阅读说明：本技术 人体组织成分参数快速估算方法以及检测系统 (Method for rapidly estimating human tissue composition parameters and detection system ) 是由 徐梅钧 焦长平 刘芳 于 2020-11-26 设计创作，主要内容包括：本发明公开了一种人体组织成分参数快速估算方法以及检测系统,分别在5KHz以及100KHz下测得大致的人体细胞内外体液阻抗,进一步以此值为基础,基于无迹滤波算法对其进行多频率下的快速精确修正,最终实现人体成分参数值的快速精确估算,进一步为人体脂率、健康指数、潜在疾病分析等给出指导性建议。通过上述方式,本发明人体组织成分参数快速估算方法以及检测系统,能够通过检测的电位差信号进行基于小波模极大值思想的滤噪预处理,降低检测信号波动带来的干扰问题,最终实现了人体组织成分参数快速、简单的高精度鲁棒估算。(The invention discloses a method for quickly estimating human tissue composition parameters and a detection system, which are used for respectively measuring approximate human body cell internal and external body fluid impedance under the conditions of 5KHz and 100KHz, further based on the values, carrying out quick and accurate correction under multiple frequencies on the human body tissue composition parameters based on an unscented filter algorithm, finally realizing the quick and accurate estimation of the human body composition parameters, and further providing guiding suggestions for human body fat rate, health indexes, potential disease analysis and the like. By the mode, the method and the system for rapidly estimating the human tissue composition parameters can perform noise filtering pretreatment based on the idea of wavelet modulus maximum through the detected potential difference signals, reduce the interference problem caused by the fluctuation of the detected signals, and finally realize rapid, simple and high-precision robust estimation of the human tissue composition parameters.)

1. A method for rapidly estimating human tissue composition parameters is characterized by comprising the following steps:

a: selecting a human body individual as a detection object, placing an excitation electrode and a detection electrode at the end part of a limb of the human body, and detecting the tissue parameters of the human body by regarding the human body as an approximate cylinder;

b: the method comprises the following steps of selecting a variable-frequency constant-current source unit as a measuring electrode to be connected with an excitation electrode of a human body, generating variable-frequency constant-amplitude current which can be borne by the human body and has a frequency range of 1KHz to 100KHz by the variable-frequency constant-current source unit, and selecting a potential difference acquisition unit to be connected with a detection electrode of the human body and used for acquiring and measuring potential difference values at two ends;

c: denoising the acquired potential difference signals based on the wavelet modulus maximum principle through a data processing unit to obtain relatively smooth potential difference stable signals;

d: the potential difference stabilizing signals obtained by the data processing unit are measured by the human body component parameter initial value calculating unit under the action of constant current sources with different frequencies generated by the variable-frequency constant current source unit, so that the rough equivalent internal resistance values of the human body extracellular fluid and the total intracellular and extracellular fluids under the action of alternating current are obtained;

e: by using a human body component parameter online correction unit under multiple frequencies, human body tissue component parameters as state quantities, multiple frequency alternating constant currents as control quantities, potential difference values measured under multiple frequency alternating currents as observed quantities, and rough equivalent internal resistance values of extracellular fluid under the action of alternating currents and rough equivalent internal resistance values of total intracellular and extracellular fluids under the action of alternating currents, which are obtained by a human body component parameter initial value calculation unit, as initial values, an unscented Kalman filtering state estimator is provided for realizing online correction and pre-estimation of the human body tissue component parameters;

f: and various health indexes are given according to the parameters according to the corresponding human tissue composition parameters stored by the human information storage and analysis unit.

2. The method according to claim 1, wherein in step d, under the action of low-frequency and high-frequency constant current sources, corresponding potential difference values are measured respectively, so as to obtain rough equivalent internal resistance values of the extracellular fluid and the total intracellular and extracellular fluids of the human body under the action of the alternating current:

（1）

in the formula (1), the reaction mixture is,，are respectively asAlternating current and potential difference signal sequences at a frequency;represents time;the approximate equivalent internal resistance value of the extracellular fluid under the action of the alternating current,the roughly equivalent internal resistance value of the total fluid inside and outside the cell under the action of alternating current is shown; the calculating unit uses the rough internal resistance valueAndas an initial value for the unscented kalman filter estimator.

3. According to claim2, the method for rapidly estimating the human tissue composition parameters is characterized in that in the step e, the human tissue composition parameters are taken as state quantities, the multi-frequency alternating constant currents are taken as control quantities, the potential differences measured under the multi-frequency alternating currents are taken as observed quantities, and the human tissue composition parameters obtained by the human tissue composition parameter initial value calculation unit are taken as state quantitiesAndfor initial value, it is proposed to utilize unscented Kalman filtering state estimator to realize human tissue composition parametersOnline correction pre-estimation; the system of state equations is as follows:

（2）

in the formula (2), the reaction mixture is,is a time constant expressed by human tissue components under the action of alternating current;is a relaxation factor; the observation equation is:

（3）

in the formula (I), the compound is shown in the specification,according to the stateEstimate of a quantity and alternating currentCalculating an estimated value of the obtained potential difference; the computing unit then uses the observed potential differencePotential difference calculated from observation equationThe deviation amount of (2) realizes the state amountThe correction equation is as follows:

（4）

in the formula (I), the compound is shown in the specification,parameters for unscented Kalman Filter Algorithm, respectivelyIn thatOne-step estimation of the time;

using the above parametersThe parameters of the equivalent circuit model of the human tissue components can be calculated according to the equivalent circuit diagram 4 of the human tissue as follows:

（5）

（6）

（7）

in the formula (I), the compound is shown in the specification,is the total resistance value of the human tissue extracellular fluid;the total resistance value of the intracellular fluid of the human tissue;is the capacitance value of human cell membrane.

4. The method for rapidly estimating the parameters of the body tissue components according to claim 3, wherein the specific process of the online correction of the body component parameters in the step e is as follows:

A. sampling: generatingAnSet of points, whereinRepresenting the dimension of the state in the equation of state;

（8）

in the formula (8), the state variable，，Taken together areA dimension matrix;

to representIn the square matrixThe columns of the image data are,；，representing the distance of the set of sample points to the mean point,as a scale factor, take0 or；

B. One-step estimation: using the pair of formula (2)Point setPerforming one-step estimation to obtain a one-step estimation value of the state；

C. And (3) state statistics: computingOne-step estimation result of point setStatistic of (2):

（9）

（10）

（11）

wherein the content of the first and second substances,2 is taken when the normal distribution is satisfied;

D. heavy miningSample preparation: mean statistics using state one-step estimationIn alternative (8)Variance statistic in equation (10)Each column inIn alternative (8)To produce newPoint set;

E. And (3) calculation of an observed value: use of formula (3) and novelPoint setCalculating the observed value to obtain a frequency ofObserved value of；

F. And (3) observation statistics: computingStatistic of (2):

（12）

（13）

（14）；

G. updating estimation: updating state estimatesAnd variance；

（15）

（16）

（17）

In formula (16)I.e. based on only one sampling frequencyParameters of human tissue compositionThe on-line one-time correction is carried out, the current frequency of the variable-frequency constant current source is continuously changed, the sampling points in the process are increased, and finally the human tissue composition parameters are realizedAnd finally, obtaining the final human tissue parameter value according to the human tissue component equivalent circuit and the formulas (5) to (7).

5. A human tissue composition parameter detection system of the human tissue composition parameter rapid estimation method according to any one of claims 1 to 4, comprising a data acquisition module, a signal generation module, a calculation module and an information display module, wherein the data acquisition module comprises a potential difference acquisition unit, the signal generation module comprises a variable frequency constant current source, the calculation module comprises a data processing unit, a human composition parameter initial value calculation unit and a human composition parameter online correction unit, the information display module comprises a human information storage and analysis unit, the input end of the potential difference acquisition unit is connected with the human body, the output end of the potential difference acquisition unit is connected with the data processing unit, the output end of the variable frequency constant current unit is respectively connected with the human body and the data processing unit, the output end of the data processing unit is respectively connected with the human composition parameter initial value calculation unit and the input end of the human composition parameter online correction unit And the output end of the human body composition parameter online correction unit is connected with the input end of the human body information storage and analysis unit.

6. The system for detecting parameters of body tissue constituents according to claim 5, wherein said variable frequency constant current source is an exciting electrode installed on the sofa pedal and the armrest, said potential difference collecting unit is a measuring electrode installed on the sofa pedal and the armrest, and said human body forms a cylindrical conductor by attaching said exciting electrode and measuring motor to the hand and foot of said human body.

Technical Field

The invention relates to the field of detection of biological tissue composition parameters, in particular to a method and a system for quickly estimating human tissue composition parameters.

Background

Health problems become one of the concerns of people, and the rapid life and overload work lead to the gradual deterioration of human physique, which is particularly reflected in abnormal metabolism, hypertension, hyperlipidemia, insomnia, alopecia and the like, and even have potential disease troubles. Therefore, the daily monitoring of human health and the early warning of potential diseases can be realized through simple household health detection equipment in daily life, and the problem becomes a popular research problem.

In clinical medicine, diseases are usually caused by the balance of body fluid in human bodies, the imbalance of metabolism, the change of tissue parameters in human bodies and the like, and subtle changes of the diseases are discovered at early stage and are important for early warning of various disease latencies. The human body is an organism, so that the liquid occupies a considerable proportion, and the quantity of intracellular fluid and extracellular fluid determines the parameters of the components of the human body, such as protein, fat content and the like. In addition, the content of the liquid in the human body can also be known for the treatment of diseases, such as hemodialysis, cancer analysis and the like. The current methods for measuring the content of liquid in the human body are roughly classified into an isotope dilution method, a neutron activation analysis method, an infrared method, an ultrasonic method, a bio-impedance method, and the like. Among them, the isotope dilution method is considered as a gold standard, but it is difficult to realize in a domestic environment. The bioimpedance method is one of the most widely studied and applied human tissue component parameter analysis methods in the field of human health detection at home due to the advantages of simplicity, safety, rapidness, no wound and the like.

At present, foreign research on the bio-impedance measurement technology focuses on the aspects of system modeling, parameter extraction and system development, for example, a relational model between the total liquid quantity of a human body and the total impedance of the human body is established, and the rationality and feasibility of analyzing the components of the human tissue by using the bio-impedance technology are verified; analyzing the electrical characteristics of human cell membranes and the like under the currents of different frequency bands; and a plurality of sets of human tissue equivalent circuit models are successfully established.

The research of domestic scholars in this respect is more biased to the application, for example, the potential relation and the internal connection between the human tissue and cranial nerves, brain tumors, blood diseases and metabolic diseases are searched; by researching the variation of the parameters of the equivalent circuit model of the human tissue, the prediction, treatment, monitoring and the like of diseases are realized. However, the following problems still exist in the current research on the detection problem of the human tissue composition parameters: at present, most household human tissue component parameter analysis based on bioimpedance measurement is measurement under single-frequency excitation, and human parameter information is relatively single to extract, so that simple human sign analysis can only be realized; even if the system equipment for realizing the bio-impedance measurement under the multi-frequency excitation mostly utilizes a simple linear parameter estimation method to realize the identification of nonlinear strong complex function parameters of a human tissue equivalent circuit, the convergence speed is low, the time consumption is long, the precision is low, the system equipment is easily influenced by signal disturbance caused by problems of human body posture, skin humidity, dryness and the like, the robustness is poor, the user experience is not ideal, and if the health prediction and the potential disease analysis are carried out by using low-quality human tissue component parameter values, larger errors can be brought.

Disclosure of Invention

The invention mainly solves the technical problem of providing a method and a system for quickly estimating human tissue composition parameters, which can reduce the fluctuation interference problem of detected potential difference signals by detecting potential difference signals under multi-frequency current and carrying out noise filtering pretreatment based on the idea of wavelet modulus maximum, further realize quick and simple high-precision robust estimation of the human tissue composition parameters by using an unscented Kalman filtering algorithm, and finally realize quick and simple high-precision robust estimation of the human tissue composition parameters.

In order to solve the technical problems, the invention adopts a technical scheme that: the method for rapidly estimating the human tissue composition parameters comprises the following steps: a: selecting a human body individual as a detection object, placing an excitation electrode and a detection electrode at the end part of a limb of the human body, and detecting the tissue parameters of the human body by regarding the human body as an approximate cylinder; b: the method comprises the following steps of selecting a variable-frequency constant-current source unit as a measuring electrode to be connected with an excitation electrode of a human body, generating variable-frequency constant-amplitude current which can be borne by the human body and has a frequency range of 1KHz to 100KHz by the variable-frequency constant-current source unit, and selecting a potential difference acquisition unit to be connected with a detection electrode of the human body and used for acquiring and measuring potential difference values at two ends; c: denoising the acquired potential difference signals based on the wavelet modulus maximum principle through a data processing unit to obtain relatively smooth potential difference stable signals; d: the potential difference stabilizing signals obtained by the data processing unit are measured by the human body component parameter initial value calculating unit under the action of constant current sources with different frequencies generated by the variable-frequency constant current source unit, so that the rough equivalent internal resistance values of the human body extracellular fluid and the total intracellular and extracellular fluids under the action of alternating current are obtained; e: by using a human body component parameter online correction unit under multiple frequencies, human body tissue component parameters as state quantities, multiple frequency alternating constant currents as control quantities, potential difference values measured under multiple frequency alternating currents as observed quantities, and rough equivalent internal resistance values of extracellular fluid under the action of alternating currents and rough equivalent internal resistance values of total intracellular and extracellular fluids under the action of alternating currents, which are obtained by a human body component parameter initial value calculation unit, as initial values, an unscented Kalman filtering state estimator is provided for realizing online correction and pre-estimation of the human body tissue component parameters; f: and various health indexes are given according to the parameters according to the corresponding human tissue composition parameters stored by the human information storage and analysis unit.

In a preferred embodiment of the present invention, in the step d, under the action of low-frequency and high-frequency constant current sources, corresponding potential difference values are measured respectively, so as to obtain rough equivalent internal resistance values exhibited by the extracellular fluid and the total intracellular and extracellular fluids of the human body under the action of the alternating current:

（1）

in the formula (1), the reaction mixture is,，are respectively asAlternating current at frequency and potential difference signalA sequence of numbers;represents time;the approximate equivalent internal resistance value of the extracellular fluid under the action of the alternating current,the roughly equivalent internal resistance value of the total fluid inside and outside the cell under the action of alternating current is shown; the calculating unit uses the rough internal resistance valueAndas an initial value for the unscented kalman filter estimator.

In a preferred embodiment of the present invention, in the step e, the human body composition parameter online correction unit uses the human body tissue composition parameter as a state quantity, uses the multi-frequency alternating constant current as a control quantity, uses the potential difference measured under the multi-frequency alternating current as an observed quantity, and uses the human body composition parameter initial value obtained by the human body composition parameter initial value calculation unit as an observed quantity under the multi-frequency alternating constant currentAndfor initial value, it is proposed to utilize unscented Kalman filtering state estimator to realize human tissue composition parametersOnline correction pre-estimation; the system of state equations is as follows:

（2）

in the formula (2), the reaction mixture is,is a time constant expressed by human tissue components under the action of alternating current;is a relaxation factor; the observation equation is:

（3）

in the formula (I), the compound is shown in the specification,based on an estimate of the state quantity and the alternating currentCalculating an estimated value of the obtained potential difference; the computing unit then uses the observed potential differencePotential difference calculated from observation equationThe deviation amount of (2) realizes the state amountThe correction equation is as follows:

（4）

in the formula (I), the compound is shown in the specification,parameters for unscented Kalman Filter Algorithm, respectivelyIn thatOne-step estimation of the time;

using the above parametersThe parameters of the human tissue component equivalent circuit model can be calculated according to the human tissue equivalent circuit diagram as follows:

（5）

（6）

（7）

in the formula (I), the compound is shown in the specification,is the total resistance value of the human tissue extracellular fluid;the total resistance value of the intracellular fluid of the human tissue;is the capacitance value of human cell membrane.

In a preferred embodiment of the present invention, the specific process of online correction of the human body composition parameter in step e by the online correction unit is as follows:

A. sampling: generatingAnSet of points, whereinRepresenting the dimension of the state in the equation of state;

（8）

in the formula (8), the state variable，，Taken together areA dimension matrix.To representIn the square matrixThe columns of the image data are,；，representing the distance of the set of sample points to the mean point,is a scale factor, generally taken as 0 or；

B. One-step estimation: using the pair of formula (2)Point setPerforming one-step estimation to obtain a one-step estimation value of the state；

C. And (3) state statistics: computingOne-step estimation result of point setStatistic of (2):

（9）

（10）

（11）

wherein the content of the first and second substances,2 is taken when the normal distribution is satisfied;

D. resampling: mean statistics using state one-step estimationIn alternative (8)Variance statistic in equation (10)Each column inIn alternative (8)To produce newPoint set;

E. And (3) calculation of an observed value: use of formula (3) and novelPoint setCalculating the observed value to obtain a frequency ofObserved value of；

F. And (3) observation statistics: computingStatistic of (2):

（12）

（13）

（14）；

G. updating estimation: updating state estimatesAnd variance；

（15）

（16）

（17）

In formula (16)I.e. the parameters of the composition of the body tissue realized according to only one sampling frequencyThe on-line one-time correction is carried out, the current frequency of the variable-frequency constant current source is continuously changed, the sampling points in the process are increased, and finally the human tissue composition parameters are realizedAnd finally, obtaining the final human tissue parameter value according to the human tissue component equivalent circuit and the formulas (5) to (7).

In order to solve the technical problem, the invention adopts another technical scheme that: the human tissue composition parameter detection system comprises a data acquisition module, a signal generation module, a calculation module and an information display module, wherein the data acquisition module comprises a potential difference acquisition unit, the signal generation module comprises a variable-frequency constant-current source, the calculation module comprises a data processing unit, a human composition parameter initial value calculation unit and a human composition parameter online correction unit, the information display module comprises a human information storage and analysis unit, the input end of the potential difference acquisition unit is connected with a human body, the output end of the potential difference acquisition unit is connected with the data processing unit, the output end of the variable-frequency constant-current unit is respectively connected with the human body and the data processing unit, the output end of the data processing unit is respectively connected with the human composition parameter initial value calculation unit and the input end of the human composition parameter online correction unit, the output end of the human body composition parameter online correction unit is connected with the input end of the human body composition parameter storage and analysis unit

In a preferred embodiment of the invention, the variable-frequency constant-current source is an excitation electrode mounted on the sofa pedal and the armrest, the potential difference acquisition unit is a measurement electrode mounted on the sofa pedal and the armrest, and the excitation electrode and the measurement motor are correspondingly attached to the hand and the foot of the human body to form a cylindrical conductor.

The invention has the beneficial effects that: the invention relates to a fast estimation method and a detection system of human tissue component parameters based on an unscented filter algorithm. The actual rough measurement values of the human tissue parameters under relatively low frequency and high frequency are used as the initial values of the unscented Kalman filtering algorithm, so that the rapid convergence near the optimal value is realized, and the global optimal convergence of the nonlinear function is ensured. The method avoids the interference of low assignment and high-frequency potential difference signals of detection due to large disturbance such as noise, provides noise filtering pretreatment based on the wavelet modulus maximum thought on the detected potential difference signals, reduces the interference problem caused by the fluctuation of the detected signals, and finally realizes the rapid and simple high-precision robust estimation of the human tissue composition parameters.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without inventive efforts, wherein:

FIG. 1 is a schematic structural diagram of a method for rapidly estimating parameters of tissue constituents of a human body according to a preferred embodiment of the present invention;

FIG. 2 is a schematic diagram of an equivalent circuit of human tissue;

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Referring to fig. 1, a method for rapidly estimating parameters of tissue components of a human body includes the following steps:

a: selecting a human body individual as a detection object, placing an excitation electrode and a detection electrode at the end part of a limb of the human body, and detecting the tissue parameters of the human body by regarding the human body as an approximate cylinder;

b: the method comprises the following steps of selecting a variable-frequency constant-current source unit as a measuring electrode to be connected with an excitation electrode of a human body, generating variable-frequency constant-amplitude current which can be borne by the human body and has a frequency range of 1KHz to 100KHz by the variable-frequency constant-current source unit, and selecting a potential difference acquisition unit to be connected with a detection electrode of the human body and used for acquiring and measuring potential difference values at two ends;

c: denoising the acquired potential difference signals based on the wavelet modulus maximum principle through a data processing unit to obtain relatively smooth potential difference stable signals;

d: the potential difference stabilizing signals obtained by the data processing unit are measured by the human body component parameter initial value calculating unit under the action of constant current sources with different frequencies generated by the variable-frequency constant current source unit, so that the rough equivalent internal resistance values of the human body extracellular fluid and the total intracellular and extracellular fluids under the action of alternating current are obtained;

e: by using a human body component parameter online correction unit under multiple frequencies, human body tissue component parameters as state quantities, multiple frequency alternating constant currents as control quantities, potential difference values measured under multiple frequency alternating currents as observed quantities, and rough equivalent internal resistance values of extracellular fluid under the action of alternating currents and rough equivalent internal resistance values of total intracellular and extracellular fluids under the action of alternating currents, which are obtained by a human body component parameter initial value calculation unit, as initial values, an unscented Kalman filtering state estimator is provided for realizing online correction and pre-estimation of the human body tissue component parameters;

f: and various health indexes are given according to the parameters according to the corresponding human tissue composition parameters stored by the human information storage and analysis unit. Such as: human body fat rate, FFM and the like, and early warning analysis and the like are carried out on human body health degree and potential diseases; and meanwhile, displaying the analysis conclusion to the test user through APP.

In addition, in step d, under the action of the low-frequency and high-frequency constant current sources, the human tissue components mainly show resistance characteristics and capacitance characteristics, so that the cell membrane showing capacitance characteristics at low frequency can be approximately regarded as an open circuit, and at the moment, the current mainly flows through extracellular fluid; at high frequencies, the cell membrane, which exhibits capacitive properties, can be viewed approximately as a pathway, where current flows through both the extracellular and intracellular fluids. The calculating unit respectively measures corresponding potential difference values by utilizing the principle, and further obtains rough equivalent internal resistance values of the human extracellular fluid and the total intracellular and extracellular fluids under the action of alternating current:

（1）

in the formula (1), the reaction mixture is,，are respectively asAlternating current and potential difference signal sequences at a frequency;represents time;the approximate equivalent internal resistance value of the extracellular fluid under the action of the alternating current,the roughly equivalent internal resistance value of the total fluid inside and outside the cell under the action of alternating current is shown; the calculating unit uses the rough internal resistance valueAndas an initial value for the unscented kalman filter estimator.

In addition, in the step e, the human body composition parameter online correction unit takes the human body tissue composition parameters as state quantity under the multi-frequency alternating constant current, takes the multi-frequency alternating constant current as control quantity, takes the potential difference measured under the multi-frequency alternating current as observed quantity, and takes the human body composition parameter initial value obtained by the human body composition parameter initial value calculation unit as observation quantityAndfor initial value, it is proposed to utilize unscented Kalman filtering state estimator to realize human tissue composition parametersOnline correction pre-estimation; the system of state equations is as follows:

（2）

in the formula (2), the reaction mixture is,is a time constant expressed by human tissue components under the action of alternating current;is a relaxation factor; the observation equation is:

（3）

in the formula (I), the compound is shown in the specification,based on an estimate of the state quantity and the alternating currentCalculating an estimated value of the obtained potential difference; the computing unit then uses the observed potential differencePotential difference calculated from observation equationThe deviation amount of (2) realizes the state amountThe correction equation is as follows:

（4）

in the formula (I), the compound is shown in the specification,parameters for unscented Kalman Filter Algorithm, respectivelyIn thatOne-step estimation of the time;

using the above parametersThe parameters of the human tissue component equivalent circuit model can be calculated according to the human tissue equivalent circuit diagram as follows:

（5）

（6）

（7）

in the formula (I), the compound is shown in the specification,is the total resistance value of the human tissue extracellular fluid;the total resistance value of the intracellular fluid of the human tissue;is the capacitance value of human cell membrane.

In addition, the specific process of online correction of the human body composition parameter online correction unit in the step e is as follows:

A. sampling: generatingAnSet of points, whereinRepresenting the dimension of the state in the equation of state;

（8）

in the formula (8), the state variable，，Taken together areA dimension matrix.To representIn the square matrixThe columns of the image data are,；，representing the distance of the set of sample points to the mean point,is a scale factor, generally taken as 0 or；

B. One-step estimation: using the pair of formula (2)Point setPerforming one-step estimation to obtain a one-step estimation value of the state；

C. And (3) state statistics: computingOne-step estimation result of point setStatistic of (2):

（9）

（10）

（11）

wherein the content of the first and second substances,2 is taken when the normal distribution is satisfied;

D. resampling: mean statistics using state one-step estimationIn alternative (8)Variance statistic in equation (10)Each column inIn alternative (8)To produce newPoint set;

E. And (3) calculation of an observed value: use of formula (3) and novelPoint setCalculating the observed value to obtain a frequency ofObserved value of；

F. And (3) observation statistics: computingStatistic of (2):

（12）

（13）

（14）；

G. updating estimation: updating state estimatesAnd variance；

（15）

（16）

（17）

In formula (16)I.e. the parameters of the composition of the body tissue realized according to only one sampling frequencyThe on-line one-time correction is carried out, the current frequency of the variable-frequency constant current source is continuously changed, the sampling points in the process are increased, and finally the human tissue composition parameters are realizedAnd finally, obtaining the final human tissue parameter value according to the human tissue component equivalent circuit and the formulas (5) to (7).

A human tissue composition parameter detection system of a human tissue composition parameter rapid estimation method comprises a data acquisition module, a signal generation module, a calculation module and an information display module, wherein the data acquisition module comprises a potential difference acquisition unit, the signal generation module comprises a variable-frequency constant-current source, the calculation module comprises a data processing unit, a human composition parameter initial value calculation unit and a human composition parameter online correction unit, the information display module comprises a human information storage and analysis unit, the input end of the potential difference acquisition unit is connected with a human body, the output end of the potential difference acquisition unit is connected with the data processing unit, the output end of the variable-frequency constant-current unit is respectively connected with the human body and the data processing unit, the output end of the data processing unit is respectively connected with the human composition parameter initial value calculation unit and the input end of the human composition parameter online correction unit, the output end of the human body composition parameter online correction unit is connected with the input end of the human body composition parameter storage and analysis unit

In addition, the variable-frequency constant-current source is an excitation electrode arranged on a sofa pedal and a handrail, the potential difference acquisition unit is a measuring electrode arranged on the sofa pedal and the handrail, and the excitation electrode and the measuring motor are correspondingly attached to the hand and the foot of the individual human body to enable the individual human body to form a cylindrical conductor.

When a user has a rest on the intelligent sofa, the exciting electrodes arranged on the sofa pedals and the armrests can generate variable-frequency alternating constant-value current, the electrodes can send alternating constant current to a human body, the alternating current flows through the human body, and the measuring electrodes arranged on the sofa pedals and the armrests detect corresponding potential differences. The right hand part and the left foot part of the human body are respectively provided with an excitation electrode and a detection electrode, so that the total number of the electrode plates is 4, and the human body is considered as an approximate cylinder to detect the tissue parameters of the human body. When the frequency conversion current signal is introduced, the potential difference signal is acquired as follows:the potential difference signal and the alternating current signal generated by the variable-frequency constant-current source unit are simultaneously transmitted to the data processing unit, and the data processing unit performs wavelet-based mode on the alternating current signalThe denoising process of the large-value principle is free of noiseTime potential difference signal。

The process is firstly carried out under low-frequency 5KHz alternating current, and potential difference signals acquired at the same time are as follows:further obtaining the rough resistance value under low frequency for the denoised potential difference signal according to the formula (1)。

Further repeating the above process under high frequency 100KHz alternating current, and collectingFurther obtaining the rough resistance value under high frequency for the denoised potential difference signal according to the formula (1)。

Further, the rough impedance value under the low frequency and the high frequency is used as an initial value of the state of the unscented Kalman filtering algorithm under the frequency conversionAnd carrying out online correction on the human tissue composition parameters, wherein the process is finished in the human composition parameter online correction unit.

The specific process of the unscented kalman filter algorithm online correction is given here as follows:

sampling-generatingAnSet of points (Representing the dimension of a state in an equation of state, in the present invention）；

（8）

In the formula (10), the state variable，，Taken together areA dimension matrix.To representIn the square matrixThe columns of the image data are,；，representing the distance of the set of sample points to the mean point,is a scale factor, generally taken as 0 or。

One-step estimation-using the pair of formula (2)Point setPerforming one-step estimation to obtain a one-step estimation value of the state；

And step 3: state statistics-computationOne-step estimation result of point setStatistic of (2):

（9）

（10）

（11）

wherein the content of the first and second substances,satisfy normal distribution and take 2.

Resampling-mean statistics using state one-step estimationIn alternative (8)Variance statistic in equation (10)Each column inIn alternative (8)To produce newPoint set。

Observation calculation-using formula (3) and newPoint setCalculating the observed value to obtain a frequency ofObserved value of；

Observation statistics-calculationStatistic of (2):

（12）

（13）

（14）

and 7: update estimation-update state estimationAnd variance。

（15）

（16）

（17）

In formula (16)I.e. the parameters of the composition of the body tissue realized according to only one sampling frequencyThe on-line one-time correction is carried out, the current frequency of the variable-frequency constant current source is continuously changed, the sampling points in the process are increased, and finally the human tissue composition parameters are realizedAnd (4) accurate correction.

Finally, obtaining the final human tissue parameter value according to the human tissue component equivalent circuit schematic diagram in fig. 2 and the formulas (5) to (7);

different from the prior art, the method and the detection system for rapidly estimating the human tissue composition parameters based on the unscented filter algorithm realize the high-precision robust estimation of the human tissue composition parameters by utilizing the closed-loop self-correction structure of the unscented Kalman filter algorithm and the unbiased estimation capability of the unscented Kalman filter algorithm on the state parameters of the nonlinear model for more than two orders. The actual rough measurement values of the human tissue parameters under relatively low frequency and high frequency are used as the initial values of the unscented Kalman filtering algorithm, so that the rapid convergence near the optimal value is realized, and the global optimal convergence of the nonlinear function is ensured. The method avoids the interference of low assignment and high-frequency potential difference signals of detection due to large disturbance such as noise, provides noise filtering pretreatment based on the wavelet modulus maximum thought on the detected potential difference signals, reduces the interference problem caused by the fluctuation of the detected signals, and finally realizes the rapid and simple high-precision robust estimation of the human tissue composition parameters.

The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by the present specification, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.