high frequency band component.SELECTED DRAWING: Figure 1

Description

The present disclosure relates to an information processing system, a server, an information processing method, a program, and a learning model that generate at least BMI information, weight information, or body fat percentage information based on measurement data obtained from a measurement device.

BACKGROUND ART Weight measuring devices that determine a user's weight or BMI (BODY MASS INDEX) based on a user's load value have been known, for example, as shown in Patent Document 1.

Japanese Patent Application Publication No. 2005-351830

However, as wearable device technology has developed in recent years, new methods of measuring BMI, weight, and body fat percentage are desired.

Therefore, the present disclosure provides an information processing system, a server, an information processing method, a program, and a learning model that generate at least BMI information, weight information, or body fat percentage information based on measurement data acquired from a measurement device worn by a user. I will explain about it.

An information processing system according to an aspect of the present disclosure is an information processing system that receives measurement data from a measurement device worn by a user via a network, and generates biological data from the measurement data, and the measurement device includes: The device includes an LED and a photodiode that emit at least one of infrared light, red, green, and blue light, and receives light amount data obtained by irradiating light from the LED onto the skin of the user and receiving the light with the photodiode. It includes a biometric data generation unit that generates at least one of BMI information, weight information, and body fat percentage information by inputting it into a corresponding learning model, and the received light amount data includes only the DC component of the received light, or the received light amount data. The data includes only the high frequency band component of the light, or only the DC component and the high frequency band component.

A server in one aspect of the present disclosure is a server that receives measurement data from a measurement device worn by a user via a network, and generates biometric data from the measurement data, and the measurement device includes at least an infrared or A learning model that includes an LED and a photodiode that emit red, green, or blue light, and that corresponds to received light amount data obtained by irradiating light from the LED onto the skin of the user and receiving the light with the photodiode. The data includes a biological data generation unit that generates at least one of BMI information, weight information, and body fat percentage information by inputting the information into The data is either only the frequency band component or only the DC component and the high frequency band component.

An information processing method according to an aspect of the present disclosure is an information processing method that receives measurement data from a measurement device worn by a user via a network, and generates biological data from the measurement data, the measurement device comprising: The device includes an LED and a photodiode that emit at least one of infrared light, red, green, and blue, and the biometric data generating unit irradiates light from the LED onto the skin of the user and receives the light at the photodiode. The received light amount data includes a step of generating at least one of BMI information, weight information, and body fat percentage information by inputting the obtained received light amount data into a corresponding learning model, and the received light amount data includes only the DC component of the received light, Alternatively, it is data of only the high frequency band component of the received light, or only the DC component and the high frequency band component.

A program according to an aspect of the present disclosure is a program that causes an information processing system to execute an information processing method for receiving measurement data from a measurement device worn by a user via a network and generating biological data from the measurement data. , the measuring device includes an LED and a photodiode that emit at least one of infrared light, red, green, and blue, and the information processing method includes a biological data generation unit that directs light from the LED to the user's skin. the step of generating at least one of BMI information, weight information, and body fat percentage information by inputting received light amount data obtained by irradiating and receiving the light with the photodiode into a corresponding learning model; The quantity data is data of only the DC component of the received light, only the high frequency band component of the received light, or only the DC component and the high frequency band component.

In one aspect of the present disclosure, the learning model irradiates the user's skin with light from the LED of a measurement device including a photodiode and an LED that emits at least one of infrared light, red light, green light, and blue light to emit light from the photodiode. When received light amount data obtained by receiving light is input, at least one of the user's BMI value, weight, and body fat percentage is estimated, and at least one of the BMI information, weight information, and body fat percentage information is calculated. A learning model for outputting,
The learning model is learned by associating the received light amount data with at least one of the BMI value, body weight, and body fat percentage as training data, and the received light amount data is based on only the DC component of the received light or only the DC component of the received light. The data includes only the high frequency band component, or only the DC component and the high frequency band component.

According to the present disclosure, the information processing system according to the present embodiment generates at least one of the user's BMI information, weight information, and body fat percentage information based on the measurement data via the biometric data generation unit 132. . Thereby, for example, it becomes possible to grasp BMI information, weight information, and body fat percentage information regarding a person to be measured using a wearable device.

FIG. 1 is a block configuration diagram showing an information processing system according to an embodiment of the present disclosure. 2 is a diagram showing the hardware configuration of the management server 100 in FIG. 1. FIG. 3 is a block diagram illustrating functions of a storage unit 120 and a control unit 130 in FIG. 2. FIG. 4 is a schematic diagram showing an example of tag information associated with the measurement data of FIG. 3. FIG. FIG. 2 is a diagram showing the correlation between a BMI value estimated from received light amount data measured by the measuring device 300 of FIG. 1 and an actual BMI value. 2 is a diagram for explaining an example of an electrocardiogram waveform and a pulse waveform measured by the measuring device 300 of FIG. 1. FIG. 2 is a flowchart illustrating an example of processing of the information processing system 1 of FIG. 1. FIG.

The contents of the embodiments of the present invention will be listed and explained. A system according to an embodiment of the present invention has the following configuration.

[Item 1]
An information processing system that receives measurement data from a measurement device worn by a user via a network, and generates biological data from the measurement data, the system comprising:
The measurement device includes at least an LED and a photodiode that emit infrared light, or red, green, or blue light,
At least one of BMI information, weight information, and body fat percentage information is input into a corresponding learning model by inputting received light amount data obtained by irradiating light from the LED onto the user's skin and receiving the light by the photodiode. Equipped with a biometric data generation unit that generates
The received light amount data is data of only the DC component of the received light, only the high frequency band component of the received light, or only the DC component and the high frequency band component,
An information processing system characterized by:
[Item 2]
The high frequency band component includes a frequency band component of 30 Hz or more or 35 Hz or more,
The information processing system according to item 1, characterized in that:
[Item 3]
A server that receives measurement data from a measurement device worn by a user via a network, and generates biological data from the measurement data,
The measurement device includes at least an LED and a photodiode that emit infrared light, or red, green, or blue light,
At least one of BMI information, weight information, and body fat percentage information is input into a corresponding learning model by inputting received light amount data obtained by irradiating light from the LED onto the user's skin and receiving the light by the photodiode. Equipped with a biometric data generation unit that generates
The received light amount data is data of only the DC component of the received light, only the high frequency band component of the received light, or only the DC component and the high frequency band component,
A server characterized by:
[Item 4]
An information processing method that receives measurement data from a measurement device worn by a user via a network, and generates biological data from the measurement data, the method comprising:
The measurement device includes at least an LED and a photodiode that emit infrared light, or red, green, or blue light,
The biometric data generation unit generates BMI information, weight information, and body fat by inputting received light amount data obtained by irradiating light from the LED onto the user's skin and receiving the light with the photodiode into a corresponding learning model. generating at least one of rate information;
The received light amount data is data of only the DC component of the received light, only the high frequency band component of the received light, or only the DC component and the high frequency band component,
An information processing method characterized by:
[Item 5]
A program that causes an information processing system to execute an information processing method for receiving measurement data from a measurement device worn by a user via a network and generating biological data from the measurement data, the program comprising:
The measurement device includes at least an LED and a photodiode that emit infrared light, or red, green, or blue light,
The information processing method includes:
The biometric data generation unit generates BMI information, weight information, and body fat by inputting received light amount data obtained by irradiating light from the LED onto the user's skin and receiving the light with the photodiode into a corresponding learning model. generating at least one of rate information;
The received light amount data is data of only the DC component of the received light, only the high frequency band component of the received light, or only the DC component and the high frequency band component,
A program characterized by:
[Item 6]
Received light amount data obtained by irradiating light onto the user's skin from the LED of a measurement device equipped with an LED and a photodiode that emit at least one of infrared, red, green, and blue light and receiving the light by the photodiode. is input, a learning model for estimating at least one of the user's BMI value, weight, and body fat percentage and outputting at least one of the BMI information, weight information, and body fat percentage information,
The learning model is learned by associating the received light amount data with at least one of a BMI value, body weight, and body fat percentage as teacher data,
The received light amount data is data of only the DC component of the received light, only the high frequency band component of the received light, or only the DC component and the high frequency band component,
A learning model characterized by:

Embodiments of the present disclosure will be described below with reference to the drawings. Note that the embodiments described below do not unduly limit the content of the present disclosure described in the claims. Furthermore, not all components shown in the embodiments are essential components of the present disclosure. Furthermore, features shown in each embodiment can be applied to other embodiments as long as they do not contradict each other.

(Embodiment 1)
<Configuration>
FIG. 1 is a block configuration diagram showing an information processing system 1 according to Embodiment 1 of the present disclosure. For example, this information processing system 1 receives user measurement data from the measurement device 300 via the network NW at the management server 100, and performs predetermined calculations on the measurement data to obtain at least BMI information or weight information. , is a system that generates either body fat percentage information.

The information processing system 1 includes a management server 100, a user terminal device 200, a measurement device 300, and a network NW. Management server 100 and user terminal device 200 are connected via network NW. The network NW includes the Internet, an intranet, a blockchain network, a wireless LAN (Local Area Network), a WAN (Wide Area Network), and the like.

The management server 100 is, for example, a device that receives user measurement data from the measurement device 300 via the network via the user terminal device 200, and performs calculations from the measurement data to biometric data. It is composed of server devices provided.

The user terminal device 200 is an information processing device owned by the user, such as a personal computer, a tablet terminal, a smartphone, a smart watch, a mobile phone, a PHS, or a PDA. It is used to display text, waveform graphs, etc. User information such as the user's identification number, date of birth, gender, height, etc. may be registered in advance in the user terminal device 200, and the user information including the age calculated from the date of birth may be associated with the measurement data. The information may also be transmitted to the management server 100 via the network NW.

Further, the user terminal device 200 may allow the user to input the state of data acquisition by the measuring device 300 using a touch panel or the like. The user terminal device 200 sets the "data acquisition state" to, for example, "running" when running, "when waking up" when waking up, "before eating" when before eating, and Tag information such as "during meal" if the user is eating, "after meal" if the user is after eating, and "at bedtime" if the user is going to bed can be input as tag information. In this case, the user terminal device 200 transmits the measurement data received from the measurement device 300 at a predetermined periodic timing to the management server 100 via the network NW in association with the tag information.

The measuring device 300 is a device that measures the user's measurement data, and this measuring device 300 measures, for example, the user's electrocardiogram, pulse wave, skin temperature (body temperature), acceleration, and angular velocity data using a known method. This is a device for making measurements at periodic timing. The predetermined period may be set in advance, or may be set arbitrarily by the user. More specifically, for example, a temporal period in units of seconds may be set, or a frequency may be similarly set. Note that the user of the user terminal device 200 and the user (person to be measured) of the measuring device 300 may be the same or different.

As an example of a specific configuration of the measurement device 300, each light is irradiated onto the skin from an LED that emits at least either infrared or red light, and the amount of light received by a photodiode (i.e., the amount of reflection, The configuration may be such that received light amount data regarding the intensity of reflected light is acquired as measurement data. Alternatively, it may be configured with a device that acquires pulse waves as pulse waveform data based on changes in the volume of blood vessels caused by the heartbeat of the user based on similar temporal changes in the intensity of the received light, and detection is performed using this method. The pulse waveform that can be used is a photoelectric volume pulse waveform. It may be configured with a device that brings two electrodes into contact with the skin and acquires the electrocardiogram as electrocardiographic waveform data based on the time change of the difference in detected potential, and the electrocardiographic waveform may be data obtained by galvanic skin reaction. good. Alternatively, the device may be configured to acquire the user's skin temperature as data using a temperature sensor that is brought into contact with the user's skin. Alternatively, it may be configured with a three-axis acceleration sensor that detects the variation state of each of the orthogonal XYZ axes, and the measurement device 300 is attached to the user's wrist or arm. In this case, the measurement device 300 acquires acceleration data as an acceleration that is a combination of the swing of the wrist, arm, etc., and the movement of the whole body. Furthermore, it may be configured with a gyro sensor (angular velocity sensor) that detects the rotational angular velocity in each of the orthogonal XYZ axes, and acquires the user's motion as angular velocity data. If so, the measuring device 300 acquires angular velocity data as an angular velocity that is a combination of the rotation of the wrist, arm, etc., and the movement of the whole body.

Bluetooth (registered trademark), Near Field radio Communication (NFC), Afero (registered trademark), Zigbee (registered trademark), Z-Wave (registered trademark) are used between the user terminal device 200 and the measurement device 300. ) or wireless LAN. Note that a wired connection may be used instead of such a wireless connection. Further, the user terminal device 200 and the measurement device 300 may be an integrated device; for example, the measurement device 300 may be equipped with a SIM to provide a communication function so as to be able to directly communicate with the management server 100. Good too.

There is one or more user terminal devices 200, which are connected to the network NW for the number of users using the measurement device 300. There is one or more measuring devices 300, and the measuring devices 300 are connected to the same number of user terminal devices 200 used by one user. When one user uses a plurality of measurement devices 300, the plurality of measurement devices 300 are connected to one user terminal device 200.

<Management server 100>
FIG. 2 is a diagram showing the hardware configuration of the management server 100. FIG. 3 is a block diagram illustrating the functions of the storage section 120 and the control section 130. Note that the illustrated configuration is an example, and other configurations may be used.

The management server 100 includes a communication section 110, a storage section 120, a control section 130, and an input/output section 140. These functional units are realized by executing a predetermined program for the management server 100.

The communication unit 110 is a communication interface for communicating with the user terminal device 200, and communicates according to a communication protocol such as TCP/IP (Transmission Control Protocol/Internet Protocol).

The storage unit 120 stores programs, input data, etc. for executing various control processes and functions within the control unit 130, and is composed of RAM (Random Access Memory), ROM (Read Only Memory), etc. Ru. Further, as shown in FIG. 3, the storage unit 120 includes a measurement data DB 121 that stores measurement data obtained by the measurement device 300 in association with user information, and a measurement data DB 121 that associates biometric data calculated and generated from the measurement data with the user information. A biometric data DB 122 that stores user status information based on measurement data or biometric data, a user status information DB 123 that stores user status information generated based on measurement data or biometric data in association with user information, and a user information DB 124 that stores user information including user identification information. Remember. Further, the user identification information may be account information generated by the data management unit 131, application identification information of an application downloaded to the user terminal device 200, or user terminal identification information of the user terminal device. good. Furthermore, the storage unit 120 temporarily stores data communicated with the user terminal device 200. Note that the data structure of the DB is not limited to this, and a part of the above-mentioned DB may be stored in the user terminal device 200 or the measurement device 300.

The control unit 130 controls the overall operation of the management server 100, and is composed of a CPU (Central Processing Unit) and the like. Further, as shown in FIG. 3, the control unit 130 includes functional units such as a data management unit 131, a biometric data generation unit 132, a user status information generation unit 133, and a data output unit 134.

The data management unit 131 sets user identification information for each user who uses the measuring device 300. For example, if the user identification information is account information, the account information is generated when the user who uses the measurement device 300 registers the account information with the user terminal device 200. Therefore, the data management unit 131 controls whether the user terminal device 200 of the user can access various DBs in the storage unit 120 for each account. The data management unit 131 stores various data such as measurement data, biological data, user status information, etc. in a corresponding DB in association with user information including user identification information. Further, at this time, the data management unit 131 may associate the measurement data with predetermined tag information and store the associated measurement data.

FIG. 4 is a schematic diagram showing an example of tag information associated with the measurement data of FIG. 3. Data D1 shown in FIG. 4 is measurement data of the measurement device 300. The tag T1 is tag information associated with the data D1, and for example, the time information when the measuring device 300 measured the data D1, or the time information when the data D1 was transmitted from the measuring device 300 to the user terminal device 200 is chronologically Stored as data. Alternatively, both the measured time information and the transmitted time information may be associated. For example, in the first line of the tag T1 shown in FIG. 4, "20180620120746144" is stored, which indicates June 20, 2018, 12:07:46 seconds and 144 milliseconds. Such time information can be obtained from communication logs. This makes it possible to understand which time period the measurement data is from.

Note that such association of measurement data using tag information is not limited to time information, and may also be performed by having the user freely write in physical information or behavioral information indicating the user's physical condition or behavior and storing it as tag information. Okay, let the user select from predetermined options (for example, in response to the question "How is your current physical condition?", let the user select one of "1: good, 2: average, 3: bad") etc.), and the selected answer may be stored. Furthermore, based on other measurement data and biological data (e.g., walking speed information, stride length information, movement information of the wearing part, posture information, center of gravity position information, heart rate information, etc.), predetermined actions (e.g., sitting, standing, etc.) , lying down, walking, running, sleeping, waking up, going to bed, eating, driving, resting, etc.) may be estimated by a known method and associated with biological data etc. as tag information (behavior type information). At this time, for example, the learning model may be estimated using a learning model learned based on the teacher measurement data, or the learning model may be personalized by performing additional learning based on the results recorded by the user described above. As a result, when the control unit 150 generates biometric data such as BMI information, weight information, and body fat percentage information, more accurate biometric data is generated by associating the tag information with the biometric data. At the same time, the user status information generated based on this can also become more personalized data.

The biometric data generation unit 132 performs predetermined calculations on the measurement data stored in the measurement data DB 121 to generate biometric data. In the generation of biometric data, new biometric data may be generated based on one or more previously generated biometric data that is not limited to data that is directly generated by calculating measurement data. The biological data includes, for example, the user's BMI information, weight information, body fat percentage information, blood pressure information, heart rate information, blood oxygen amount information, maximum oxygen intake information, electrocardiogram information, breathing rate, body temperature information, blood sugar level information, Data such as step count information, stride length information, center of gravity position information, posture information, activity type information, stress information, amount of exercise information, exercise load information, travel distance information, walking speed information, amount of activity information, information on movement of hands or legs, etc. Although it is calculated from measurement data using a known method, it is not limited thereto. The biometric data generated by the calculation is stored in the biometric data DB 122. Note that the biometric data generation section 132 is not limited to being provided in the control section of the management server, but may be provided in the control section within the measuring device 300 or the control section within the user terminal device 200.

In addition, using a known learning device, for example, measurement data, biological data generated based on the measurement data (such as BMI information, weight information, body fat percentage information, heart rate information, blood pressure information, etc.) and positive biological data ( For example, the correspondence between BMI information, weight information, body fat percentage information, heart rate information, blood pressure information, etc. based on known measuring devices, medical equipment, calculation formulas, etc. (for example, information indicating the degree and range of error, etc.) The biometric data generating unit 132 creates a learning model in advance based on the teacher data associated with the learning model (which may be included), and generates the biometric data by using the above-mentioned predetermined calculation (analysis) as a determination using the learning model. may be generated.

Here, each light is irradiated onto the skin from an LED that emits at least either infrared or red light, and the amount of light received by the photodiode (i.e., the amount of reflection, the intensity of the reflected light) is measured data. A method of generating at least one of BMI information, weight information, and body fat percentage information, which are biological data, from received light amount data regarding the subject will be exemplified. In FIG. 5, a BMI estimation learning model is created based on training data that correlates the received light amount data measured by the measuring device 300 attached to the wrist of the subject and the actual BMI value of the subject. The estimated BMI value (BMI information) can be calculated by inputting the received light amount data separately measured by the measuring device 300 into the BMI estimation learning model. This is a diagram showing the correlation when the vertical axis is the actual BMI value of the subject corresponding to the received light amount data and the horizontal axis is the actual BMI value of the subject (approximately 400 subjects). The received light amount data here may be data obtained by extracting only the DC component from among the components of the received light. The DC component is a component that is absorbed in a substantially constant amount by non-pulsating blood vessels, skin, tissues, etc. In contrast, the AC component is a variable amount of light absorbed by arteries and the like. These components are commonly used components and are extracted by known methods. In addition, instead of or in addition to the DC component, the received light amount data is a high frequency component (for example, a component in a frequency band of 30 Hz or more, more preferably a component in a frequency band of 35 Hz or more). Good too. In this way, the applicant has discovered that there is a correlation between received light amount data (in particular, individual differences in absorption rate shown in the DC component and differences in the expression of high frequency components) and BMI values. By creating a BMI estimation learning model based on training data in which BMI values are correlated with each other, it is possible to generate BMI information using only the received light amount data acquired from the measuring device 300 (for example, a wearable device). In addition, although the received light amount data is exemplified using an LED that emits either infrared or red light, it is also possible to use a BMI estimation learning model that is trained based on data from both LEDs that emit infrared and red light. Therefore, the accuracy of estimating the BMI value is further improved. Note that any light emitter (especially an LED) that emits light in a wavelength band that can capture pulse waves may be used. For example, a light emitter that emits green or blue light (especially an LED) may be used. You can. Furthermore, considering that it is known that the BMI value is generally calculated by dividing the weight in kg by the square of the height in m, the weight can be calculated from the estimated BMI information and the user's height information registered in advance. It is also possible to generate information. In addition, a weight estimation learning model is created based on teacher data in which received light amount data and weight values are associated with each other, and is stored in a storage unit inside or outside the management server. Estimated weight information may be generated by inputting the received light amount data into a weight estimation learning model. Furthermore, a body fat percentage estimation learning model is created based on the teacher data in which the received light amount data and the body fat percentage are associated with each other, and is stored in a storage unit inside or outside the management server, and is separately installed in the measuring device 300. Estimated body fat percentage information may be generated by inputting the received light amount data measured in the body fat percentage estimation learning model. In the above example, the measurement device 300 is worn on the wrist, but the measurement device 300 is not limited to this, and can be worn on any part of the person to be measured by appropriately adjusting the amount of light emitted. For example, it may be the fingers, arms, neck, head, ears, temples, head, chest, torso, waist, back, abdomen, legs, ankles, soles, etc. In addition, although the user is a human in the above example, the user (target to be measured) is not limited to this, but by appropriately adjusting the amount of light emitted, it is possible to etc.).

Furthermore, a method for calculating systolic blood pressure and diastolic blood pressure, which are biological data, from measurement data will be exemplified. FIG. 6 is a diagram for explaining an example of an electrocardiogram waveform and a pulse wave measured by the measuring device 300 in FIG. The app shows a photoelectric volume pulse waveform, a velocity pulse waveform obtained by the first differentiation of the photoelectric volume pulse waveform with respect to time, and an acceleration pulse waveform obtained by the second order differentiation of the photoelectric volume pulse waveform with respect to time. FIG. 6 shows, from top to bottom, an electrocardiographic waveform, a photoelectric volume pulse waveform, a velocity pulse waveform, and an acceleration pulse waveform. The vertical axis indicates the intensity of each waveform, and the electrocardiographic waveform and photoelectric plethysmographic waveform are expressed in mV, which indicates the electric potential. The horizontal axis shows the passage of time, and the passage of time is shown from left to right.

An electrocardiogram waveform is a waveform that shows periodic changes in the electrical signals that cause a person's heart to beat. The electrocardiographic waveform is assigned names such as P wave, Q wave, R wave, S wave, and T wave to the inflection points of its shape, and indicates one cycle of heartbeat. The P wave represents atrial contraction, the Q wave, the R wave, and the S wave represent the state of ventricular contraction, and the T wave represents the start of ventricular expansion.

The photoelectric volume pulse waveform is a waveform that shows changes in blood pressure and volume within the peripheral vascular system as the human heart beats. In the photoelectric volume pulse waveform, names of A wave, P wave, V wave, and D wave are assigned to the inflection points of the shape, respectively, indicating one cycle of a heartbeat. Using the A wave as the reference point at the time when the arterial pulse wave occurs, the P wave is a percussion wave (shock wave) generated by left ventricular ejection, the V wave is a Valley wave (wave due to overlapping bulge) generated when the aortic valve closes, and the D wave. indicates a dicrotic wave (overlapping wave) which is a reflected vibration wave.

The velocity pulse waveform is obtained by first-order differentiation of the photoelectric volume pulse waveform with respect to time. The acceleration pulse waveform is the first-order differential of the velocity pulse waveform with respect to time, that is, the second-order differential of the photoelectric volume pulse waveform. As shown in Fig. 6, the accelerated pulse waveform includes wave a (positive wave in early systole), wave b (negative wave in early systole), wave c (re-rising wave in mid-systole), and wave d (systolic wave) at each peak of the waveform. The names are assigned as follows: late re-downward wave), e wave (diastolic early positive wave), and f wave (diastolic early negative wave).

The ratio of the intensity of the b-wave to the intensity of the a-wave, and the ratio of the intensity of the f-wave to the intensity of the e-wave are parameters indicating the stretchability or elasticity of the blood vessel. The main components of blood vessels are vascular endothelium, elastic fibers, proteins, and smooth muscles. Each of these components has different properties, and Collagen and Elastin have a strong influence on the elasticity of blood vessels during systolic blood pressure and systolic blood pressure, respectively. Therefore, the elasticity that varies depending on the blood pressure value can be expressed by the parameters of (b/a), which is the ratio of the intensity of the b wave to the intensity of the a wave, and (f/e), which is the ratio of the intensity of the f wave to the intensity of the e wave. These values also vary depending on age, gender, and environmental variables. Therefore, the values of (b/a) and (f/e) can be calculated as characteristic information of the accelerated pulse waveform.

As shown in FIG. 6, the time equal to the difference between the time Tr when the R wave occurs and the time Tp when the P wave occurs becomes the ventricular systolic pulse wave propagation time PTT_SYS. The difference between the time Tt when the T wave occurs and the time Td when the D wave occurs becomes the ventricular diastolic pulse wave propagation time PTT_DIA. That is, from the R-wave time Tr and T-wave time Tt of the electrocardiogram waveform, the T-wave time Tp and the D-wave time Td of the photoelectric pulse waveform, as shown in equations (1) and (2), Then, the ventricular systolic pulse wave propagation time PTT_SYS and the ventricular diastolic pulse wave propagation time PTT_DIA can be calculated.

PTT_SYS=Tp-Tr...(1)

PTT_DIA=Td-Tt...(2)

The first and second electrodes that measure the electrocardiogram waveform and the optical sensor module that measures the photoelectric volume pulse waveform are placed opposite to each other across the wrist, and by separating the placement distance, the electrocardiogram waveform detection site and the photoelectric sensor module are placed facing each other across the wrist. This means that the measurement site of the volume pulse waveform is separated. Therefore, by creating a time lag between the respective characteristic waveforms, it is possible to lengthen the absolute calculation time of the ventricular systolic pulse wave propagation time PTT_SYS and the ventricular diastolic pulse wave propagation time PTT_DIA. Therefore, when obtaining change information on the ventricular systolic pulse wave propagation time PTT_SYS and the ventricular diastolic pulse wave propagation time PTT_DIA, the accuracy of the change information can be improved.

Here, the formula for calculating blood pressure will be explained.

The pulse wave propagation velocity equation (Moens-Korteweg equation) shown below (3) shows the relationship between the pulse wave propagation velocity and the longitudinal elastic modulus of the arterial wall.

L/T_PTT=√(E・h/(2・r・ρ)) ...(3)

The parameters in equation (3) are: L: distance between measurements, T_PTT: pulse wave propagation time, r: blood vessel inner diameter, E: longitudinal elastic modulus of blood vessel, h: blood vessel thickness, and ρ: blood density.

It is known that there is a correlation between longitudinal elastic modulus and blood pressure values.

E=E ₀・exp(α・P)...(4)

It can be shown as Here, P: blood pressure value, α: constant, E ₀ : initial value.

From equations (3) and (4),

P=(-2・ln(T_PTT)+ln(2・r・ρ・L ² /(E ₀・h)))/α...(5)

can be derived. ln indicates natural logarithm. At this time, since "r·ρ" is proportional to the blood volume at the measurement site, it can be indicated by the high value (Vp, Vd) shown by the photoelectric plethysmogram. Moreover, since "E ₀ · h" is a value proportional to the elasticity of the blood vessel, it can be replaced using (b/a) and (f/e), which are parameters indicating elasticity.

Therefore, the systolic blood pressure BP_SYS (Blood Pressure_Systolic) and the diastolic blood pressure BP_DIA (Blood Pressure_Diastolic) can be expressed by the following equations (6) and (7).

BP_SYS=A1・ln(PTT_SYS)+A2・ln(Vp)+A3・ln(b/a)+A4...(6)

BP_DIA=A5・ln(PTT_DIA)+A6・ln(Vd)+A7・ln(f/e)+A8...(7)

A1 to A8 are constants determined by conditions. The systolic blood pressure BP_SYS, which can be calculated using equation (6), is the natural logarithm of the ventricular systolic pulse wave propagation time PTT_SYS multiplied by a constant A1, and the natural logarithm of the P wave intensity Vp multiplied by a constant A2. It can be determined by the sum of the natural logarithm of (b/a) multiplied by the constant A3 and the constant A4. The diastolic blood pressure BP_SYS, which can be calculated using equation (7), is the natural logarithm of the ventricular systolic pulse wave propagation time PTT_DIA multiplied by a constant A5, and the natural logarithm of the D wave intensity Vd multiplied by a constant A6. It can be determined by the sum of the natural logarithm of (f/e) multiplied by the constant A7 and the constant A8. Systolic blood pressure BP_SYS and diastolic blood pressure BP_DIA can be determined by determining each constant based on the characteristics of the device, the person to be measured, and the like. However, when checking the change status of systolic blood pressure BP_SYS and diastolic blood pressure BP_DIA, it is not necessary to finalize all the constants, and while using provisional values instead, information on systolic blood pressure BP_SYS and diastolic blood pressure BP_DIA can be used as information. It is possible to obtain the value. The natural logarithm of the P wave intensity Vp and the natural logarithm of the D wave intensity Vd are terms that take into account the influence of blood density. Further, the natural logarithm of (b/a) and the natural logarithm of (f/e) are terms that take into consideration the influence of the longitudinal elastic modulus of the artery wall. Therefore, depending on the measurement conditions, information regarding the systolic blood pressure BP_SYS and information regarding the diastolic blood pressure BP_DIA may be calculated by selecting one of the terms and making the other terms constant.

Furthermore, the biometric data generation unit 132 can obtain temperature information as biometric data from skin temperature information of the subject measured by a temperature sensor (thermistor, etc.) of the measuring device 300 worn by the subject, for example. .

Furthermore, a method for calculating walking speed information, which is biological data, from measurement data will be exemplified. For example, known calculation methods may be used alone or in combination (for example, averaging, weighting, etc.) from the waveform data of acceleration data measured by the measurement device 300 worn on the user's wrist. Walking speed information can be obtained, for example, by integrating acceleration data at predetermined time intervals. In addition, the user terminal device 200 or the measurement device 300 is equipped with a GPS function, and more accurate walking speed information can be obtained by referring to walking speed information calculated from GPS location information and time information related to the location information. You may also obtain However, GPS accuracy generally decreases, especially when the location is indoors or underground, or when the travel distance is relatively short (for example, within a few meters). It may be possible to select whether to use the information in combination (or to use it in place of the calculation by the acceleration sensor), or the user terminal device 200 may use the GPS reception sensitivity information or the distance calculated by the GPS or the acceleration sensor. Based on the information, it may be possible to select whether to use the information from the GPS. By calculating walking speed information in this way, it is possible to easily (especially on a daily basis) measure ``walking speed'', which is one of the known indicators for measuring human muscle mass, which is necessary to understand for frailty prevention etc. Is possible. Additionally, smartphones and mobile phones have acceleration sensors, but since these devices are difficult to identify where the body is moving, measurement devices such as wearable devices that are worn on the user's wrist, etc. 300, it is possible to easily track behavioral information.

Furthermore, a method for calculating step length information, which is biological data, from measurement data will be exemplified. For example, known calculation methods may be used alone or in combination (for example, averaging, weighting, etc.) from the waveform data of acceleration data measured by the measurement device 300 worn on the user's wrist. For example, since we wave our hands like a pendulum when we walk, we can obtain step length information using Since the interval between steps can be determined based on the timing at which the acceleration component in the direction perpendicular to the direction is the smallest or the timing at which the vertical direction changes, etc., step length information can be obtained by using time information. Alternatively, for example, when kicking off the ground, acceleration components in the kicking direction are combined, so it is also possible to determine the interval between one step based on the generation timing of the acceleration component in the direction. By calculating stride length information in this way, it is possible to accurately grasp the timing at which stride length begins to shorten and the degree of variation in stride length, which is useful for estimating the degree of progression of dementia. Furthermore, the decline in the sense of balance may be calculated by calculating the amount of left and right sway (for example, sway width, sway speed, etc.) during walking from the above-mentioned acceleration data.

Furthermore, the amount of exercise information calculated based on the distance information calculated by the above-mentioned GPS and acceleration sensor, the above-mentioned stride length information and step count information (for example, based on the information from the acceleration sensor, which can be acquired by a known method), etc. It may be calculated as data.

In addition, the biological data generation unit 132 determines the location where the measurement device 300 is worn (for example, the wrist, ankle, etc.) from the acceleration data and angular velocity data measured by the measurement device 300 worn by the subject. Motion information such as how fast and at what angle the animal is moving can be obtained as biometric data.

In addition, the biological data generation unit 132 performs frequency analysis on the acceleration data measured by the measuring device 300 worn by the subject, for example, and associates the high and low frequencies with the high and low activity frequencies, and Activity amount information can be obtained as biometric data by calculating based on predetermined conditions such as what percentage of the day the activity occupies.

Furthermore, the biological data generation unit 132 can identify acceleration data during exercise, including walking, using a known calculation method, for example, from acceleration data measured by the measuring device 300 worn by the subject. The amount of exercise information can be obtained as biological data by calculating it under predetermined conditions using, for example, frequency analysis. Further, by further using additional information such as angular velocity information, it is possible to obtain more accurate momentum information.

The biological data generation unit 132 also generates, for example, heart rate information that increases with exercise load, for example, with respect to the activity amount information and the exercise amount information derived from acceleration data measured by the measuring device 300 worn by the subject. By weighting the information, exercise load amount information can be obtained as biological data. In addition, for example, if vector information of acceleration data is taken into consideration, state information such as walking environment (slope, stairs, etc.) and posture (standing position, sitting position, etc.) can also be specified, and thus the state information may be further used. Further, by further using additional information such as angular velocity information, it is possible to obtain more accurate exercise load amount information.

Furthermore, the biological data generation unit 132 uses the heart rate information to generate maximum oxygen uptake information using a known formula such as VO ₂ max = 15 x (220 - age) ÷ heart rate (particularly resting heart rate). It can be obtained as biological data.

The user status information generation unit 133 generates information on the comparison result of each biometric data or the overall The comparison results may be combined to evaluate the user condition (for example, health condition evaluation, diet progress evaluation, etc.), and evaluation information according to the evaluation may be generated as user condition information. The evaluation information may be, for example, information indicating which reference biometric data is exceeded (or not exceeded), or which group is set with the reference biometric data as the boundary. It may also be information indicating whether it is applicable (for example, evaluation groups of A to F, graded evaluation classification groups such as healthy, reserve, and caution group).

The data output unit 134 transmits measurement data, biological data, user status information, etc. to the user terminal device 200 (here, the user may be the person to be measured, or to other different persons (for example, close relatives, doctors, caregivers). It may be a researcher, researcher, etc.)). In the user terminal device 200, the output data may be displayed on the screen via a dedicated application, for example, so that the user can easily check the output data. In addition, when setting multiple reference biometric data and creating multiple groups as described above, if the group falls under the predetermined reference biometric data or higher, a predetermined notification (for example, BMI value, weight, Send (e.g., select from various notifications in a notification information DB (not shown) according to predetermined conditions The contents of the notification may be changed depending on which reference biometric data has been exceeded (for example, the higher the reference value exceeded, the more the notation, color, text, etc.) (e.g., the display becomes a highlighted notification). Alternatively, for example, using tag information, extract at least one of the biometric data when waking up, the biometric data before meals, the biometric data after eating, and the biometric data when going to bed, and extract biometric data related to one or more days (for example, If the number of days is two or more, statistical values such as the transition of the extracted biometric data) or the average value or median value of the extracted biometric data for each predetermined period may be output.

The input/output unit 140 is information input devices such as a keyboard and mouse, and output devices such as a display.

<Processing flow>
The flow of processing of the information processing method executed by the information processing system 1 will be described with reference to FIG. 7. FIG. 7 is a flowchart illustrating an example of processing of the information processing system 1 of FIG. 1.

As processing in step S101, the data management unit 131 generates account information for each user who uses the measurement device 300, and acquires predetermined user information from the user terminal device 200 and the like. The registered user information is stored in the user information DB 124 by the data management unit 131. The process in step S101 may be performed as pre-processing before the user uses the measuring device 300, or may be performed when the user uses the measuring device 300 for the first time.

As the process in step S102, when the user uses the measuring device 300, measurement data is transmitted from the measuring device 300 to the management server 100 via the user terminal device 200, and is received via the communication unit 110. The data management unit 131 stores measurement data in association with user information in the measurement data DB 121 of the storage unit 120.

As the process of step S103, the biometric data generation unit 132 reads the measurement data, and generates biometric data by a predetermined calculation or the like. The generated biometric data is stored in the biometric data DB 122 by the data management unit 131.

As the process in step S104, the user status information generation unit 133 reads measurement data or biometric data and generates user status information. The generated user status information is stored in the user status information DB 123 by the data management unit 131.

As the process in step S<b>105 , the data output unit 134 reads at least one of measurement data, biological data, and user status information, and outputs it to the user terminal device 200 .

<Effect>
As described above, the information processing system according to the present embodiment generates, in particular, at least one of the user's BMI information, weight information, and body fat percentage information via the biometric data generation unit 132 based on the measurement data. Thereby, for example, it becomes possible to grasp BMI information, weight information, and body fat percentage information regarding a person to be measured using a wearable device.

Although the disclosed embodiments have been described above, they can be implemented in various other forms, and can be implemented with various omissions, substitutions, and changes. These embodiments and modifications, as well as omissions, substitutions, and changes, are included within the technical scope of the claims and their equivalents.

1 Information processing system 100 Management server 200 User terminal device 300 Measuring device NW network

Claims (6)

An information processing system that receives measurement data from a measurement device worn by a user via a network, and generates biological data from the measurement data, the system comprising:
The measurement device includes at least an LED and a photodiode that emit infrared light, or red, green, or blue light,
At least one of BMI information, weight information, and body fat percentage information is input into a corresponding learning model by inputting received light amount data obtained by irradiating light from the LED onto the user's skin and receiving the light by the photodiode. Equipped with a biometric data generation unit that generates
The received light amount data is data of only the DC component of the received light, only the high frequency band component of the received light, or only the DC component and the high frequency band component,
An information processing system characterized by: The high frequency band component includes a frequency band component of 30 Hz or more or 35 Hz or more,
The information processing system according to claim 1, characterized in that: A server that receives measurement data from a measurement device worn by a user via a network, and generates biological data from the measurement data,
The measurement device includes at least an LED and a photodiode that emit infrared light, or red, green, or blue light,
At least one of BMI information, weight information, and body fat percentage information is input into a corresponding learning model by inputting received light amount data obtained by irradiating light from the LED onto the user's skin and receiving the light by the photodiode. Equipped with a biometric data generation unit that generates
The received light amount data is data of only the DC component of the received light, only the high frequency band component of the received light, or only the DC component and the high frequency band component,
A server characterized by: An information processing method that receives measurement data from a measurement device worn by a user via a network, and generates biological data from the measurement data, the method comprising:
The measurement device includes at least an LED and a photodiode that emit infrared light, or red, green, or blue light,
The biometric data generation unit generates BMI information, weight information, and body fat by inputting received light amount data obtained by irradiating light from the LED onto the user's skin and receiving the light with the photodiode into a corresponding learning model. generating at least one of rate information;
The received light amount data is data of only the DC component of the received light, only the high frequency band component of the received light, or only the DC component and the high frequency band component,
An information processing method characterized by: A program that causes an information processing system to execute an information processing method for receiving measurement data from a measurement device worn by a user via a network and generating biological data from the measurement data, the program comprising:
The measurement device includes at least an LED and a photodiode that emit infrared light, or red, green, or blue light,
The information processing method includes:
The biometric data generation unit generates BMI information, weight information, and body fat by inputting received light amount data obtained by irradiating light from the LED onto the user's skin and receiving the light with the photodiode into a corresponding learning model. generating at least one of rate information;
The received light amount data is data of only the DC component of the received light, only the high frequency band component of the received light, or only the DC component and the high frequency band component,
A program characterized by: Received light amount data obtained by irradiating light onto the user's skin from the LED of a measurement device equipped with an LED and a photodiode that emit at least one of infrared, red, green, and blue light and receiving the light by the photodiode. is input, a learning model for estimating at least one of the user's BMI value, weight, and body fat percentage and outputting at least one of the BMI information, weight information, and body fat percentage information,
The learning model is trained as teacher data by associating the received light amount data with at least one of a BMI value, body weight, and body fat percentage,
The received light amount data is data of only the DC component of the received light, only the high frequency band component of the received light, or only the DC component and the high frequency band component,
A learning model characterized by: