Health assessment system, health assessment program, and health assessment server

Abstract

To provide a system, a health degree determination program, and a health degree determination server capable of easily determining health degree.SOLUTION: A health degree determination system 520 includes: a bioelectrical impedance acquisition unit 53 for acquiring bioelectrical impedance of a plurality of body parts of a user; a body composition estimation unit 54 for estimating a body composition of a plurality of items of the user from the bioelectrical impedance of the plurality of body parts; and a health degree determination unit 56' for determining health degree of the user based at least in part, indirectly or directly on the body composition of the plurality of items.SELECTED DRAWING: Figure 8

Description

[Technical Field]

[0001] The present invention relates to a health assessment system for assessing a user's health, a health assessment program, and a health assessment server. [Background technology]

[0002] Generally, in a health checkup, in addition to height, weight, chest circumference, etc., measurements are taken of chest X-ray, urinary protein, urinary occult blood, GOT, GPT, LDL-C, HDL-C, TG, uric acid level, HbA1c, FPG, hematocrit, RBC, CRP, systolic blood pressure, diastolic blood pressure, etc. Based on the results of these health checkups, a doctor may give advice on the subject's health level or health risk, or a method has been proposed in which the subject's health level or health risk is determined using a predetermined calculation formula from the results of the health checkup (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2002-063278 Summary of the Invention [Problem to be solved by the invention]

[0004] However, health checkups require subjects to visit a medical institution for testing, and are generally conducted only once a year. It is advantageous to make those living a lifestyle that increases health risks (decreases health) aware of those risks in a shorter period of time, while it is also advantageous to show those living a lifestyle that reduces health risks (improves health) the results in a shorter period of time.

[0005] Therefore, an object of the present invention is to provide a system that can easily determine health status.

[0006] Since health level and health risk are concepts that are in an anti-correlated relationship, assessing health level means assessing health risk. In this application, the present invention and its embodiments are described using the concept of health level, but the present invention also falls within the scope of the present invention when health level is replaced with health risk. Health risk includes the possibility of contracting a disease. [Means for solving the problem]

[0007] One aspect of the health assessment system of the present invention is configured to include a bioelectrical impedance acquisition unit that acquires the bioelectrical impedance of one or more body parts of a user, and a health assessment unit that assesses the health of the user based, at least indirectly or directly, on the bioelectrical impedance of the one or more body parts.

[0008] Body composition scales, which are commonly used in ordinary households, measure the bioelectrical impedance of one or more parts of the user's body, so with this configuration, it is possible to easily determine the level of health using a body composition scale that is in the home, for example.

[0009] The above-mentioned health assessment system may further include a body composition estimation unit that estimates multiple items of the user's body composition from the bioelectrical impedance of the one or more body parts, and the health assessment unit may assess the user's health based, at least indirectly or directly, on the multiple items of body composition.

[0010] The above-mentioned health assessment system may further include a health check estimation unit that estimates the health check results of the user for multiple items from the body composition for the multiple items, and the health assessment unit may assess the health of the user based at least in part on the health check results for the multiple items.

[0011] In the above-described health assessment system, each of the plurality of health checkup results may have a significant correlation with the plurality of body composition items.

[0012] In the above health level determination system, the health level determination unit may determine whether the subject is healthy, unhealthy, or uncertain.

[0013] The above-mentioned health assessment system may further include an additional information acquisition unit that requests the input of additional information that was not used to assess the health level when the result of the assessment is uncertain, and the health assessment unit may further assess the health level using the additional information.

[0014] The above-mentioned health assessment system may further include an output unit that outputs the results of the health assessment, and the output unit may display the results of the assessment by pointing out the results on a clock-like graphic.

[0015] In the above-described health assessment system, the output unit may indicate the predicted future assessment result on the graphic.

[0016] A health assessment program of one embodiment of the present invention is configured to, when executed by a processor, cause the processor to function as a health assessment unit that assesses the health of the user based, at least in part, indirectly or directly, on bioelectrical impedance obtained from one or more parts of the user's body.

[0017] A health assessment server according to one embodiment of the present invention is configured to include a health assessment unit that assesses the health of a user based, at least in part, indirectly or directly, on bioelectrical impedance obtained from one or more body parts of the user. [Brief explanation of the drawings]

[0018] [Figure 1] FIG. 1 is a schematic diagram showing the external configuration of a health assessment system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a diagram showing a usage mode of the measurement device according to the first embodiment of the present invention. [Figure 3]FIG. 3 is a block diagram showing the hardware configuration of the health information providing system according to the first embodiment of the present invention. [Figure 4] FIG. 4 is a block diagram showing the functional configuration of the health information providing system according to the first embodiment of the present invention. [Figure 5] FIG. 5 is a graph showing the correlation between HDL cholesterol (vertical axis), which is a biochemical test value, and fat mass (kg) (horizontal axis), which is body composition information, according to the first embodiment of the present invention. [Figure 6] FIG. 6 is a graph showing the correlation between HDL cholesterol (vertical axis) and visceral fat mass (cm 2 ) (horizontal axis), which is body composition information, according to the first embodiment of the present invention. [Figure 7] FIG. 7 is a graph showing the accuracy of estimation using multiple types of body composition information according to the first embodiment of the present invention. [Figure 8] FIG. 8 is a block diagram showing the functional configuration of a health level determination system 520 according to the second embodiment of the present invention. [Figure 9] FIG. 9 is a graph showing the accuracy of health level determination using multiple items of body composition information according to the first and second embodiments of the present invention. [Figure 10A] FIG. 10A is a graph for explaining the accuracy of determining a health level using multiple items of body composition information according to the first and second embodiments of the present invention. [Figure 10B] FIG. 10B is a graph for explaining the accuracy of determining a health level using multiple items of body composition information according to the first and second embodiments of the present invention. [Figure 11] FIG. 11 is a diagram showing an example of a health advice screen according to an embodiment of the present invention. [Figure 12] FIG. 12 is a diagram showing another example of the health advice screen according to the embodiment of the present invention. [Figure 13] FIG. 13 is a block diagram showing the functional configuration of a health assessment system according to the third embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0019] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Note that the embodiment described below shows an example of how the present invention can be implemented, and the present invention is not limited to the specific configuration described below. When implementing the present invention, a specific configuration corresponding to the embodiment may be appropriately adopted.

[0020] The health estimation system of the embodiment described below measures the bioelectrical impedance of one or more body parts of a user and determines the health of the user based, at least indirectly or directly, on the measured bioelectrical impedance of the one or more body parts. Methods for determining health indirectly based on bioelectrical impedance include a method for estimating multiple body composition items from the bioelectrical impedance of one or more body parts and determining the health based on the estimated multiple body composition items, and a method for estimating multiple body composition items from the bioelectrical impedance of one or more body parts, further estimating the results of health checkups for the multiple items from the estimated multiple body composition items, and determining the health based on the estimated multiple health checkup results.

[0021] Whether the method of determining health level is one that estimates health checkup results or one that determines health level without estimating health checkup results, it is desirable to determine health level based on multiple items of body composition (multiple items may be different types of body composition, such as fat mass and muscle mass, or multiple items may be different body composition parts, such as the whole body and the right leg), but it may also be based on a single item of body composition and at least one item of the user's biometric information (height, age, sex, weight) described below.

[0022] In the following, a method of estimating body composition from bioelectrical impedance, further estimating health check results from the body composition, and determining health level from the health check results will be described as a first embodiment, a method of estimating body composition from bioelectrical impedance and determining health level from the body composition without estimating health check results will be described as a second embodiment, and a method of determining health level directly from bioelectrical impedance without estimating body composition and health check results will be described as a third embodiment.

[0023] In any of the embodiments, health risks are determined from in vivo information (bioelectrical impedance) that reveals continuous lifestyle habits, and therefore, compared to determining health risks through health checkups that are burdensome for users and medical institutions and cannot be performed frequently, or determining health risks through genetic testing services that can test by simply sending saliva or the like but can only evaluate from a congenital perspective, this embodiment places less of a burden on users and makes it possible to determine health risks in the course of daily life while also taking lifestyle habits into consideration. Not only is this embodiment less burdensome, but it is also expected that health risks can be determined more accurately by using this embodiment, which allows daily measurements and tracking changes, than by determining health risks solely through health checkups that are only taken about once a year and which may not necessarily represent daily conditions.

[0024] (First embodiment) FIG. 1 is a diagram showing a health assessment system according to a first embodiment of the present invention. FIG. 2 is a diagram showing how the measurement device according to the first embodiment of the present invention is used. In this embodiment, health assessment system 510 comprises measurement device 10 and information processing terminal (hereinafter referred to as "user terminal") 20. Measurement device 10 is a body composition analyzer that applies the measurement principle known as bioelectrical impedance analysis (BIA) to measure the potential difference caused by a weak current flowing inside the body, thereby measuring body composition such as body fat.

[0025] The measuring device 10 comprises a main body 11 and a handle unit 12. The main body 11 and the handle unit 12 are electrically connected by a connection cord 13. The handle unit 12 can be housed in a housing portion 14 provided in the main body 11. When the handle unit 12 is housed in the housing portion 14, the connection cord 13 is wound up by a winding mechanism (not shown) inside the main body 11 and housed inside the main body 11.

[0026] The main body 11 is provided with a current-carrying electrode 111R and a measurement electrode 112R for the right foot on the right side of the upper surface, and a current-carrying electrode 111L and a measurement electrode 112L for the left foot on the left side of the upper surface.

[0027] The handle unit 12 has a generally rod-like shape and includes a handle body 15 at its center, with grips 16R and 16L provided on either side of the handle body 15. A display panel 17 and operation buttons 18A to 18D are provided on the handle body 15. Furthermore, the grip 16R includes a current-carrying electrode 161R and a measurement electrode 162R for the right hand, and the grip 16L includes a current-carrying electrode 161L and a measurement electrode 162L for the left hand.

[0028] 2, a user can measure body composition by standing barefoot on main body 11, stretching their arms, and gripping handle unit 12 with both hands. At this time, the base of the right toes contacts current-carrying electrode 111R, the heel of the right foot contacts measurement electrode 112R, the base of the left toes contacts current-carrying electrode 111L, the heel of the left foot contacts measurement electrode 112L, the fingers of the right hand contacts current-carrying electrode 161R, the palm of the right hand contacts measurement electrode 162R, the fingers of the left hand contacts current-carrying electrode 161L, and the palm of the left hand contacts measurement electrode 162L.

[0029] The main body 11 also has a load cell therein for measuring body weight. As shown in FIG. 2, the weight of a user standing on the main body 11 can be measured. The load cell is composed of a strain element made of a metal member that deforms in response to a load, and a strain gauge attached to the strain element. When a user stands on the measuring device 10, the strain element of the load cell bends due to the user's load, causing the strain gauge to expand and contract. The resistance value (output value) of the strain gauge changes in response to this expansion and contraction. The measuring device 10 calculates the body weight from the difference between the output value of the load cell when no load is applied (zero point) and the output value when a load is applied. The configuration for measuring body weight using a load cell may be the same as that of a general weighing scale.

[0030] The user terminal 20 is a portable terminal equipped with a processor capable of executing application programs, memory for temporarily storing data required for processor processing, internal storage such as flash memory for storing application programs and generated data, a touch panel, various interfaces, etc. The user terminal 20 also includes a wireless communication device for connecting to the Internet and a short-range communication device for connecting to other nearby devices. The measurement device 10 also includes a short-range communication device for connecting to other nearby devices. The measurement device 10 and the user terminal 20 can transmit and receive various information via short-range wireless communication (e.g., Bluetooth (registered trademark)) by pairing with each other.

[0031] FIG. 3 is a block diagram showing the hardware configuration of a health assessment system 510 according to the first embodiment of the present invention. The health assessment system 510 includes an input unit 501, a weight measurement unit 502, a bioelectrical impedance measurement unit 503, a storage unit 504, a control unit 505, and an output unit 506. Of these components, the weight measurement unit 502 and the bioelectrical impedance measurement unit 503 are provided in the measurement device 10. However, since the measurement device 10 and the user terminal 20 can communicate with each other as described above, the input unit 501, the storage unit 504, the control unit 505, and the output unit 506 may be provided in either the measurement device 10 or the user terminal 20, or may be provided in both. In this embodiment, as an example, the weight measurement unit 502 and the bioelectrical impedance measurement unit 503 are provided in the measurement device 10, and the input unit 501, the storage unit 504, the control unit 505, and the output unit 506 are provided in the user terminal 20.

[0032] The input unit 501 accepts operation inputs from the user. In this embodiment, the height, age, and gender of each user are particularly input to the input unit 501. The operation buttons 18A to 18D of the measuring device 20 and the touch panel of the user terminal 20 can all be the input unit 501. The weight measuring unit 502 corresponds to the load cell of the measuring device 10.

[0033] The bioelectrical impedance measuring unit 503 also includes the electrodes 111R, 111L, 112R, 112L, 161R, 161L, 162R, and 162L provided in the measuring device 10, the current-carrying electrodes 161R, 161L, 111R, and 111L, and a current control circuit that applies a weak constant AC current to these electrodes. The current control circuit can apply constant AC current at a plurality of frequencies.

[0034] The storage unit 504 stores, for each user, information input from the input unit 501, the weight measured by the weight measurement unit 502, the bioelectrical impedance measured by the bioelectrical impedance measurement unit 503, the body composition calculated by the control unit 505, etc. The storage unit 504 also stores a measurement program for measuring the weight and body composition, and a health assessment program according to this embodiment, as well as data generated by these programs and various information used in these programs (for example, an inference model, which will be described later).

[0035] Control unit 505 controls each unit of health assessment system 510 in accordance with the measurement program, and calculates body composition and determines health based on information input to input unit 501, weight measured by weight measurement unit 502, and bioelectrical impedance measured by bioelectrical impedance measurement unit 503 in accordance with the health assessment program. Output unit 506 corresponds to display panel 17 of measuring device 20 or the touch panel of user terminal 20. Under the control of control unit 505, output unit 506 displays a screen for inputting information to input unit 501, a screen for controlling control unit 505, a screen showing the results of calculations performed by control unit 505, and the like. Note that the measurement program and health assessment program may be provided to measurement device 10 or user terminal 20 by user terminal 20 downloading them via a communication network. Alternatively, the measurement program and health assessment program may be recorded on a non-transitory recording medium, and measurement device 10 or user terminal 20 reads the measurement program and health assessment program from the recording medium, thereby providing the measurement program and health assessment program to measurement device 10 or user terminal 20.

[0036] FIG. 4 is a block diagram showing the functional configuration of health determination system 510 according to the first embodiment of the present invention. In health determination system 510, various functions are realized by control unit 505 executing various programs. FIG. 4 particularly shows the functions realized by executing the health determination program of this embodiment. Health determination system 510 includes height / age / gender acquisition unit 51, weight acquisition unit 52, bioelectrical impedance acquisition unit 53, body composition estimation unit 54, health check result estimation unit 55, health determination unit 56, health advice provision unit 57, and output unit 58.

[0037] The height / age / gender acquisition unit 51 acquires the user's biometric information, such as height, age, and gender, by accepting user operation input via the input unit 501. The weight acquisition unit 52 acquires the user's biometric information, such as weight, by measuring the user's weight via the weight measurement unit 502. The bioelectrical impedance acquisition unit 53 measures the bioelectrical impedance of the user's whole body and each body part (hereinafter, the whole body will be referred to as "body part").

[0038] The bioelectrical impedance acquisition unit 53 measures the bioelectrical impedance of a plurality of body parts of the user, for example, in the following manner. (1) To measure the bioelectrical impedance of the whole body, current is supplied using current-carrying electrode 161L and current-carrying electrode 111L, and the potential difference between measurement electrode 162L in contact with the left hand and measurement electrode 112L in contact with the left foot is measured along the current path that flows through the left hand, left arm, chest, abdomen, left leg, and left foot. (2) To measure the bioelectrical impedance of the right leg, current is supplied using current-carrying electrode 161R and current-carrying electrode 111R, and the potential difference between measurement electrode 112L in contact with the left foot and measurement electrode 112R in contact with the right foot is measured along the current path that flows through the right hand, right arm, chest, abdomen, right leg, and right foot. (3) To measure the bioelectrical impedance of the left leg, current is supplied using current-carrying electrode 161L and current-carrying electrode 111L, and the potential difference between measurement electrode 112L in contact with the left foot and measurement electrode 112R in contact with the right foot is measured along the current path that flows through the left hand, left arm, chest, abdomen, left leg, and left foot. (4) To measure the bioelectrical impedance of the right arm, current is supplied using current-carrying electrode 161R and current-carrying electrode 111R, and the potential difference between measurement electrode 162L in contact with the left hand and measurement electrode 162R in contact with the right hand is measured along the current path that flows through the right hand, right arm, chest, abdomen, right leg, and right foot. (5) To measure the bioelectrical impedance of the left arm, current is supplied using current-carrying electrode 161L and current-carrying electrode 111L, and the potential difference between measurement electrode 162L in contact with the left hand and measurement electrode 162R in contact with the right hand is measured along the current path that flows through the left hand, left arm, chest, abdomen, left leg, and left foot. In the case of a system that does not have electrodes that come into contact with the hands and only has electrodes that come into contact with the left and right feet, the bioelectrical impedance of only one body part is measured (the measurement location is different from that of a system that has electrodes that come into contact with the hands, but it is the bioelectrical impedance that corresponds to the whole body), but even in this case it is possible to estimate multiple items of body composition.

[0039] In this manner, the bioelectrical impedance acquisition unit 53 passes a constant AC current from each current-carrying electrode to a predetermined part of the user's body and measures the potential difference generated in this current path. Then, based on the values ​​of the current and potential difference, the bioelectrical impedance of the user's multiple body parts is calculated. The configuration for measuring the bioelectrical impedance may be the same as that of a general body composition analyzer. Note that the bioelectrical impedance for each of the multiple body parts is calculated when a constant AC current of a standard frequency (e.g., 50 kHz) is passed, when a constant AC current of a high frequency (e.g., 250 kHz) is passed, and when a constant AC current of a low frequency (e.g., 5 kHz) is passed.

[0040] The body composition estimation unit 54 acquires the height, age, and gender acquired by the height / age / gender acquisition unit 51, the weight acquired by the weight acquisition unit 52, and the bioelectrical impedance acquired by the bioelectrical impedance acquisition unit 53, and performs calculations using this information to estimate the user's body composition. The body composition estimation unit 54 applies the biometric information and bioelectrical impedance to a predetermined regression equation to perform calculations, thereby acquiring body composition information such as fat percentage, fat mass, fat-free mass, muscle mass, visceral fat mass, visceral fat level, visceral fat area, subcutaneous fat mass, basal metabolic rate, bone mass, total body water percentage, BMI (Body Mass Index), intracellular fluid volume, and extracellular fluid volume. Furthermore, an index value indicating muscle quality may be acquired as body composition information. The configuration for calculating body composition information may be similar to that of a general body composition monitor. The body composition estimation unit 54 outputs the acquired body composition information to the medical examination result estimation unit 55 and to the output unit 58.

[0041] The medical checkup result estimation unit 55 acquires biological information including height, age, and sex acquired by the height / age / sex acquisition unit 51 and weight acquired by the weight acquisition unit 52, and body composition information calculated by the body composition estimation unit 54, and estimates the medical checkup result based on this information.

[0042] Health checkups involve testing a variety of items. For example, according to the "2020 One-Day Health Checkup Basic Test Item List" published by the Japan Society of Ningen Dock, essential items for a health checkup include body measurements, physiology, X-rays and ultrasound, biochemistry, hematology, serology, urine, stool, interview and examination, and assessment and guidance. The health checkup result estimation unit 55 estimates health checkup results from body composition information, but at this time, for body measurements, biometric information such as height and weight has already been obtained. Furthermore, the correlation between body composition and physiology, X-rays and ultrasound, hematology, serology, urine, and stool is relatively weak. Therefore, when estimating health checkup results from body composition, the health checkup result estimation unit 55 of this embodiment particularly estimates values ​​for biochemistry test items, which have a relatively strong correlation with body composition among health checkup items.

[0043] According to the "2020 One-Day Medical Checkup Basic Test Item List" above, biochemistry test items include total protein, albumin, creatinine, eGFR (estimated glomerular filtration rate), uric acid, total cholesterol, HDL (High Density Lipoprotein) cholesterol, LDL (Low Density Lipoprotein) cholesterol, non-HDL cholesterol, triglycerides, total bilirubin, AST (GOT) (Aspartate transaminase (Glutamic oxaloacetic transaminase)), ALT (GPT) (Alanine aminotransferase (Glutamic pyruvic transaminase)), γ-GT (γ-GTP) (γ-Glutamyl transaminase), and γ-GTP (γ-Glutamyl Glutamate). These include gamma-glutamyl transpeptidase (ALP), alkaline phosphatase (ALP), blood glucose (fasting), and hemoglobin A1c (HbA1c). The medical examination result estimation unit 55 estimates some or all of these.

[0044] The medical examination result estimation unit 55 further estimates items corresponding to hematology and serology among the essential items of a medical examination. Hematology test items include red blood cells, white blood cells, hemoglobin, hematocrit, MCV (Mean Corpuscular Volume), MCH (Mean Corpuscular Hemoglobin), MCHC (Mean Corpuscular Hemoglobin Concentration), and platelet count, while serology test items include CRP (C-Reactive Protein), blood type (ABO·Rh), and HBs antigen. The medical examination result estimation unit 55 estimates some or all of these items.

[0045] Before describing the operation of the medical examination result estimation unit 55, the relationship between the biological information, body composition information, and biochemical test values ​​will be described.

[0046] 5 is a graph showing the correlation between HDL cholesterol (vertical axis), which is a biochemical test value, and fat mass (kg) (horizontal axis), which is body composition information, according to the first embodiment of the present invention. As can be seen from the graph in FIG. 5, there appears to be a linear correlation between HDL cholesterol and fat mass, with the greater the fat mass, the lower the HDL cholesterol. However, the variability is relatively large, making it difficult to accurately estimate HDL cholesterol from fat mass.

[0047] FIG. 6 shows a graph of HDL cholesterol (vertical axis) and visceral fat mass (cm ) as body composition information according to the first embodiment of the present invention. 2 6 shows the correlation between HDL cholesterol and the square of visceral fat mass. As can be seen from the graph in Figure 6, there appears to be a correlation between HDL cholesterol and the square of visceral fat mass, but even in this case the correlation is weak and there is too much variability to estimate HDL cholesterol from visceral fat mass.

[0048] 7 is a graph showing the accuracy of estimation using multiple items of body composition information according to the first embodiment of the present invention. The vertical axis of the graph in FIG. 7 represents measured HDL cholesterol, and the horizontal axis represents estimated HDL cholesterol estimated by multiple regression analysis using multiple items of body composition information. By performing statistical analysis such as multiple regression analysis using multiple items of body composition information, HDL cholesterol can be estimated with high accuracy, as shown in FIG. 7.

[0049] The health check result estimation unit 55 estimates health check results from the biometric information and body composition information using a multiple regression equation obtained by performing statistical analysis, such as multiple regression analysis, on a large number of pairs of biometric information and body composition information and health check results. This multiple regression equation uses the biometric information and body composition information as explanatory variables and the health check results as the target variable. This multiple regression equation can also be considered an inference model obtained by learning using pairs of a large number of pairs of biometric information and body composition information and health check results obtained from actual health checks as training data. Note that the inference model is not limited to a multiple regression equation, and other inference models generated by learning, such as a decision tree or a neural network, may also be used. For example, with regard to age and gender, different multiple regression equations may be prepared and used for each age and gender without using them as explanatory variables.

[0050] The medical examination result estimation unit 55 may, for example, set the fat mass (kg) to x1 and the visceral fat mass (cm 2 ) is x2, HDL cholesterol Y is estimated by multiple regression analysis using the following multiple regression equation (1): Note that, to improve the accuracy of estimating biochemical test values, for example, HDL cholesterol Y may be estimated using a multiple regression equation that takes into account the amount and / or direction of change in x1 and x2 and adds them as explanatory variables. Y=ax1 / x2 2 +b (1)

[0051] The health check result estimation unit 55 stores a plurality of multiple regression equations such as those described above for estimating biochemical test values ​​for a plurality of items. The biological information and body composition information used to estimate each biochemical test value are generally different. The health check result estimation unit 55 estimates each biochemical test value by substituting the biological information and body composition information required for the regression equation for estimating each biochemical test value. The health check result estimation unit 55 outputs the estimated health check results to the health degree assessment unit 56 and also to the output unit 58.

[0052] The health degree determining unit 56 determines the health degree based on the estimated health check result values ​​of multiple items estimated by the health check result estimating unit 55. However, as described above, in this embodiment, the results of not all items obtained in an actual health check are estimated as health check results, but biochemical test values ​​in particular. Therefore, the health degree determining unit 56 determines the health degree based on this information.

[0053] Health level determining section 56 may determine, as the health level, a lifestyle-related disease overall prevalence risk (hereinafter simply referred to as "disease risk") that indicates the overall risk of lifestyle-related disease-related illnesses.

[0054] According to conventional methods, the prevalence of disease is determined by calculating a total score from, for example, chest X-ray, urinary protein, urinary occult blood, AST (GOT), γ-GT (γ-GTP), LDL cholesterol, HDL cholesterol, triglycerides, uric acid, HbA1c, FPG, hematocrit, red blood cells, CRP, systolic blood pressure, diastolic blood pressure, etc. However, the health condition determination unit 56 of this embodiment determines the prevalence of disease based on, for example, AST (GOT), γ-GT (γ-GTP), LDL cholesterol, HDL cholesterol, triglycerides, uric acid, HbA1c, hematocrit, and CRP obtained by estimation in the health check result estimation unit 55.

[0055] The health degree determination unit 56 may determine the prevalence risk as a health degree using a total score calculation formula that follows the conventional total scoring method and is modified for health checkup results (e.g., chest X-ray, blood pressure, etc.) for which no estimation result can be obtained by the health checkup result estimation unit 55. The modification may be to simply omit items for which no estimation result can be obtained in the conventional total scoring method.

[0056] Here, depending on the item, the health checkup results estimated from biometric information and body composition information may be less accurate than the health checkup results obtained by actually conducting a health checkup. Therefore, when the prevalence risk is calculated from the estimated health checkup results using a total score calculation formula that follows conventional methods, the prevalence risk value may deviate from the actual prevalence risk.

[0057] To prevent such cases from occurring, the health level assessment unit 56 may use a multiple regression equation obtained by performing statistical analysis, such as multiple regression analysis, on pairs of health levels and health check results estimated from multiple sets of biometric information and body composition information to determine health levels from the health check results estimated by the health check result estimation unit 44. This multiple regression equation uses the health check results estimated from the biometric information and body composition information as explanatory variables and the health level as the target variable. This multiple regression equation can also be considered an inference model obtained by learning using pairs of health check results estimated from multiple sets of biometric information and body composition information and the health levels of the subjects from whom the biometric information and body composition information were obtained as training data. The inference model is not limited to a multiple regression equation, and may be other inference models generated by learning, such as a decision tree or a neural network. For example, for age and gender, different multiple regression equations may be prepared and used for each age and gender without using them as explanatory variables. In addition, in the health assessment system 510, not only estimated health check results but also biological information and body composition information are obtained, so not only the estimated health check results but also the biological information and body composition information may be used as explanatory variables.

[0058] The health level in the training data used for this learning may be (1) a disease risk value calculated by a conventional comprehensive scoring system from the results of an actual health check (including information from a chest X-ray, etc.), (2) information on whether or not the patient actually has a lifestyle-related disease, or (3) information on actual medical expenses, with medical expenses considered to indicate the presence or severity of a lifestyle-related disease. Even if a user does not know where their health level falls out of, for example, 10 levels, they can simply know whether they are in the "normal range" or the "dangerous range," and as long as the accuracy of the two types of discrimination is high, they can recognize risk using a measuring device they use at home on a daily basis, and find the system quite useful.

[0059] In any of the cases (1) to (3), the multiple regression equation that inputs the estimated health check results and outputs the health score is different from the calculation equation used for conventional overall scoring. Through the learning described above, this equation is adjusted to an index that combines health check result items using a different method than conventional methods to strengthen the correlation based on the properties of the bioelectrical impedance being measured. The ability to adjust the equation for calculating health score from health check results in this way means that even if the estimation accuracy based on body composition information is low for each of the multiple health check result items, the health score index combining those multiple items can be selected from a large number of combinations to achieve high accuracy in limited situations (i.e., high degree of agreement with actual health risks from a specific perspective). This makes it possible to more reliably determine a user's health score by measuring bioelectrical impedance. This will be explained in more detail again using Figure 9.

[0060] Health level determination unit 56 outputs the determined health level to health advice providing unit 57 and also to output unit 58.

[0061] Health advice providing section 57 generates information for presenting to the user the health level determined by health level determining section 56 or information based on that health level, and outputs the information to output section 58.

[0062] The output unit 58 displays the body composition acquired by the body composition estimation unit 54, the health check result estimated by the health check result estimation unit 55, the health degree determined by the health degree determination unit 56, and the information generated by the health advice provision unit 57. Each piece of health information may be displayed separately or all at once.

[0063] As described above, the memory unit 504 and the control unit 505 may be provided in either the measurement device 10 or the user terminal 20, but one or both of the memory unit 504 and the control unit 505 may be provided in a server that can communicate with the user terminal 20 via a communication network. Furthermore, if the measurement device 10 has a function for communicating with a communication network, the memory unit 504 and the control unit 505 may be provided in the server, thereby eliminating the need for the user terminal 20. In other words, the body composition estimation unit 54, the health check result estimation unit 55, the health degree determination unit 56, and the health advice provision unit 57 may be provided in a server that can communicate with the measurement device 10 or the user terminal 20 via a communication network.

[0064] (Second embodiment) The health determination system of the second embodiment estimates a plurality of body composition items from the bioelectrical impedance of a plurality of body parts, and determines a health level based at least in part on the estimated body composition items.

[0065] 8 is a block diagram showing the functional configuration of health determination system 520 according to the second embodiment of the present invention. In health determination system 520, components similar to those in health determination system 510 according to the first embodiment are assigned the same numbers and descriptions thereof will be omitted where appropriate. Health determination system 520 includes height / age / gender acquisition unit 51, weight acquisition unit 52, bioelectrical impedance acquisition unit 53, body composition estimation unit 54, health determination unit 56', health advice provision unit 57', output unit 58, and additional information acquisition unit 59.

[0066] The configurations and operations of height / age / gender acquisition unit 51, weight acquisition unit 52, bioelectrical impedance acquisition unit 53, and body composition estimation unit 54 are the same as those in the first embodiment. Body composition estimation unit 54 outputs the acquired body composition information to health level determination unit 56′ and output unit 58.

[0067] Health level determination unit 56' determines the health level based on the height, age, and gender information acquired by height / age / gender acquisition unit 51, the weight information acquired by weight acquisition unit 52, and the body composition information acquired by body composition estimation unit 54. Health level determination unit 56' may determine the disease risk as the health level, similar to the first embodiment.

[0068] As described above, according to conventional methods, the disease risk is determined by calculating a total score based on, for example, chest X-ray, urinary protein, urinary occult blood, AST (GOT), γ-GT (γ-GTP), LDL cholesterol, HDL cholesterol, triglycerides, uric acid, HbA1c, FPG, hematocrit, red blood cells, CRP, systolic blood pressure, diastolic blood pressure, etc. However, the health level determination unit 56' of the second embodiment does not estimate any of these medical examination results, but determines the disease risk from biological information and body composition.

[0069] The health level assessment unit 56' assesses the health level from the biometric information and body composition information using a multiple regression equation obtained by performing statistical analysis, such as multiple regression analysis, on a large number of pairs of biometric information and body composition information and health levels. This multiple regression equation uses the biometric information and body composition information as explanatory variables and the health level as a target variable. This multiple regression equation can also be considered an inference model obtained by learning using as training data pairs of a large number of pairs of biometric information and body composition information and health levels calculated from the results of actual health checks of subjects from whom the biometric information and body composition information were obtained. As with the first embodiment, the inference model is not limited to a multiple regression equation and may be other inference models generated by learning, such as a decision tree or a neural network. For example, for age and gender, different multiple regression equations may be prepared and used for each age and gender without using them as explanatory variables.

[0070] 9 is a graph showing the accuracy of health assessment using body composition information for multiple items according to the second embodiment of the present invention. The vertical axis of the graph in FIG. 9 represents the total disease risk score (true value) calculated, for example, from health checkup results using a conventional method, and the horizontal axis represents the disease risk score (determined value) calculated by health assessment unit 56' by inputting body composition information for multiple items into a multiple regression equation. In this graph, line 91 represents a function showing the relationship between the true value and the corresponding determined value; the closer a point on the graph is to line 91, the closer the determined value is to the true value (higher the determination accuracy).

[0071] As described above, the health level (true value) on the vertical axis of Figure 9 may be (1) a disease risk value calculated by a conventional comprehensive score system based on actual health checkup results including information such as chest X-rays, (2) information on whether or not a person actually has a lifestyle-related disease, or (3) information on actual medical expenses, with medical expenses considered to indicate the presence or severity of a lifestyle-related disease. The health level (determined value) on the horizontal axis of Figure 9 may be a value calculated by inputting multiple items of body composition information into a multiple regression equation, or a value calculated by inputting one or more items of body composition information and one or more items of biological information into a multiple regression equation.

[0072] Here, if a boundary 92 between healthy and unhealthy is set for the true value of the disease risk (health level), a boundary 93 of the judgment value corresponding to the boundary 92 can be obtained using a line 91. The judgment value at this boundary 93 is assumed to be y0. In the example of FIG. 9, there are cases where the judgment value is y0 or greater but the true value indicates that the subject is healthy, i.e., false positives 94 shown in FIG. 9. There are also cases where the judgment value is y0 or less but the true value indicates that the subject is unhealthy, i.e., false negatives 95 shown in FIG. 9.

[0073] Therefore, health level determination unit 56' sets boundary 96 corresponding to determination value y1 so as to include false positives 94, with boundary 93 at the center, and sets boundary 97 corresponding to determination value y2 so as to include false negatives 95, and sets the range from y2 to y1 as an uncertain region. Note that the uncertain region does not necessarily need to include all false positives or all false negatives in the learning data, but it is desirable to set it so as to reduce false positives and false negatives as much as possible.

[0074] As described above, health assessment unit 56' of this embodiment assesses health level using three values ​​(three ranges): healthy, unhealthy, and uncertain. However, it is desirable to narrow the range of values ​​set as described above and resulting in an uncertain assessment as narrow as possible. To achieve this, the multiple regression equation, which inputs values ​​for multiple items of body composition information (and biological information), is adjusted by performing the statistical analysis including the learning described above so as to narrow the uncertain range. Narrowing the uncertain range in this way means that the accuracy of determining the two types of "health risk (dangerous range)" and "no health risk (normal range)" can be improved, making it possible to achieve useful health level assessment using a body composition monitor used on a daily basis.

[0075] Determining health level into three categories and adjusting the multiple regression equation for calculating health level so as to narrow the uncertainty region (and thus making it possible to determine health level into two categories, healthy and unhealthy) are similarly applied to health level determination unit 56 in the first embodiment. Moreover, by applying the method described below, it becomes possible to further narrow the uncertainty region in both health level determination unit 56 in the first embodiment (which uses medical examination results estimated from body composition information) and health level determination unit 56' in the second embodiment (which directly uses body composition information).

[0076] 10A and 10B are graphs showing an example (example of the second embodiment) in which the vertical axis represents the health level (true value) and the horizontal axis represents the health level (determined value) calculated directly based on body composition information and biological information, such as body fat percentage, visceral fat, muscle mass, basal metabolic rate, age, and gender. This method not only adjusts the multiple regression equation for calculating the health level (determined value) (the horizontal axis of FIG. 9 ) based on the medical checkup results estimated from the body composition information or the body composition information itself, but also adjusts the equation for calculating the health level (true value) (the vertical axis of FIG. 9 ), which serves as a basis for adjusting the multiple regression equation. In other words, while changing not only the equation for calculating the health level (determined value) on the horizontal axis but also the equation for calculating the health level (true value) on the vertical axis, a state is searched for in which the multiple sets of plotted data are as close as possible to line 91 (reducing the number of data points that are false positives 94 and false negatives 95 and their distance from line 91).

[0077] Figure 10A shows a graph in which the disease risk assessment value obtained from actual health checkup results using the conventional total score calculation formula is used as the true value, and Figure 10B shows a graph in which the disease risk assessment value obtained by performing the above search is used as the true value.

[0078] For this search, if the formula for calculating the health index (true value) on the vertical axis is experimentally changed, the plot on the vertical axis will be plotted using a value different from the value calculated from actual health checkup results using the conventional total score calculation formula. The value to be used as the true value, i.e., the formula for calculating the health index (true value), can be determined so as to have a strong correlation with information on whether the subject having the body composition information (and biological information) actually has a lifestyle-related disease and / or information on the actual medical expenses incurred by the subject having the body composition information (and biological information) (which may be medical expenses extracted by limiting the type of disease or total medical expenses). Such a health index (true value) will be an assessment of the risk of related diseases using a combination of indicators that have a high affinity with body composition, such as TG, LDL-C, HDL-C, and HbA1c.

[0079] As shown in Figure 10A, if health level (true value) is determined based on a risk assessment from general health checkup results without adjusting the vertical axis, the health level (assessed value) on the horizontal axis will include indicators that do not have a strong relationship with the body composition information that forms the basis of the health level (assessed value), and so even if there is a relationship between the vertical and horizontal axes, the margin of error will be large. However, by adjusting the way the health level (true value) is determined to minimize error as much as possible, in order to obtain a health level (true value) that is in line with whether or not a specific disease is actually present and whether or not medical expenses are actually being spent, from a combination of indicators that have a high affinity with body composition information, the effectiveness of health level assessment from body composition information can be increased, as shown in Figure 10B.

[0080] In the example of Figure 10B, a high correlation was obtained by limiting the disease risks used as the health level as a result of the search. Thus, if the specific disease focused on when adjusting the method for determining the health level (true value) is a disease closely related to body composition (e.g., arteriosclerosis), the effectiveness of the assessment is effectively increased. In other words, the health level assessment unit 56 or 56' is highly effective when assessing, in addition to disease risk, other health levels, such as metabolic syndrome warning risk indicating the risk of metabolic syndrome, vascular damage risk indicating the risk of damage to blood vessels, glucose metabolism disorder (diabetes) risk indicating the risk of glucose metabolism disorder (diabetes), liver function warning risk indicating the risk of liver dysfunction or liver disease, and sarcopenia / frailty risk indicating the risk of muscle weakness and physical frailty.

[0081] When these health indicators are calculated using conventional methods based on a doctor's judgment or a predetermined algorithm, metabolic syndrome risk is generally calculated from blood lipids (TG, TC, LDL-C, HDL-C), glucose tolerance index (fasting blood glucose, HbAic, insulin), blood pressure, and waist circumference; vascular damage risk is calculated from blood lipids, blood pressure, glucose tolerance, and lifestyle habits such as smoking habits; glucose metabolism abnormality (diabetes) risk is calculated from HbA1c, blood glucose (fasting), and obesity level; liver function risk is calculated from AST (GOT), ALT (GPT), and γ-GTP; and sarcopenia / frailty risk is calculated from whole body muscle mass, weight, grip strength, height, walking speed, gait, center of gravity sway, and lower limb muscle strength.

[0082] When determining these health levels according to the first embodiment, the various types of information mentioned above that form the basis for determining the health levels are estimated from body composition information (and biological information). For example, blood lipids, blood pressure, abdominal circumference, some lifestyle habits, grip strength, walking speed, gait, center of gravity sway, etc. can be estimated with relatively high accuracy using body composition information (and biological information), and obesity level, whole body muscle mass, weight, height, lower limb muscle strength, etc. can be obtained as body composition information (and biological information) itself.

[0083] In contrast, in the second embodiment, these health levels are determined directly from body composition information (and biological information), and the second embodiment may be able to provide a determination with a narrower uncertainty range than the first embodiment. In particular, the risk of diseases, the probability of which is affected by daily living habits, can be measured naturally on a daily basis under normal conditions, and direct determination from body composition, which reflects subtle daily changes, is expected to be more suitable for detecting true warning signals. Therefore, even if it is difficult to accurately estimate each of the liver function indicators obtained from blood tests in health checkups from body composition information (and biological information), for example, the health level determination system 520 can capture the state and temporal changes of body composition, which is measured daily, and issue a warning signal indicating a liver function warning risk.

[0084] Whether or not a useful judgment can be made for the user depends on the inference model. Therefore, it is advantageous to select an appropriate type of inference model (multiple regression analysis, principal component analysis, decision tree, neural network, Bayesian network, etc.) and to prepare a sufficient amount of good training data to perform appropriate training. As good and sufficient training data, not only are pairs of body composition information (and biological information) and health status (true value) obtained from information at the time the body composition was measured (such as the overall score based on the results of a health checkup actually performed at that time, the presence or absence of actual illnesses at that time, or actual medical expenses at that time) prepared, but also pairs of body composition information (and biological information) measured at multiple times over a specified period and health status (true value) obtained from information at the end of that period prepared. This makes it possible to accurately issue warning signals indicating a health risk based on daily changes in body composition measured on a daily basis.

[0085] 10A and 10B, the usefulness can be improved by adjusting the health level (true value) used as training data so that it has a high correlation with body composition information. For example, in the case of vascular damage risk, instead of using the value calculated using conventional methods from information such as blood lipids, blood pressure, glucose tolerance, and smoking habits obtained in an actual health check as the true value, an inference model can be created that closely matches the value determined from body composition information (and biological information) while adjusting the way the true value is determined based on whether the symptoms of arteriosclerosis are actually progressing or whether medical expenses are actually being spent on treating arteriosclerosis.

[0086] Depending on the type of health level, for example, sarcopenia / frailty risk calculated from whole-body muscle mass / body weight, whole-body muscle mass / height, etc., may be calculated using a theoretically defined function or algorithm (rule-based) rather than an inference model. Such health levels can be determined using biological information and body composition information as they are. In other words, muscle mass, which is part of body composition information, can be used as whole-body muscle mass for calculating sarcopenia / frailty risk.

[0087] Furthermore, if the body composition estimation unit 54 has the function of estimating muscle fiber density, muscle fullness, and muscle cell size based on the frequency characteristics of electrical resistance, the ratio of resistance to reactance, the phase angle, etc., this evaluation value may be used to determine the risk of sarcopenia or frailty. In the body composition monitor commercialized by the present applicant, this evaluation value is output as a muscle quality score (trademark pending). Thus, even when not relying on an inference model, determining the risk from body composition information may be more direct and accurate than determining the risk from health checkup results.

[0088] Of course, sarcopenia and frailty risk can also be determined using an inference model. In this case, rather than using a rule-based value as the health level (true value), the true value is set based on whether sarcopenia symptoms are actually progressing, whether medical expenses are actually being spent on treating sarcopenia, etc., and the inference model is developed to closely match the value determined from body composition information (and biological information). The body composition information used in the inference model includes indicators of not only muscle condition but also fat condition, making it possible to determine sarcopenia and frailty risk with greater accuracy.

[0089] As described above, the health level determination unit 56 or 56' in this embodiment determines each of the above health levels as three values: presence or absence of risk and uncertainty, or as two values: presence or absence of risk if the uncertainty range can be narrowed sufficiently. When the health level is obtained as a continuous value, the range of the continuous value can be divided into healthy (no risk), uncertainty, and unhealthy (risky) in that order for determination. Furthermore, when a probabilistic model such as a Bayesian network is used as the inference model, the health level does not necessarily need to be calculated as a continuous value, and healthy (no risk), uncertainty, or unhealthy (risky) can be determined according to the obtained probability.

[0090] The health level determination unit 56 or 56' outputs the determined health level to the health advice providing unit 57 or 57' and the output unit 58.

[0091] Health advice providing unit 57 or 57' determines health advice in accordance with the health level determined by health level determination unit 56 or 56'. The health advice is output as screen information to output unit 58, and is output (displayed) from output unit 58. If the health level determination result is healthy or unhealthy, health advice providing unit 57 or 57' generates screen information notifying the determination result to that effect, and if the determination result is uncertain, generates screen information requesting additional information about the user.

[0092] The user can easily add information by requesting, as additional information, lifestyle habits (drinking habits, smoking habits, sleep duration, etc.), as well as information that can be measured at home using measuring devices without going to a medical facility, such as values ​​from an activity monitor, a blood pressure monitor, or a sleep monitor. The health advice providing unit 57 or 57' generates a screen for inputting additional information, and the screen is displayed on the output unit 58, thereby configuring the additional information acquisition unit 59. The user can input additional information via the additional information acquisition unit 59. The additional information acquisition unit 59 may request, as additional information, information related to heartbeats, such as heart rate (pulse rate) and heartbeat fluctuations. Heartbeat-related information can be acquired using a wearable device equipped with a heartbeat sensor, or the measuring device 10 of the health assessment system 510 may calculate the heartbeat from bioelectrical impedance.

[0093] The health level determination unit 56 or 56' determines the health level using the additional information added to the additional information acquisition unit 59. In this case, if the health level based on the body composition information is obtained as a continuous value, the health level (continuous value) may be adjusted by modifying the health level based on the additional information. Furthermore, if the health level based on the body composition information is obtained using a probabilistic model, the health level determination unit 56 or 56' may use an inference model in which the biological information, body composition information, and additional information are explanatory variables (inputs) and the health level is the objective variable (output). Note that even when additional information is provided, the health level determination unit 56 or 56' may determine the health level as one of three values: healthy, unhealthy, or uncertain. However, when additional information is provided, the health level determination unit 56 or 56' may determine the health level as a binary value: healthy or unhealthy without determining it as uncertain. Alternatively, the health level determination unit 56 or 56' may determine the health level as one of three values: healthy, caution required, or unhealthy.

[0094] Fig. 11 is a diagram showing an example of a health advice screen according to an embodiment of the present invention. Output unit 58 displays the health advice screen. Health advice providing unit 57 or 57' generates screen information for displaying the health advice screen on output unit 58. In the example of Fig. 10, health level determination unit 56 or 56' determines the health level as one of three values: healthy, need attention, and unhealthy, and health advice screen 90 shows the determined health levels with advice of "Safe," "Need Attention," and "Vigilance," respectively.

[0095] FIG. 11 shows the result of assessing disease risk as a health level. On the health advice screen 100, the disease risk is shown by a clock-like dial 101 and hands 102 called a health meter. On the dial 101, a sector-shaped area from midnight to a predetermined time is designated as a safe area 111, a sector-shaped area following the safe area 111 up to the predetermined time is designated as a caution area 112, and a sector-shaped area following the caution area 112 up to 12 o'clock (midnight) is designated as an alert area 113. The hands 102 indicate the assessment result. In the example of FIG. 10, the assessment result is "safe," which is relatively close to "caution required."

[0096] The health advice screen 100 also shows a predicted value three years from now with a needle 103. The predicted value three years from now may be calculated by advancing the current judgment value (hand 102) by a predetermined time (e.g., 1.5 hours) on the health advice screen 100, taking into account age, gender, etc., or may be calculated from a history of past judgment values ​​if such a history is available. The health meter may also show how far the needle will move back by taking action to improve the health.

[0097] The health advice providing unit 57 or 57' similarly generates the health advice screen 100 of the health meter for other health levels in accordance with the judgment results of the health level judgment unit 56 or 56', and the output unit 58 displays it.

[0098] FIG. 12 is a diagram showing another example of a health advice screen according to an embodiment of the present invention. FIG. 12 shows an example in which the user terminal 20 is a wearable information processing device. Specifically, the user terminal 20′ as a wearable information processing device is a smart watch or smart wristband worn on the arm. In this example, the user terminal 20′ has a liquid crystal display device in the dial portion, and this display device corresponds to the output units 506 to 58. In this case, for example, of the configuration shown in FIG. 3, the input unit 501, weight measurement unit 502, bioelectrical impedance measurement unit 503, memory unit 504, and control unit 505 are provided in the measurement device 10, and the output unit 506 is provided in the user terminal 20′.

[0099] The user terminal 20' can communicate with the measurement device 10 via short-range wireless communication, and screen information showing the health meter is sent from the measurement device 10 to the user terminal 20'. The user terminal 20 can switch between displaying a clock and other information and displaying the health meter.

[0100] The user terminal 20' may include an accelerometer for measuring the acceleration of the user's body. The user terminal 20' may also include a pulse meter for detecting the user's pulse. The user terminal 20' may move the pointer of the health meter by combining information such as the user's body movements and activity tendencies detected by the accelerometer, energy consumption, and pulse rate measured by the pulse meter. For example, if the user exercises a lot and consumes a lot of energy, the pointer may return to a safer side, and conversely, if the user exercises less than the average, the pointer may move to a higher risk side.

[0101] Furthermore, the user terminal 20' as such a wearable information processing device may be paired as a second user terminal with the handheld user terminal 20 as a first user terminal. In this case, for example, the weight measurement unit 502 and the bioelectrical impedance measurement unit 503 are provided in the measurement device 10, the input unit 501, the storage unit 504, and the control unit 505 are provided in the user terminal 20, and the output unit 506 is provided in the user terminal 20 and the user terminal 20'. Then, the terminal that displays the health meter can be switched between the user terminal 20 and the user terminal 20'.

[0102] As described above, the health degree determination system 520 according to the second embodiment can determine a health degree without estimating the medical examination results from body composition information. Therefore, errors due to estimation of the medical examination results from body composition information and errors due to determination of the health degree from the medical examination results do not overlap, and the accuracy of the health degree determination can be improved.

[0103] Furthermore, health degree determination system 510 or 520 of this embodiment determines health degree using three values: healthy, uncertain, and unhealthy. Therefore, a definitive determination can be made when there is no (or only a small) suspicion of a false positive or false negative, and an inconclusive determination can be made when there is a suspicion of a false positive or false negative. Furthermore, health degree determination system 50 of this embodiment requests additional information when the determination result is uncertain, but does not request additional information when a definitive determination can be made without additional information, thereby enabling a simple health degree determination. Furthermore, even when an inconclusive determination result is obtained, by requesting additional information that can be obtained at home, such as values ​​from an activity meter or a blood pressure monitor, it is possible to obtain a health degree determination result without going to a medical facility.

[0104] Furthermore, in the health degree determination system 510 or 520 of this embodiment, the health degree determination result is displayed on a health advice screen 100 including a clock-like graphic, so that the user can intuitively grasp their own health degree by checking this health advice screen. As described above, the health advice screen 100 may be displayed on the user terminal 20, on the user terminal 20', or on the display panel 17 of the measurement device 20 (in this case, the user can know their own health degree simply by standing on the measurement device 20).

[0105] (Third embodiment) The health assessment system of the third embodiment assesses health based at least in part on the bioelectrical impedance of multiple body parts. In the second embodiment, health was assessed from body composition information, but this body composition information is calculated by estimation based on the user's biometric information (height, weight, etc.) and the bioelectrical impedance of multiple body parts. Therefore, by assessing health from the primary information, i.e., biometric information, and the bioelectrical impedance of multiple body parts, it may be possible to assess health with higher accuracy than in the second embodiment.

[0106] 13 is a block diagram showing the functional configuration of a health degree determination system 530 according to the third embodiment of the present invention. In health degree determination system 530, components similar to those in health degree determination system 520 according to the second embodiment are assigned the same numbers and descriptions thereof will be omitted where appropriate. Health degree determination system 530 includes height / age / gender acquisition unit 51, weight acquisition unit 52, bioelectrical impedance acquisition unit 53, health degree determination unit 56", health advice provision unit 57", and output unit 58.

[0107] As described above, the bioelectrical impedance acquisition unit 53 obtains the bioelectrical impedance of five body parts, namely the whole body, right leg, left leg, right arm, and left arm, for each high-frequency and low-frequency current, resulting in a total of 10 types of impedance values.

[0108] The health level determination unit 56'' determines the health level using some or all of these 10 types of impedance values ​​and biometric information (height, weight, etc.). The health level determination unit 56'' determines the health level using an inference model. In this inference model, the biometric information and impedance values ​​are used as explanatory variables (inputs), and the health level is used as the objective variable (output). To train this inference model, a large number of pairs of biometric information and impedance values ​​and health levels (presence or absence of illness, medical expenses, etc.) can be used as training data.

[0109] As in the second embodiment, health degree determination unit 56" of the third embodiment determines health degree using three values: healthy, unhealthy, and uncertain. As in health degree determination system 520 of the second embodiment, health degree determination system 530 may request additional information when the determination result is uncertain, and make a more accurate determination based on the additional information. As in the second embodiment, body composition estimation unit 54 may be provided to estimate body composition from biological information and bioelectrical impedance.

[0110] Everything explained in the second embodiment about the relationship between body composition information (and biological information) and various health indicators (true value, judgment value) can be applied to the relationship between bioelectrical impedance (and biological information) and various health indicators (true value, judgment value) in the third embodiment. Everything explained above about the display of health indicators, the configuration of the health advice screen, etc. can also be applied to the third embodiment.

[0111] As described above, according to the health assessment systems of the first to third embodiments, health is assessed based at least in part, indirectly or directly, on bioelectrical impedance, so that health can be easily assessed using a device such as a body composition analyzer.

[0112] In the above embodiment, a body composition monitor with eight electrodes is used, but a body composition monitor with four electrodes for only the feet or only the hands, or a body composition monitor with nine or more electrodes may also be used. Furthermore, although less accurate than a body composition monitor, a simplified body fat monitor in which a pair of current and voltage electrodes is held between the fingers or palms of both hands, or a body fat monitor or subcutaneous fat monitor that measures local bioelectrical impedance of a part of the upper arm or part of the abdomen, may also be used. [Explanation of symbols]

[0113] 10. Measurement equipment 11 Main body 12 Handle unit 13 Connection cord 14. Storage compartment 15 Handle body 16L,R···Grip 17 Display panel 18A~D···Operation buttons 20. User terminal 50 Health Information System 51 Height, age, and gender acquisition section 52 Weight acquisition unit 53 Bioimpedance acquisition unit 54...Body Composition Estimation Department 55. Biochemical Test Value Estimation Section 56,56´,56´´···Health Level Judgment Department 57 Health Advice Department 58 Output section 111L,R...electrode for energization 112L,R...Measurement electrode 161L,R...electrode for energization 162L,R...Measurement electrode 501 Input section 502 Weight measurement unit 503 Bioimpedance measurement unit 504...Storage section 505 Control unit 506···Output section

Claims

1. A health assessment system, comprising: a bioelectrical impedance acquisition unit that acquires bioelectrical impedance of one or more body parts of a user; a body composition estimation unit that estimates a plurality of body composition items of the user from the bioelectrical impedance of the one or more body parts; a health check estimation unit that inputs the body composition of the multiple items into a first inference model and estimates the health check results of the user of the multiple items as an output of the first inference model; a health assessment unit that inputs the health check results of the multiple items into a second inference model and obtains the user's health level as the output of the second inference model; A health assessment system equipped with the above.

2. The health assessment system according to claim 1, A health assessment system, wherein each of the health checkup results for the plurality of items has a significant correlation with the body composition for the plurality of items.

3. The health assessment system according to claim 1 , wherein the health level is a level reflecting information on the presence or absence of lifestyle-related diseases or a level of medical expenses involved.

4. The health assessment system according to any one of claims 1 to 3, The health assessment system, wherein the health assessment unit assesses whether the subject is healthy, unhealthy, or uncertain.

5. The health assessment system according to claim 4, an additional information acquisition unit that requests input of additional information that was not used to determine the health level when the result of the determination is inconclusive; The health assessment unit further assesses the health level using the additional information.

6. The health assessment system according to claim 4 or 5, an output unit that outputs the result of the health level determination, The output unit displays the result of the assessment by indicating the result on a clock-like graphic.

7. The health assessment system according to claim 6, The output unit indicates the predicted future assessment result on the graphic.

8. A health assessment program executed by a processor, the program causing the processor to: a body composition estimation unit that estimates a plurality of body composition items of the user from bioelectrical impedance acquired from one or more body parts of the user; a health check estimation unit that inputs the body composition of the plurality of items into a first inference model and estimates the health check results of the plurality of items of the user as an output of the first inference model; and a health assessment unit that inputs the results of the health checkups for the multiple items into a second inference model and obtains the health level of the user as the output of the second inference model; A health assessment program that functions as a

9. A health assessment server, a body composition estimation unit that estimates a plurality of body composition items of the user from bioelectrical impedance acquired from one or more body parts of the user; a health check estimation unit that inputs the body composition of the multiple items into a first inference model and estimates the health check results of the user of the multiple items as an output of the first inference model; a health assessment unit that inputs the health check results of the multiple items into a second inference model and obtains the user's health level as the output of the second inference model; A health assessment server equipped with the above.

Images