• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    Intelligent platform and method for monitoring body composition and assessing the nutritional and hydration status of a user. The present invention refers to a platform, portable and wireless, formed by three devices that makes measurements of the module and phase of the bioimpedance of a user in multiple configurable frequencies, and, from them, an estimation of the body composition and a assessment of nutritional status and hydration. The invention allows a distributed processing that optimizes resources in multi-user environments and provides a greater capacity for personalization; a method for estimating body composition based on a three dispersion bioimpedance model, more precise and more robust to disturbances, noise and parasitic effects; a method for resolution of parameters of the presented bioimpedance model; and capabilities for detection of alarm situations and integration of information in an e-Health system. (Machine-translation by Google Translate, not legally binding)
    公开号:ES2682059A2
    申请号:ES201730354
    申请日:2017-03-16
    公开日:2018-09-18
    发明作者:Laura María Roa Romero；Luis Javier Reina Tosina；David Naranjo Hernández
    申请人:Universidad de Sevilla；Centro de Investigacion Biomedica en Red CIBER；
    IPC主号:
    专利说明:

    OBJECT OF THE INVENTION
    The object of the invention described herein is included in the area of Information Technology and Communications (ICTs).
    More specifically, the object of the invention has a place in biomedical engineering and medical technology, for the development of portable electronic devices for monitoring physiological variables of people and their health status.
    BACKGROUND OF THE INVENTION
    The bioimpedance analysis (BIA) is a set of methods applied in the estimation of body composition through the measurement of impedance. The principle on which these methods are based is the determination of the impedance of an electric current that passes through the body, which can be estimated at one or multiple frequencies. The BIA analysis has become a standardized technique in estimating body composition since it does not present the restrictions of dissolution methods and provides more accurate estimates than anthropometric methods. In addition, it is a simple, safe and non-invasive technique. Thanks to the B1A analysis it is possible to obtain an estimate of the volumes of body fluids and body composition in both normal and pathological states.
    Bioimpedance methods also have many practical advantages: the instrumentation is portable, relatively cheap, and measurements can be performed quickly with minimal operator training; Bioimpedance methods require little maintenance and the measurements are safe, and easy to perform; It is a non-invasive technique, requiring only the placement of electrodes in the body; The results are obtained immediately, and the measurements can be repeated as often as desired, with a high inter-observer reproducibility. The clinical utility of the analysis of body composition through bioimpedance techniques has been demonstrated in numerous studies: in nephrology (identification of dry weight in renal patients, improvement of cardiovascular management, monitoring of fluid transfer during ultrafiltration, volume estimation of urea distribution in the calculation of the KW parameter and nutritional evaluation, etc.), in nutrition (chronic malnutrition, obesity, cachexia, sarcopenia, etc.), during pregnancy and lactation, in the risk assessment of various pathologies , as a marker or direct cause of diseases, during the decision-making process in a disease, during aging or a rehabilitation process, as a complement in the diagnosis and monitoring of conditions related to the cardiovascular system (fluid accumulation after surgery cardiac, hyponatremia, etc.), the immune system (patients with AIDS, dengue fever, hemorrhagic fever, chronic inflammation), in the evaluation of nutritional status in conditions related to the nervous system (Alzheimer's disease, mental anorexia, mental disability), in pediatrics, in oncology (evaluation of the patient's condition, early diagnosis, etc. .), in the postoperative period, in critically ill patients (monitoring of physiological tendencies in intensive care, sepsis, hemodynamic resuscitation, acute respiratory distress syndrome, etc.) in bedridden patients, in patients with liver cirrhosis, in gerontology; and even in sports science (evaluation of the effectiveness of a training program, detection of anomalies in the distribution of liquids, etc.).
    The BIA analysis is also useful in chronic respiratory diseases, where loss of body weight and decreased muscle mass have been recognized as risk factors associated with increased morbidity (inflammation, cachexia, anorexia, skeletal muscle dysfunction, increase of dyspnea, worsening health status, increased risk of exacerbations, decreased exercise capacity), mortality and a deterioration in the quality of life.
    This analysis is of great interest in patients with chronic or acute kidney disease, where excess fluid is also a condition related to increased

    morbidity and mortality.
    There are multiple documents related to bioimpedance devices and methods applied in the estimation of body composition. In this sense you have to:
    Document US2005 / 0192488 describes a monitoring system that also includes bioimpedance sensors.
    US7133716 shows a method for determining muscle and fat body masses from compartmental estimates of extracellular and intracellular water. Document US7783344 describes the hardware used by a bioimpedance device for estimating body volumes, although it does not show the method used in said determination.
    Document US2005 / 0101875 presents a device for monitoring non-homeostatic body composition. In this case a portable sensor of
    low cost and limited battery (can be disposable), called a risk monitor, which performs one or more bioimpedance measures on the user, as well as other biomedical signals such as cardiography, blood pressure, respiratory rate and body mass. The monitor derives, from these measurements and other manually entered data, evidence of fat and obesity or an anorexia or bulimia index, showing the results in real time. Normally, the determination of body composition is made from estimates of bioimpedance in a single frequency (SFBIA). An example is the device described in document US2008 / 0086058, which estimates the percentage of fat from the bioimpedance measurement at a single frequency. The use of a four electrode measurement configuration is also common, as is the case described in US6532384. The injected current normally circulates through the two external electrodes. The voltage measurement on the internal electrodes then allows the bioimpedance value to be established.
    In monofrequency devices, the determination of body composition is normally obtained from an equation that describes a regression line. This regression line minimizes the mean square error of body composition estimates with respect to a reference method. Regression equations usually respond to the following expression: Y = k1. H2 I Z + k2, H being the height of the subject and Z the value of the measured bioimpedance module. An example of application of said equation is in US6125297. However, there are also other proposals such as the equations described in US5615689, which respond to the following expression: Y = EXP (k,. LOG (Z) + k,. LOG (H) + k,].
    In US5449000 the bioimpedance is measured by a Kelvin bridge, then estimating the body composition from equations that depend linearly on the bioimpedance and anthropometric measurements of various body sections.
    US5788643 uses the module and the bioimpedance phase at the typical 50 kHz frequency for the calculation of body composition. They also use regression equations based on linear approximations. In contrast, in US6125297 a frequency of 25 kHz and linear regression equations are used to estimate total water, extracellular water, total blood volume and plasma volume.
    On other occasions, for the estimation of body composition, bioimpedance measures are used in multiple frequencies (MFBIA, from English Multi Frequency Bioelectrical Impedance Analysis). In US5449000 a body composition estimation model based on anthropometric equations using the bioimpedance module at different frequencies between 5 and 150 kHz is described.
    In US7783344, the multi-frequency bioimpedance module is also used, but in this case between 20 KHz and 100 KHz. Said device incorporates wireless communication capabilities.
    The multifrequency devices that use the Cale model are based on the bioimpedance spectroscopy (BIS) technique, which is used by the best performance devices on the market. At US6151523 at least four frequencies are used; and then the
    Cale model to extrapolate the resistance values at zero frequency and infinite frequency. It proposes equations based on the values of the impedances at the different frequencies, including zero and infinite frequency, weighted by height and a geometric term that depends on the measured body segment. It is not a portable device to the extent that it is powered by electric current and is connected to a computer, where it shows the results.
    The device described in US6532384 uses three frequencies. From the three bioimpedance values calculate the center and radius of the Cale model curve that fits the bioimpedance values. The zero frequency resistance, the infinite frequency resistance, the impedance and the frequency at the maximum imaginary value are used in the estimation of extracellular water, intracellular water, fat mass and fat free mass. This device is portable and battery powered. In addition, it has a user interface consisting of buttons and a screen.
    The measurement system described in US7917202 is composed of a multi-frequency bioimpedance measurement and acquisition unit from 1 kHz to 1000 kHz. Their models and equations allow to determine the overhydration of the subject under study.
    US7945317 details how they use at least 10 frequencies, although they recommend 50, in the range of 5 to 1000 KHz, preferably logarithmically distributed. The bioimpedance team includes a microprocessor programmed to develop at least part of the analysis procedures. Apply a regression equation obtained from a calibration study to inform the subject of his body composition.
    The invention described in US6615077 comprises the resources for the determination and monitoring of the resistivity of the body of renal patients and the determination of the dry weight or the desired hydration status of a patient undergoing dialysis. The measurement system consists of a unit of measurement and acquisition of bio-impedance mulli-frequency, for which the Xitron 4200s device (Xitron Technologies, United States) is proposed. Connected to the bioimpedance measurement unit, the system includes an electrical output that connects to the body segment, which consists of at least two current injector electrodes in the segment. The system can apply a single current frequency, or optionally multiple frequencies from 1 kHz to 1000 kHz. It also contains an electrical input for measuring the voltage produced and its transmission to the bioimpedance analysis unit.
    In US2006 / 0122540 the problem of estimating the dry weight of the renal patient is also addressed. In the determination of body composition the type of model used in the analysis has a special relevance.
    The base of US2007 / 0027402 is a circuit model formed by three branches in parallel: a path formed by a resistor and a series capacitor that models the path of the electric current through the muscle, another path also formed by a resistor and a condenser that models the passage through grease, and a last resistive path. Once the model parameters associated with the bioimpedance values of the measured body segment have been identified, the muscle mass is estimated from the resistance associated with the muscle through equations obtained by linear correlation with magnetic resonance imaging. The fat mass is obtained in an equivalent way, but from the resistance associated with the fat.
    In US8257280 another circuit model described by two parallel branches is described: one formed by a resistor and a series capacitor that models the path of the electric current through the grease, and another consisting of a resistor that represents the component of extracellular fluid
    In US2004 / 0171963, a method of estimating body composition that corrects bioelectrical impedance values is presented by using a variable that represents the ratio of intracellular fluids to extracellular fluids. As detailed in this document, this method reduces the influence of rapid changes in extracellular fluid distribution and estimates body water and body composition more accurately.
    In US2011 / 0275922 an improved electrical model is described capable of explaining, in a wide range of frequencies, the electrical properties by different proportions of body tissues.
    In the model described in US2008 / 0086058 intracellular water is determined from an intracellular resistance, to which each class of tissue, whether adipose or lean, contributes differently.
    The device described in US2008 / 0086058 is formed by a unit of measurement, for the calculation of extracellular and intracellular resistances, and an evaluation unit, for the calculation of body parameters.
    Algorithms are described in US2010 / 0081960 to determine the fraction of body water, and from it, lean mass and fat mass.
    Another aspect discussed in different documents is the location of the electrodes in the bioimpedance measurement system. In US2011 / 0275922 methods and apparatus for determining the muscle content, fat content, and / or extracellular fluid content of a segment of the subject's body are described, said segment being the entire body.
    In US2006 / 0122540, a segmental bioimpedance measurement system in the calf is proposed for continuous monitoring of body composition.
    In US7930021 a device for measuring body composition is detailed by means of a bioimpedance analysis, by means of electrodes arranged in the handle of the device, which is held by both hands of the subject.
    In US2011 / 0245712 a method for monitoring pulmonary edema by bioimpedance is described, in which the electrodes are located in the torso of the subject.
    The device and method described in patent EP2863794 are based on 50 kHz segmental bioimpedance measurements and weighted equations according to the geometric factors of various segments.
    In EP2752158 and W020121132516 the electrodes are located at the waist.
    Another aspect dealt with in the different documents referring to BIA devices for the estimation of body composition are communications. The portable device described in US2012 / 0035494 can communicate wirelessly with a remote center through an intermediate device (cell phone, internet, etc.).
    The devices described in W02012 / 132516 and EP2752158 also include wireless communication capabilities.
    Finally, it is worth highlighting some particular applications of the BIA analysis. In US5788643, the ratio between extracellular water and fat-free mass is applied to establish a diagnostic index in chronic congestive heart failure. The device described in US2015 / 0025353 estimates the composition of a body region by electrical impedance tomography.
    In the case of US2015 / 0164370, the body composition analysis device is part of a system for the controlled realization of peritoneal dialysis.
    US2014 / 0243699 describes a method for determining the overhydration of a patient and other body parameters from bioimpedance measures.
    The sensors and monitoring systems proposed in US2010 / 0081960 and US8406865 use bioimpedance measures in combination with water percentage measures to estimate body composition parameters. To determine the percentage of water use the spectrophotometric methods.
    DESCRIPTION OF THE INVENTION
    The present invention relates to a platform, portable and wireless, formed by a series of devices, preferably by three devices, which perform measurements of the
    module and the bioimpedance phase of a user in multiple configurable frequencies, and, from them, calculates an estimate of body composition and an assessment of nutritional status and hydration.
    The improvements provided by the invention are distributed processing that optimizes resources in a multi-user environment and provides greater customization capacity; a method of estimating body composition based on a three dispersion bioimpedance model, more precise and more robust in the face of disturbances, noise and parasitic effects; an efficient method for the quasi-analytical resolution of the parameters of the bioimpedance model presented; and capabilities for detecting alarm situations and integrating information into an e-system.
    Health.
    The advantages of the object of the invention are represented in the set of claims that accompany this description. Compared to other proposals, the advantages of the object of the invention are several: its robustness against parasitic effects
    both low and high frequency, as well as in phase; its simplicity and low computational load using a quasi-analytical solution that takes advantage of the particular characteristics of the frequency behavior of bioimpedance; its greater precision when approaching bioimpedance values, even in those cases in which some type of artifact alters the measurements.
    DESCRIPTION OF THE DRAWINGS
    To complement the description that is being made and in order to help a better understanding of the characteristics of the invention, according to a preferred example of practical implementation thereof, a set of drawings is attached as an integral part of said description. where, for illustrative and non-limiting purposes, the following has been represented:
    Figure 1.- Shows a diagram of the basic architecture of the "intelligent platform for
    the monitoring of the body composition and the assessment of the nutritional status and hydration of the user "and devices that compose it.
    Figure 2.- Shows a diagram of the basic architecture of the portable bioimpedance sensor.
    5 Figure 3.-Shows a diagram of the basic architecture of the personal monitoring device.
    Figure 4.-Shows a diagram of the basic architecture of the multi-user monitoring device.
    Figure 5.-Shows a diagram of platform configurations.
    Figure 6.- Shows a diagram of the architecture of the processing stage of the platform.
    Figure 7.- Illustrates the method of estimating body modeling parameters.
    Figure 8.- Illustrates the method of estimating the parameters of a dispersion model.
    PREFERRED EMBODIMENT OF THE INVENTION
    In a possible embodiment of a first aspect of the invention proposed herein
    25 shown in figure 1, there is an intelligent platform (1) for monitoring body composition and assessing the nutritional and hydration status of the user, which in a preferred embodiment comprises three devices: a portable bioimpedance sensor (2 ), a personal monitoring device (3) and a multi-user monitoring device (4).
    30 The portable bioimpedance sensor (2) is capable of measuring the module and the bioimpedance phase at multiple configurable frequencies based on the joint operation of the following modules integrated in said portable bioimpedance sensor (2) and which can be seen in Figure 2:
    • An injection module (12) intended to inject an electric current of configurable frequency and phase in the body or body section (8) to be measured through two electrodes (9).
    • A measuring module (13) intended to measure the amplitude of the voltage generated by the circulation of said current through two other electrodes (10).
    • A first computing module (14) designed to manage all data capture hardware and, for each frequency set, uses the voltage measured in two different phases to estimate the module and the impedance phase.
    • A first two-way wireless communications module (15), which allows the portable bioimpedance sensor (2) to receive commands for frequency settings and activation of measurements, and to send the results of the computing module.
    • A first data storage module (16), for the temporary storage of information in the event of communications failure, or for the persistent recording of bioimpedance measures.
    The information generated by the portable bioimpedance sensor (2) is transmitted wirelessly to the personal monitoring device (3), with which it maintains a two-way communication link. The measurement start time can be activated locally by means of a push button (17) on the portable bioimpedance sensor (2) or it can be activated remotely by sending a command from the personal monitoring device (3 ). Also, through another command, the temporary instants in which automatic bioimpedance measurements would be carried out could be previously configured.
    In a preferred embodiment of the invention the personal monitoring device
    (3) It is portable, although in other possible embodiments it can also be fixed installation.
    In said personal monitoring device (3), with higher capacities, both
    hardware as software, that the portable bioimpedance sensor (2), methods that allow estimating body composition are developed. It can also handle the processing and management of information from other portable sensors connected to it, which may be related to other variables of interest (respiratory rate, heart rate, ECG, temperature, pulse oximetry, activity, falls, glucose, etc. .). Its operation responds to the joint operation of the following modules, referring to Figure 3:
    • A second communications module (18) intended to establish two-way wireless with at least the portable bioimpedance sensor
    (2)
    • A second computing module (19) for distributed processing, on a personal level, to estimate body composition, the assessment of hydration and nutritional status. It also executes algorithms for the detection of alarm situations or to be considered in attention.
    • An interface module (20) for displaying and managing the information of the portable bioimpedance sensor (2) and the results of the second computing module (19) in a way adapted to the user: tactile (20.a), visual (20. b), auditory (20.c), voice (20.d), etc.
    • A third module of communications (21), wireless and bidirectional, of personal information, which in a mono-user environment will be with an external service provider (11), and in a multi-person environment with the multi-user monitoring device (4 ).
    • A second data storage module (22), for the temporary storage of personal information in case of communications failure, or for the persistent recording of body composition estimates.
    In the multi-user monitoring device (4) it is only present in a multi-user environment, forming a multi-person monitoring network. Said multi-user monitoring device (4), which can be portable or fixed, is configured to establish a two-way and wireless communications link with each or more personal monitoring devices (3) of the network, in addition to the link with the external provider of services (11). It also handles the management
    of the information of all monitored users, and can also perform a
    Additional level of processing on biomedical information, as well as the detection of alarm events or situations of interest. Its operation responds to the joint operation of the following modules, referring to Figure 4:
    • A fourth communications module (23), wireless and bidirectional, with one or more personal monitoring devices (3).
    • A third computing module (24) for distributed processing, at the multi-user level, for estimating body composition, assessing hydration and nutritional status and detecting alarm situations.
    • A second interface module (25) to display and manage in an adapted way the information and alarms of all monitored users: touch (25.a), visual (25.b), auditory (25.c), voice ( 25.d), etc.
    • A fourth communications module (26), wireless and bidirectional, with an external service provider (11).
    • A third data storage module (27), for storage
    Temporary information of multiple users in case of communications failure, or for the persistent registration of said information, which allows future access without the need for a remote connection to an external database.
    If an alarm event is detected, the interface includes warning means adapted (luminous, acoustic, vibrations, etc.) to the specific application of use of the platform (1). The specialized user could then deactivate or silence the alarm while managing and reviewing the information of the assisted user.
    The multi-user monitoring device (4) can also manage the information of all users in an autonomous way, including alarm management, establishing communications in a transparent way to the user with an external service provider (11) to integrate the information and the alarms provided by the platform (1) in an e-Health system.
    The intelligent platform (1) for monitoring body composition is of special relevance in the field of bioimpedance devices both in the
    Multi-level distribution of the devices involved (data collection, personal monitoring, multi-person monitoring), as well as their functionalities (management of sensory information, information management of one user, information management of multiple users).
    Likewise, the multilevel distribution of the processing favors energy saving and reduces the computational load, provides capabilities for the detection of alarm situations or to be considered in attention, as well as the integration of information in an e-Health system.
    The structural and functional modularity of the platform (1) facilitates the integration of the devices in special configurations of the platform (1), providing it with greater flexibility and customization capacity. In a particular configuration of the platform (1) shown in Figure 5, the portable bioimpedance sensor (2) can be integrated together with the personal monitoring device (3) in a first
    single hyperdispositive (5), may or may not coexist with the multi-user monitoring device (4). In this case, communications between the two can be made directly or wired (not wireless) or wireless. In addition, the portable bioimpedance sensor (2) and the personal monitoring device (3) can share physical components in the first hyper-device (5), in which they are integrated.
    In another particular configuration of the platform (1) also shown in Figure 5, the personal monitoring device (3) can be integrated next to the multi-user monitoring device (4) in a single second hyper-device (6). As in the previous case, communications between the two can be done directly (not wireless) or wireless. In addition, the personal monitoring device (3) and multi-user monitoring device (4) can also share physical components in the second hyper-device (6). As this configuration only makes sense in a multi-user environment, in the second hyper-device (6), the personal monitoring devices (3) of all users will coexist in parallel as abstract data processing entities.
    In another configuration of the platform (1), both the portable bioimpedance sensor
    (2), such as the personal monitoring device (3) and the multi-user monitoring device (4) are integrated into a single third hyper-device (7). In this case, bioimpedance measures can only be performed on a particular user at a given time. However, the third hyper-device (7) can be used in multiple users, through sequential or deferred management of the measures. As in the previous configurations, the communications between the portable bioimpedance sensor (2) and the personal monitoring device (3), on the one hand, and the communications between the personal monitoring device (3) and multi-user monitoring device (4 ) on the other, they can be carried out directly (not wirelessly) or wirelessly. All of them can also share physical components in the third hyper-device (7) as an integrator.
    In addition to the components and elements that make up the platform (1) it is also characterized by the methods used by the devices that comprise it for the estimation of body composition and the management of said platform.
    information.
    These methods form a second aspect of the invention and comprise the processing stage (28) of the platform (1), which can be distributed in different levels, as illustrated in Figure 6: a processing level in the portable sensor of bioimpedance (29), a level of processing in the personal monitoring device
    (30) and a processing level in the multi-user monitoring device (31).
    The intelligent platform (1) for the monitoring of body composition thus establishes a methodology and distributed processing architecture, which is advantageous at the level of computing and energy saving. At the computing level, because said multilevel structure allows to compensate the processing load between the different devices to avoid computational overload. At the energy level, because the higher energy consumption in portable devices is related to sending data wirelessly. How multilevel processing reduces and abstracts the
    wireless information to be transmitted, energy saving is thus favored.
    As Figure 6 illustrates, the processing stage (28) of the platform (1) is structured in the following processing modules: a first processing module (32) for the estimation of bioimpedance values, a second processing module (33) for the estimation of body modeling parameters, a third processing module (34) for the estimation of body composition and a fourth processing module (35) for alarm management. Depending on the needs, these processing modules can be executed locally on a single platform device (1), which can be the device (2), (3) or (4), or distributed in two
    or more devices of it.
    The first processing module (32) for the estimation of the bioimpedance values is responsible for managing the temporary instants in which the bioimpedance measures will be carried out. The processing of this module is distributed among the different devices of the platform (1). In the multi-user monitoring device (4), it is in charge of coordinating the performance of the bioimpedance measures of multiple users according to a pre-established plan, which can be configured by an expert user locally through the device interface or remotely through telematic services of the e-Health system. Based on the information provided by the multi-user monitoring device, the personal monitoring device is responsible for coordinating the bioimpedance measurements of a single user. This measure will be activated on the portable bioimpedance sensor (2) by sending a command. The portable bioimpedance sensor (2), the personal monitoring device (3) and the multi-user monitoring device (4) will maintain a real-time timing system to manage the instants in which the different measurements must be performed. A hierarchical procedure will be established from the multi-user monitoring device (4) to the portable bioimpedance sensor
    (2) based on the sending of commands for synchronization of the timing systems of: the portable bioimpedance sensor (2), the personal monitoring device (3) and the multi-user monitoring device (4). Different users, both experts and monitored users, can also activate the instantaneous performance of a bioimpedance, multi-or single frequency measurement. This instantaneous activation can be carried out from any of the interfaces of the devices (2,3,4) of the platform (1) (portable bioimpedance sensor (2), personal monitoring device (3) or multi-user monitoring device (4 "The first processing module (32) is also responsible for coordinating and activating the different frequencies that make up a bioimpedance measurement. In the portable bioimpedance sensor (2) and for each activated frequency, it takes care of the hardware control necessary to The generation of an electric current, with a defined phase, and the measurement of the voltage amplitude, also for each frequency is responsible for processing the voltage measured in two different phases to estimate the module and the impedance phase.
    The first processing module (32) provides the following results: for each measured frequency a set of three variables is generated, the first variable corresponding to the measured frequency, the second variable to the bioimpedance module and the third variable to the phase.
    The second processing module (33) for the estimation of the body modeling parameters is responsible for estimating the parameters of a current flow body model related to the bioimpedance values obtained by the first processing module (32). As bioimpedance measurements are made on a section of the human body, the model obtained incorporates properties that depend in part on the characteristics of the monitored tissues. The parameters of said model allow to abstract information related to the composition of the measured body section.
    In a preferred embodiment, a novel three dispersion model is proposed which results from the extension of the generic bioimpedance model (ColeCale model). According to this model, biological tissues can be considered formed by cells, which are separated by a conductive extracellular aqueous medium due to the presence of ions. As cell membranes have a low conductivity, they have a behavior similar to that of an electrical capacity.
    A simple model that collects this conduction phenomenon is represented by a parallel circuit in which one of the branches represents the path of the current through the extracellular medium and the other the path through the intracellular medium. The extracellular path is modeled through a resistance (Re, extracellular resistance) and the intracellular path through another resistance (Rí, intracellular resistance) in series with a capacity (Cm, membrane capacity). The impedance of this circuit can be expressed as:
    (Ec. 1)
    Where R ~ is the resistance that corresponds to bioimpedance at an infinite frequency, equivalent to the parallel of the extracellular Re and intracellular resistors R¡:
    R. · cR,
    (Ec. 2)
    Ro is the resistance that corresponds to bioimpedance at a zero frequency, which is equivalent to the extracellular resistance R &.
    R,
    (Ec. 3)
    w is the angular frequency, which is equivalent to:
    (j) 2. J [· = frequency
    (Ec. 4)
    and T represents the circuit time constant, which shows the dependence of the impedance with the frequency as a consequence of the capacitive dispersion of the membrane.
    (Ec. 5)
    If the real part of the bioimpedance is represented against an imaginary part in absolute value, the values obtained as a function of the frequency correspond to a semicircle in the first quadrant with center on the real axis in (Ro + R ...) / 2 And radio (Ro-R ..) / 2. The cut-off point furthest from the center of the semicircle with the real axis corresponds to the value R or, while the closest to the value R ". As the frequency increases from zero to infinity, all values are traversed. of the semicircle.
    An improved bioimpedance model (Cale-Cale model) includes the effects derived from the variability of cell membranes, which are not perfect electrical capacities and also have different shapes, characteristics and sizes in the different tissues that make up the human body . This effect can be modeled as the superposition of a multitude of dispersive effects, each with a different time constant. In the model, this translates into a displacement of the center of the semicircle below the real axis, which is mathematically expressed by the following equation:
    z
    Where a (O: s; a: s; 1) is the dispersion coefficient, which is a characteristic parameter that is related to the displacement of the center of the semicircle below the real axis:
    2 · 8
    a =
    7 [=
    (Ec.7)
    Being the angle that the real axis forms with the line that joins R .. with the center of the semicircle.
    In the present invention an extension of the bioimpedance model is described which includes a first low frequency dispersion and a third high frequency dispersion that allows modeling possible parasitic effects on the main dispersion (second dispersion). The model presented responds to the following expression:
    R + MI, + MI,
    oo: 1 {~ l = UF 1 (.), = ",
    + Ij · w · rfl- +: ¡"OJ-rr-
    in which, T; is a time constant of the dispersion i, ai is the dispersion coefficient i, and
    k 1,2,3
    (Ec. 10)
    This model is extended with a phase delay Td that increases linearly
    15 with frequency to more realistically model the effects derived from delays in the signals caused by electrodes, cables and hardware. Finally, the bioimpedance model proposed for the patent has the following
    expression:
    - "f1R, f1R, M,
    z -n + + +
    (= = 1+ (:) '. (O "",) l; oa, 1 + (: 1' (O "",) l; oa, 1 + (: 1 '(0 = 1'3)' to,
    (Ec.1 1) ) - .. ~ r
    - e d
    =
    The intelligent platform (1) for the monitoring of body composition uses this new model as an improvement of the bioimpedance spectroscopy (BIS) technique, which is used by the devices with better performance and greater precision.
    The second processing module (33) for the estimation of the body modeling parameters is responsible for identifying the parameters of the bioimpedance model proposed for the invention. To do this, it executes a method that minimizes the error between the bioimpedance values obtained and the values that would be obtained with the model at the corresponding frequencies. In commercial devices, this algorithm is usually an iterative process of successive approximations that may require a high processing time. The MP2 module (33) executes a novel method that allows to reduce the computational load of the process through a unique approach that takes advantage of the characteristics of the model itself. Said processing module can be executed locally on any of the devices of the platform (1), or it can be executed in a distributed manner by distributing the processing load between them.
    The biggest problem in the process of identifying parameters is the presence of some kind of disturbance, noise or parasitic effect. These effects can significantly affect the parameters obtained, leading to errors in the estimation of body composition. The vast majority of the algorithms proposed in the literature are applied to the Cale-Cale model of a single dispersion. In this context, the main novelty of the proposed procedure is a simple and robust method for the identification of the parameters, which can be applied to the proposed model of three dispersions, but also to the generic model. The two additional dispersions and the phase delay allow to include in the modeling external alterations as time constants different from those of the own body environment, although also possible internal alterations, both at low and high frequency, as well as the temporary delays caused by the Bioimpedance portable sensor hardware and cables.
    In an example of implementation of the object of the invention, the estimation of the body modeling parameters can be summarized in the following operations, referred to Figures 7 and 8:
    1) Operation 1.A (36): Reading of the N complex impedance values provided by the portable bioimpedance sensor (2) at N consecutive frequencies: Z¡, i = 1 ... N, i = 1 for the frequency lowest (tI) ei = N for the most
    high (fN).
    2) Operation 2.A (37): The basis of the processing is an iterative search of the solution to the model in a single parameter, the parameter Td (from T d.mir¡ to Td .rmu in increments of LlTd). The main advantage of this approach is that the search for the solution depends only on a single parameter, the Td parameter, so that the number of necessary iterations decreases significantly.
    3) Operation 3.A (38): For each value of Td, the bioimpedance values Z¡ are corrected according to the expression Zc, /, in which the index i is associated with the frequency (¡. Z Cl is the impedance that would be obtained if the influence of the delay modeled by Td was eliminated. Wi is the corresponding angular frequency (2 "1,).
    Iw. ·· T ..
    . and
    I w
    ZC, i z. ,
    (Ec. 12)
    4) Operation 4.A (39): The bioimpedance values are grouped into Ns sectors, which can overlap. The algorithm goes through the different sectors to find the best fit with the second dispersion, where s indicates the number of the sector.
    5) Operation 5.A (40): For each sector, estimation of the parameters of the model of a dispersion that best fits the corrected impedances Z C. i, which are proposed as possible parameters of the second dispersion.
    6) Operation 6.A (41): For each sector and if s is greater than 1, estimate the parameters of the model of a dispersion that best fits the remnant of low frequency bioimpedance, which are proposed as possible parameters of The first dispersion. The low frequency bioimpedance remainder is defined as the result of subtracting the bioimpedance values corresponding to the proposed parameters from the corrected impedances Z C.I
    for the second dispersion, considering only the lowest frequencies.
    7) Operation 7.A (42): For each sector and if s is less than Ns, estimate of
    5 parameters of a dispersion model that best fits the remainder ofhigh frequency bioimpedance, which are proposed as possibleparameters of the third dispersion. The bioimpedance remnant is definedhigh frequency as the result of subtracting the corrected impedances Z C.ithe bioimpedance values corresponding to the proposed parameters
    10 for the second dispersion, considering only the highest frequencies.
    8) Operation 8.A (43): For each sector, calculation of the mean square error between the measured impedance values and those obtained from the parameters proposed for the three dispersion model. If the error is the smallest of all
    In the cases analyzed above, the parameters evaluated are proposed as the optimal solution for the model up to that point. Finally, the algorithm loop closes with the evaluation of a new parameter Td again in step 2.A (37).
    20 The method of estimating the parameters of the dispersion model, which is applied in operations 5.A (40), 6.A (41) and 7A (42), can be summarized in the following operations, referring to the figure 8.
    1) Operation LB (44): The method is based on an iterative search of the
    The solution to the model, where N; ter is the number of iterations (a higher value in N i / er increases the probability of finding the optimal parameters, but also increases the execution time).
    2) Operation 2.B (45): The analyzed impedance values are grouped into
    30 each iteration in groups of three elements: one corresponding to low frequency (iP1 index), another to medium frequency (iP2 index) and the last one to high frequency (iP3 index). The values of these indices are preconfigured to scan all frequencies in a pseudo-random way,

    avoiding the repetition of triplets.
    3) Operation 3.8 (46): If the real part is represented against the absolute value of the imaginary part of the bioimpedances of each of the triplets, three points are obtained in the first quadrant. Once three points have been defined in the plane it is possible to determine a fourth point that is at the same distance from the other three. This property is used to calculate the radius Re and the coordinates (Creal, Cmag) corresponding to the center of the circle that passes through the three points, taking into account that the center corresponds to the fourth point and the radius with the distance mentioned.
    4) Operation 4.8 (47): For each impedance triplet the angle e that forms the real axis is calculated, at the cut-off point of the circumference closest to the origin, with the center of the circumference.
    5) Operation 5.8 (48): For each impedance triplet the values are calculated
    of the parameters a, Ro, R ", using the following equations:
    2E
    a = -
    1T = (Ec. 13)
    (Ec. 14) I ~ C ~ I - "e · ..ose · (Ec.1 5)
    6) Operation 6.8 (49): Once the semicircle of impedances is defined, each of the triplet impedances will define a time constant. The proposed time constant T will be the arithmetic mean of the three time constants. The value of the time constant is calculated from the following expressions:
    C, Re (Z) - = R = (Ec 16)
    C,
    (Ec. 17)
    to · Jr e, (2 · e¡ - €,) COS (-2-)
    (Ec. 18)
    (Ec. 19) - €, + ~ e3 '-04 · e¡. and,
    is 2 · e ¡
    (Ec.20)
    - ~
    T =
    (j) = (Ec. 21)
    7) Operation 7.9 (50): Calculation of the mean square error between the impedance values and those obtained from the parameters proposed for the dispersion model. If the error is the smallest so far, the parameters analyzed are proposed as the optimal model solution
    10 until that time. Finally, the loop closes with a new iteration in operation 1.8 (44).
    A third processing module (34) for the estimation of body composition is responsible for estimating the body composition, nutritional status and hydration of the user, based on the parameters of the model identified by the second processing module (33 ). The processing of this module is also distributed among the different devices of the platform. This module is also responsible for storing the information resulting from the estimation, its wireless delivery for integration into an e-Health system, as well as its
    20 correct presentation through the user interface, either in the personal monitoring device (3) or in the multi-user monitoring device (4). The user interface will be adapted to its characteristics, to provide an accessible means of accessing platform information based on the user's capabilities (visual, auditory, etc.).
    25 Another novel aspect of the platform is the possibility of configuration and customization, local or remote, of the algorithms used in the body estimation, depending on the user's own characteristics, their preferences or depending on the license available for their management, which can restrict, limit, adapt and / or customize the use of some algorithms. The algorithms can be stored locally on the platform for possible use, but they can also be updated remotely when some of the user characteristics are modified, when a customization of them is necessary
    or there are new improvements or updates in them. In the same estimate, one or more algorithms can be used, in which case the results will be complementary and the user interface will uniquely identify the source algorithm. The algorithms used in the estimation of body composition can be based on different methods of analysis:
    i. Quantitative Methods Basedinequations that depend on the
    values frombioimpedance,whoseparametersminimizeheerror
    Mean quadratic estimates in a baseline study.
    ii. Quantitative methods based on equations derived from the model
    employed in the second processing module (33), whose parameters
    they minimize the mean square error of the estimates in a study of
    reference.
    iii. Qualitative methods based ongraphic analysis ofvariables that
    They depend on bioimpedance.
    The variables of body composition can be referred to the whole body in the case of a global bioimpedance measure, or to a body section in the case of a segmental measure of the bioimpedance. Likewise, the parameters of these methods depend on the anthropometric and particular characteristics of the user such as age, height, perimeter of body sections, weight, sex, race or diseases suffered.
    The body composition variables that will be presented in the user interface can be configured and customized. The analysis method will define on a first level the variables that can be displayed. On a second level, the user (monitored or specialized) has the possibility to configure and customize the body composition variables to be displayed depending on the user's own characteristics
    monitored / specialized, your needs or your preferences. Some of the body composition variables considered are:
    - = Extracellular water volume.
    - = Intracellular water volume.
    _ = Total water volume.
    - = Water volume in excess or deficit.
    - = Total cell mass.
    - = Adipose tissue mass.
    _ = Lean tissue mass.
    - = Fat mass.
    - = Fat free mass.
    - Muscle mass
    - = And other indexes and variables that can be defined from the methods
    of body composition analysis.
    It is also possible to select the method of representation of the body composition variable in the user interface: text, graphic, auditory, etc. or a multiple selection of them. In addition, this proposal adds the possibility of selecting the user classification method, based on the results of estimation of body composition. The selected classification method will establish thresholds based on one or several body composition variables, which will allow the user to be classified at different levels, for example: high, normal, low. The thresholds, levels and the result of the classification will be shown in accordance with the representation method selected for the body composition variable (text, graphic, auditory, etc. or a multiple selection of them). The classification method assumes prior clinical knowledge and classification standards to provide direct information on the user's status and thus facilitate its evaluation and diagnosis.
    It is possible to include a qualitative analysis of the patient's condition based on a graphical representation in the plane that faces the fat mass index versus the corrected fat free mass index. This method allows the user status to be assessed based on the disposition of the point obtained in the graph with respect to characteristic regions defined from a reference population. The fat mass index is defined as the fat mass (kg) divided by the user's squared height (m2) and the corrected fat free mass index is the total weight minus the
    5 fat and excess water (kg) divided by the user's squared height (m2).
    It also considers the possibility of historical monitoring of the
    body composition variables in the different measures of a user. Saying
    historical record will be shown in a manner similar to the method of representation
    10 selected for the body composition variable (text, graphic, auditory, etc. or a multiple selection of them). In each of the measures, the date and time at which the measurement was made can be identified.
    The object of the invention may comprise an additional processing on the register 15 of the measures which aims to automatically establish trends, patterns and predictions in the history of the measures, which may be
    Notified to the user.
    The fourth processing module (35) is responsible for alarm management: this module
    20 integrates systems for the early detection of undesirable situations, which, if detected, would generate a series of alarms at local and remote levels that would allow a preventive action on the user. In said fourth processing module (35) there is a library of configurable indicators, locally or remotely, and a table with critical values for generating related alarms
    25 with these indicators. These indicators may be associated with a specific measure of bioimpedance, but also with an analysis of trends, patterns and predictions of the history of the measures. The logic and decision rules that govern the activation of alarms can also be configured to relate one or more of the indicators.
    权利要求:
    Claims (10)
    [1]
    1.-Intelligent platform (1) for monitoring body composition and
    assessment of nutritional status and hydration of a user, platform (1)
    characterized in that it comprises: - = a portable bioimpedance sensor (2) which in turn comprises: - = an injection module (12) intended to inject an electric current, of configurable frequency and phase through two electrodes, - = a measuring module (13) intended to measure the amplitude of the voltage generated by the circulation of said current through two other electrodes, - = a first computing module (14) intended to manage all data capture hardware and which, for each frequency set, uses the voltage measured in two different phases to estimate the module and the impedance phase. _ = a first communications module (15) that allows the portable bioimpedance sensor (2) to receive commands for the configuration of the frequencies and the activation of the measurements, and to send the results of the computing module, and - = a first module of data storage (16), for the temporary storage of information in case of communications failure, or for the persistent recording of bioimpedance measures, - = a personal monitoring device (3) which in turn comprises: _ = a second communications module (18) intended to establish two-way wireless communications with at least the portable bioimpedance sensor (2), - = a second computing module (19) for distributed processing, on a personal level, of estimating body composition, assessment of hydration and nutritional status, - = an interface module (20) to display and manage the information of the portable bioimpedance sensor (2) and results of the second computing module (19),
    e = a third communications module (21) of personal information, for
    establish two-way personal information communications, and e = a second data storage module (22), for the temporary storage of personal information, and e = a multi-user monitoring device (4) which in turn comprises:
    e = a fourth communications module (23), intended to establish communication with one or more personal monitoring devices (3), and a third computing module (24) for distributed processing of
    estimation of body composition, assessment of hydration and nutritional status and detection of alanna situations in a multi-user environment.
    and a second interface module (25) for displaying and managing in a way
    adapted the information and alarms of all monitored users, e = a fourth communications module (26) intended to establish communication with an external service provider (11), and
    e = a third data storage module (27), for the temporary storage of the information of the multiple users.
    [2]
    2.-Platafonna (1) according to claim 1, characterized in that the portable bioimpedance sensor (2) and the personal monitoring device (3) are integrated in a single first hyper-device (5).
    [3]
    3.-Platafonna (1) according to claim 1, characterized in that the personal monitoring device (3) and the multi-user monitoring device (4) are integrated into a single second hyper-device (6).
    [4]
    4. Platform (1) according to claim 1, characterized in that the portable sensor of
    Bioimpedance (2), the personal monitoring device (3) and the multi-user monitoring device (4) are integrated into a single third hyper-device (7).
    [5]
    5. Platform (1) according to any one of claims 1 to 4, characterized in that it additionally comprises:
    and =afirst moduleprocessing (32) fortheestimatefromthevaluesfrom
    bioimpedance,
    and =a second processing module (33) for the estimation of the parameters of
    body modeling,
    and =athirdmoduleprocessed (34)forthe estimatefromthe composition
    body, nutritional status and hydration of the user, and
    and =a fourth processing module (35) for alarm management.
    [6]
    6. Platform (1) according to any one of the preceding claims characterized in that it additionally comprises sensors for measuring variables that are selected from: respiratory rate, heart rate, ECG, temperature, activity, falls, glucose and pulse oximetry.
    [7]
    7. Method for estimating bioimpedance values using the platform (1) described in any one of claims 1 to 6, the method being characterized in that it comprises:
    and coordinate and activate the performance of bioimpedance measures, instantaneously or according to a pre-established plan configurable locally or remotely,
    and coordinate and activate impedance measurement frequencies,
    e for each frequency, control the hardware of the portable bioimpedance sensor for the generation of an electric current with a defined phase and the measurement of the voltage amplitude, and
    e for each frequency, process the voltage measured in two different phases for the estimation of the module and the impedance phase.
    [8]
    8. Method according to claim 7, characterized in that a three dispersion bioimpedance model defined by the following expression is used:
    z
    where R_ is the impedance at infinite frequency, ilR¡ is the increment of the modulus of
    dispersion impedance i (i = 1,2,3), w is the angular frequency, Ti is a time constant of the dispersion i, aj is the dispersion coefficient i and Td is the phase delay.
    [9]
    9. Method according to claim 7 or 8, characterized in that it further comprises carrying out:
    - = a reading of the N complex impedance values (36) provided by the portable bioimpedance sensor (2), in N consecutive frequencies: Z¡, i = I ... N, i = 1 for the lowest frequency ei = N for the highest,
    _ = an iterative search of the solution to the model in parameter 7d (37), - = a correction of the N bioimpedance values Z¡ (38) for each value of Td, according to the following expression:
    where COi is the angular frequency of the corresponding frequency (2 -1fj;) - = a grouping of the bioimpedance values that are grouped into Ns sectors (39), - = estimate, for each sector, parameters of the model of a dispersion that best fits the corrected impedances Z CJ which are proposed as possible parameters of the second dispersion (40), _ = estimate for each sector, provided that it is different from that of lower frequencies, model parameters of a dispersion that best conforms to the remnant of low frequency bioimpedance, which are proposed as possible parameters of the first dispersion (41), which comprises defining a remnant of low frequency bioimpedance as the result of subtracting the corrected impedances Z CJ from the values of bioimpedance corresponding to the parameters proposed for the second dispersion, considering only the lowest frequencies, _ = estimate for each sector, provided that it is different from the higher frequencies, parameters of a dispersion model that best fits the high frequency bioimpedance remnant, which are proposed as possible parameters of the third dispersion (42) which comprises defining a high bioimpedance remnant frequency as the result of subtracting the bioimpedance values corresponding to the parameters proposed for the second dispersion from the corrected impedances Z CJ, considering only the highest frequencies, and - = calculating for each sector an average square error (43) between values of
    5 impedance measured and those obtained from the parameters proposed for thethree dispersion model.
    [10]
    10. Method according to claim 8, characterized in that the estimation of the parameters of the model of a dispersion comprises: 10 _ = an iterative search of the solution to the model (44), according to:
    z
    where N, 'er is the number of iterations R .. is the resistance at infinite frequency, Ro is the resistance at zero frequency, w is the angular frequency, T is the time constant of the dispersion and a is the coefficient of
    15 dispersion
    - = select in each iteration three impedance values (45) in a pseudo-random way, _ = for each impedance triplet, representing the real part versus the absolute value of the imaginary part, calculate the radius Rc and the coordinates (C. ... J, C; mag) 20 (46) of the center of the circle that passes through the three impedances,
    • for each triplet, carry out an estimate of the angle O (47) that forms the real axis, at the cut-off point of the circumference closest to the origin, with the center of the circumference,
    _ for each triplet, calculate parameters a, R Ol R ", (48) according to the following 25 expressions: 2 · 8
    a = -
    1 [=
    类似技术:
    公开号 | 公开日 | 专利标题
    US20170340238A1|2017-11-30|Hydration status monitoring
    Khalil et al.2014|The theory and fundamentals of bioimpedance analysis in clinical status monitoring and diagnosis of diseases
    US10463273B2|2019-11-05|Hydration monitor
    ES2362218T3|2011-06-29|METHOD AND DEVICE FOR DETERMINING THE STATE OF HYDRATION AND / OR NUTRITION OF A PATIENT.
    Shiga et al.2009|A new simple measurement system of visceral fat accumulation by bioelectrical impedance analysis
    JPWO2010095709A1|2012-08-30|Biological information measuring device, biological information measuring method, and body composition measuring device
    CA2615845A1|2007-01-25|Index determination
    US20130172775A1|2013-07-04|Body condition information processing apparatus, non-transitory computer readable recording medium, and method for processing body condition information
    AU2017220381B2|2022-01-20|Body state classification
    ES2583755T3|2016-09-22|Method and device for determining an overhydration parameter or a body composition parameter
    KR20170111762A|2017-10-12|System for measuring body composition and body weight
    ES2682059B1|2019-09-13|SMART PLATFORM AND METHOD FOR MONITORING CORPORAL COMPOSITION AND THE VALUATION OF THE NUTRITIONAL STATE AND HYDRATION OF A USER
    Raharja et al.2018|Design of monitoring tool heartbeat rate and human body temperature based on WEB
    US20190357804A1|2019-11-28|Heart failure indicator
    Phillips et al.2021|A Deeper Understanding of Physics, Physiology, Experimental Methodology, and Statistics is Essential for Valid Comparison of USCOM 1A and cMRI
    Pierleoni et al.2021|Biological age estimation using an eHealth system based on wearable sensors
    JP4679993B2|2011-05-11|Trunk fat measurement method and apparatus
    JP6157067B2|2017-07-05|Health management device, health management program, and health management method
    KR20220030816A|2022-03-11|A health care system for toilet cover that can analyze body fat and method use the same
    US20210145303A1|2021-05-20|Indicator determination
    US20210022663A1|2021-01-28|Hydration assessment using a binary multispectral sensor
    WO2022025131A1|2022-02-03|Health level determination system, health level determination program, and health level detemination server
    Solovei et al.2012|The development of portable system for unobtrusive perspiration monitoring
    AU2006301927B2|2012-05-17|Hydration status monitoring
    Singhal et al.2014|GSM based heart beat bio monitoring system
    同族专利:
    公开号 | 公开日
    ES2682059R1|2018-09-26|
    ES2682059B1|2019-09-13|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    EP1217942A1|1999-09-24|2002-07-03|Healthetech, Inc.|Physiological monitor and associated computation, display and communication unit|
    EP2200512A1|2007-09-14|2010-06-30|Corventis, Inc.|Adherent device for respiratory monitoring and sleep disordered breathing|
    ES2537351B1|2013-11-04|2015-12-03|Universidad De Sevilla|Intelligent bioimpedance sensor for biomedical applications|
    US10548528B2|2015-08-07|2020-02-04|Ryan James Appleby|Smartphone device for body analysis|
    法律状态:
    2018-09-18| BA2A| Patent application published|Ref document number: 2682059 Country of ref document: ES Kind code of ref document: A2 Effective date: 20180918 |
    2018-09-26| EC2A| Search report published|Ref document number: 2682059 Country of ref document: ES Kind code of ref document: R1 Effective date: 20180919 |
    2019-09-13| FG2A| Definitive protection|Ref document number: 2682059 Country of ref document: ES Kind code of ref document: B1 Effective date: 20190913 |
    2020-03-20| PC2A| Transfer of patent|Owner name: UNIVERSIDAD DE SEVILLA Effective date: 20200316 |
    优先权:
    申请号 | 申请日 | 专利标题
    ES201730354A|ES2682059B1|2017-03-16|2017-03-16|SMART PLATFORM AND METHOD FOR MONITORING CORPORAL COMPOSITION AND THE VALUATION OF THE NUTRITIONAL STATE AND HYDRATION OF A USER|ES201730354A| ES2682059B1|2017-03-16|2017-03-16|SMART PLATFORM AND METHOD FOR MONITORING CORPORAL COMPOSITION AND THE VALUATION OF THE NUTRITIONAL STATE AND HYDRATION OF A USER|
    [返回顶部]