CN106295179A - A kind of influenza Forecasting Methodology and device - Google Patents

Abstract

The invention discloses a kind of influenza Forecasting Methodology and device, described Forecasting Methodology sets up influenza forecast model with the health data of human body, the health data being based on includes heart rate, body temperature, contraction pressure, diastolic pressure and sex, and by a number of somatic data of random acquisition, solve based on the sample data gathered to obtain and forecast model respectively affects coefficient, determine the expression formula of forecast model.When being predicted, the health data of personnel to be measured is substituted in forecast model, then can calculate acquisition personnel to be measured and suffer from grippal probability, it be suffered from influenza and makes prediction.Influenza Forecasting Methodology of the present invention and device, suffer from influenza according to human heart rate, body temperature, contraction pressure, diastolic pressure and sex etc. to human body and make prediction, and enables people to popularity flu and makes prevention in time.

Description

A kind of influenza Forecasting Methodology and device

Technical field

The present invention relates to medical statistics and technical field of data processing, particularly relate to a kind of influenza Forecasting Methodology And device.

Background technology

Influenza is a kind of common disease, and common influenza shows as fear of cold high heat, and body temperature reaches as high as 39 DEG C～40 DEG C, and with the symptom such as increased heart rate, slight Hypertension.

Common people can arrive hospital and go to a doctor when feeling not quite the thing, but now patient has had manifest symptom, the most just Being to say that now patient has confirmed that to catch a cold, medical worker can only provide corresponding treatment means according to the concrete patient's condition of patient, as eaten Medicine, having an injection, therefore patient still can experience ill process and the therapeutic process of pain, still influences whether that people's is normal raw Live, or delay work.

Based on this, it is provided that a kind of effective influenza Forecasting Methodology, there is important value and significance.

Summary of the invention

It is an object of the invention to provide a kind of influenza Forecasting Methodology and device, according to human heart rate, body temperature, contraction Human body is suffered from influenza and is made prediction by pressure, diastolic pressure and sex etc., allows one to timely prevention of catching a cold.

For achieving the above object, the present invention provides following technical scheme:

A kind of influenza Forecasting Methodology, including:

Setting up influenza forecast model with the health data of human body, described health data includes heart rate, body temperature, contraction Pressure, diastolic pressure and sex, described influenza forecast model uses equation below to describe:

$P=\frac{e^{{\mathrm{\β}}_0+{\mathrm{\β}}_1{x}_1+{\mathrm{\β}}_2{x}_2+{\mathrm{\β}}_3{x}_3+{\mathrm{\β}}_4{x}_4+{\mathrm{\β}}_5{x}_5}}{1+e^{{\mathrm{\β}}_0+{\mathrm{\β}}_1{x}_1+{\mathrm{\β}}_2{x}_2+{\mathrm{\β}}_3{x}_3+{\mathrm{\β}}_4{x}_4+{\mathrm{\β}}_5{x}_5}},$

Wherein, P is for suffering from grippal probability, β₀For constant, β₁For Heart rate influences coefficient, β₂For temperature influence coefficient, β₃Coefficient, β is affected for shrinking pressure₄Coefficient, β is affected for diastolic pressure₅For Effect of gender coefficient, variable x₁For heart rate value, variable x₂For Body temperature value, variable x₃For shrinking pressure value, variable x₄For diastolic blood pressure values, variable x₅For sex；

Random acquisition m group somatic data, described data are with (x₁, x₂, x₃, x₄, x₅, y) describe, ask based on described m group data Solving to obtain and respectively affect coefficient in described forecast model, wherein y represents the disease state of gathered person；

The health data of personnel to be measured is substituted in described forecast model, calculate and obtain predicting the outcome of personnel to be measured.

Alternatively, chaos APSO algorithm is used to solve to obtain described forecast model respectively affects coefficient.

Alternatively, described random acquisition m group somatic data, described data are with (x₁, x₂, x₃, x₄, x₅, y) describe, based on institute State m group data to solve to obtain described forecast model respectively affects coefficient, including:

S201: gather m group somatic data, with (x_i1, x_i2, x_i3, x_i4, x_i5, y_i) describe, i=1,2 ..., m；

S202: setup control parameter, including setting population size N, maximum iteration time K_max, wherein N, K_maxIt is and is more than The positive integer of zero；

Variable declarations, including: current iteration number of times k, current optimal value P of i-th particle_besti, current global optimum G_best, inertia weight ω, Studying factors c1 and c2, random number ξ, η, wherein k is the positive integer more than zero；

Particle encodes, and affects factor beta to each₀、β₁、β₂、β₃、β₄、β₅Encoding, the coding of i-th particle includes position Coding and velocity encoded cine, position encoded for β_i=(β_i0, β_i1, β_i2, β_i3, β_i4, β_i5), velocity encoded cine is v_i=(v_i0, v_i1, v_i2, v_i3, v_i4, v_i5)；

Definition fitness function f (β), wherein β=(β₀, β₁, β₂, β₃, β₄, β₅) it is particle position；

S203: initialize, specifically includes: make k=0, it is thus achieved that the initial position β of i-th particle_i ⁽⁰⁾=(β_i0 ⁽⁰⁾, β_i1 ⁽⁰⁾, β_i2 ⁽⁰⁾, β_i3 ⁽⁰⁾, β_i4 ⁽⁰⁾, β_i5 ⁽⁰⁾), initial velocity v_i=(v_i0 ⁽⁰⁾, v_i1 ⁽⁰⁾, v_i2 ⁽⁰⁾, v_i3 ⁽⁰⁾, v_i4 ⁽⁰⁾, v_i5 ⁽⁰⁾), wherein β_ij ⁽⁰⁾、v_ij ⁽⁰⁾It is the random number of [-1000,1000], j=1,2,3,4,5；

S204: run iteration, for all i=1,2 ..., N, update speed and the position of i-th particle, for kth time Iteration, if f is (β_i ^(k))>P_besti, then P is made_besti=f (β_i ^(k)), if max{P_best1, P_best2..., P_bestN}>G_best, then make G_best=max{P_best1, P_best2..., P_bestN}；

S205: work as k=K_maxTime, export G_bestAnd the particle position β=(β of correspondence₀, β₁, β₂, β₃, β₄, β₅)。

Alternatively, described step S204 also includes: iteration secondary for kth, calculating population's fitness:

${\mathrm{\σ}}^2=\frac{1}{N}\underset{i=1}{\overset{N}{\Σ}}{\left(\frac{f\left({{\mathrm{\β}}_i}^{\left(k\right)}\right)-\frac{1}{N}\underset{i=1}{\overset{N}{\Σ}}f\left({{\mathrm{\β}}_i}^{\left(k\right)}\right)}{f}\right)}^2,$

f(β_i ^(k)) representing the i-th particle fitness when iterations k, f is fitness evaluation value, is specifically described as:

$f=\max \{1,\max \{|f\left({{\mathrm{\β}}_i}^{\left(k\right)}\right)-\frac{1}{N}\underset{i=1}{\overset{N}{\Σ}}f\left({{\mathrm{\β}}_i}^{\left(k\right)}\right)|\left\}\right\};$

Judge whether σ²＜ σ² _minIf otherwise running+1 iteration of kth, σ² _minFor default minimum population fitness, for just Number.

Alternatively, if σ²＜ σ² _min, then judge that population enters Premature Convergence state, is carried out the particle that fitness is the highest Random disturbance, specifically includes:

One original chaotic vector of S300: stochastic generation, is described as z₀=(z₀₀,z₀₁,z₀₂,z₀₃,z₀₄,z₀₅), wherein z₀₀、 z₀₁、z₀₂、z₀₃、z₀₄、z₀₅Span be [0,1]；

S301: chaos iteration generates Q chaos vector, and the l vector description is z_l=(z_l0,z_l1,z_l2,z_l3,z_l4, z_l5), l=1,2, ..., Q, 0 < Q < N, Q is positive integer；

S302: produce Q particle, the l particle is described as β_l=(β_l0,β_l1,β_l2,β_l3,β_l4,β_l5)；

S303: the particle that fitness is the highest is carried out random disturbance, is described as:

${z_l}^{\&prime;}=|\frac{e^{\frac{l}{Q}}-1}{e-1}-\frac{1}{l}{|}^{\frac{3}{2}}{z}^{*}+(1-|\frac{e^{\frac{l}{Q}}-1}{e-1}-\frac{1}{l}{|}^{\frac{3}{2}})z_l;$

Wherein, z_l' for applying the chaos vector that (β 0, β 1, β 2, β 3, β 4, β 5) is corresponding after random disturbance, z^*For optimal value β^*=(β₀ ^*, β₁ ^*, β₂ ^*, β₃ ^*, β₄ ^*, β₅ ^*) it is mapped to the corresponding vector that [0,1] is formed afterwards, z_lFor the chaos vector after iteration l time, l For chaos iteration number of times, Q is maximum chaos iteration number of times, and z* is specifically described as:

$z^{*}=\frac{{\mathrm{\&beta;}}^{*}-{\mathrm{\&beta;}}_{min}}{{\mathrm{\&beta;}}_{\max }-{\mathrm{\&beta;}}_{min}};$

β_maxFor the maximum of particle coding, β_minMinima for particle coding.

A kind of influenza prediction means, including:

Information acquisition module, including:

Heart rate sensor, for gathering the heart rate information of personnel to be measured；

Pressure transducer, for gathering contraction pressure and the diastolic pressure of personnel to be measured；

Body temperature trans, for gathering the body temperature of personnel to be measured；

User operation module, for receiving the gender information of the personnel to be measured of user's input；

Data analysis module, for the health data according to the personnel to be measured gathered, health data includes heart rate, body Temperature, contraction pressure, diastolic pressure and sex, pass through set up influenza forecast model calculating acquisition personnel to be measured and suffer from popularity The probability of flu, described influenza forecast model describes with equation below:

$P=\frac{e^{{\mathrm{\&beta;}}_0+{\mathrm{\&beta;}}_1{x}_1+{\mathrm{\&beta;}}_2{x}_2+{\mathrm{\&beta;}}_3{x}_3+{\mathrm{\&beta;}}_4{x}_4+{\mathrm{\&beta;}}_5{x}_5}}{1+e^{{\mathrm{\&beta;}}_0+{\mathrm{\&beta;}}_1{x}_1+{\mathrm{\&beta;}}_2{x}_2+{\mathrm{\&beta;}}_3{x}_3+{\mathrm{\&beta;}}_4{x}_4+{\mathrm{\&beta;}}_5{x}_5}},$

Wherein, P is for suffering from grippal probability, β₀For constant, β₁For Heart rate influences coefficient, β₂For temperature influence coefficient, β₃Coefficient, β is affected for shrinking pressure₄Coefficient, β is affected for diastolic pressure₅For Effect of gender coefficient, variable x₁For heart rate value, variable x₂For Body temperature value, variable x₃For shrinking pressure value, variable x₄For diastolic blood pressure values, variable x₅For sex.

Alternatively, described user operation module is additionally operable to output and display predicts the outcome.

Alternatively, described user operation module is additionally operable to receive and record the personal information of the personnel to be measured of user's input, Including name, sex, age and detection record.

By technique scheme it can be seen that a kind of influenza Forecasting Methodology provided by the present invention, with human body Health data sets up influenza forecast model, the health data being based on include heart rate, body temperature, contraction pressure, diastolic pressure and Sex, then gathers a number of sample data, i.e. random acquisition a number of individual health data, and often group data include The heart rate of gathered person, body temperature, contraction pressure, diastolic pressure, sex and disease state, based on sample data solve in model each Affect coefficient, determine the expression formula of forecast model.When predicting influenza, the health data of personnel to be measured is substituted into described In forecast model, so can calculate acquisition personnel to be measured suffer from grippal probability, it is suffered from influenza and makes prediction.

Influenza Forecasting Methodology of the present invention, according to human heart rate, body temperature, contraction pressure, diastolic pressure and sex etc. to people Body is suffered from influenza and is made prediction, and obtains P, enables people to popularity flu and makes prevention in time.

Accompanying drawing explanation

In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing In having technology to describe, the required accompanying drawing used is briefly described, it should be apparent that, the accompanying drawing in describing below is only this Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, it is also possible to Other accompanying drawing is obtained according to these accompanying drawings.

The flow chart of a kind of influenza Forecasting Methodology that Fig. 1 provides for the embodiment of the present invention；

The each method flow diagram affecting coefficient solving in forecast model that Fig. 2 provides for the embodiment of the present invention；

The schematic diagram of a kind of influenza prediction means that Fig. 3 provides for the embodiment of the present invention.

Detailed description of the invention

For the technical scheme making those skilled in the art be more fully understood that in the present invention, real below in conjunction with the present invention Execute the accompanying drawing in example, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described enforcement Example is only a part of embodiment of the present invention rather than whole embodiments.Based on the embodiment in the present invention, this area is common The every other embodiment that technical staff is obtained under not making creative work premise, all should belong to present invention protection Scope.

A kind of influenza Forecasting Methodology that the embodiment of the present invention provides, including:

Setting up influenza forecast model with the health data of human body, described health data includes heart rate, body temperature, contraction Pressure, diastolic pressure and sex, described influenza forecast model uses equation below to describe:

$P=\frac{e^{{\mathrm{\&beta;}}_0+{\mathrm{\&beta;}}_1{x}_1+{\mathrm{\&beta;}}_2{x}_2+{\mathrm{\&beta;}}_3{x}_3+{\mathrm{\&beta;}}_4{x}_4+{\mathrm{\&beta;}}_5{x}_5}}{1+e^{{\mathrm{\&beta;}}_0+{\mathrm{\&beta;}}_1{x}_1+{\mathrm{\&beta;}}_2{x}_2+{\mathrm{\&beta;}}_3{x}_3+{\mathrm{\&beta;}}_4{x}_4+{\mathrm{\&beta;}}_5{x}_5}},$

Wherein, P is for suffering from grippal probability, β₀For constant, β₁For Heart rate influences coefficient, β₂For temperature influence coefficient, β₃Coefficient, β is affected for shrinking pressure₄Coefficient, β is affected for diastolic pressure₅For Effect of gender coefficient, variable x₁For heart rate value, variable x₂For Body temperature value, variable x₃For shrinking pressure value, variable x₄For diastolic blood pressure values, variable x₅For sex；

Random acquisition m group somatic data, described data are with (x₁, x₂, x₃, x₄, x₅, y) describe, ask based on described m group data Solving to obtain and respectively affect coefficient in described forecast model, wherein y represents the disease state of gathered person；

The health data of personnel to be measured is substituted in described forecast model, calculate and obtain predicting the outcome of personnel to be measured.

By foregoing it can be seen that the embodiment of the present invention provide influenza Forecasting Methodology, with the health of human body Data set up influenza forecast model, and the health data being based on includes heart rate, body temperature, contraction pressure, diastolic pressure and sex, Then gathering a number of sample data, i.e. random acquisition a number of individual health data, often group data include being adopted The heart rate of collection person, body temperature, contraction pressure, diastolic pressure, sex and disease state, solve each impact in model based on sample data Coefficient, determines the expression formula of forecast model.When predicting influenza, the health data of personnel to be measured is substituted into described prediction In model, so can calculate acquisition personnel to be measured suffer from grippal probability, it is suffered from influenza and makes prediction.

Influenza Forecasting Methodology of the present invention, according to human heart rate, body temperature, contraction pressure, diastolic pressure and sex etc. to people Body is suffered from influenza and is made prediction, and obtains P, enables people to popularity flu and makes prevention in time.

Below influenza Forecasting Methodology of the present invention is described in detail.

Refer to Fig. 1, for the flow chart of a kind of influenza Forecasting Methodology that the embodiment of the present invention provides, this prediction side Method includes:

S1: set up influenza forecast model with the health data of human body, described health data include heart rate, body temperature, Shrinking pressure, diastolic pressure and sex, described influenza forecast model uses equation below to describe:

$P=\frac{e^{{\mathrm{\&beta;}}_0+{\mathrm{\&beta;}}_1{x}_1+{\mathrm{\&beta;}}_2{x}_2+{\mathrm{\&beta;}}_3{x}_3+{\mathrm{\&beta;}}_4{x}_4+{\mathrm{\&beta;}}_5{x}_5}}{1+e^{{\mathrm{\&beta;}}_0+{\mathrm{\&beta;}}_1{x}_1+{\mathrm{\&beta;}}_2{x}_2+{\mathrm{\&beta;}}_3{x}_3+{\mathrm{\&beta;}}_4{x}_4+{\mathrm{\&beta;}}_5{x}_5}},$

Wherein, P is for suffering from grippal probability, β₀For constant, β₁For Heart rate influences coefficient, β₂For temperature influence coefficient, β₃Coefficient, β is affected for shrinking pressure₄Coefficient, β is affected for diastolic pressure₅For Effect of gender coefficient, variable x₁For heart rate value, variable x₂For Body temperature value, variable x₃For shrinking pressure value, variable x₄For diastolic blood pressure values, variable x₅For sex.

In the present embodiment, β₀、β₁、β₂、β₃、β₄、β₅Span be [-1000,1000].Wherein, variable x₁Unit: Beat/min, span: x₁∈ [0,200], variable x₂Unit: DEG C, span: x₂∈ [0,50], variable x₃Unit: mmHg, Span: x₃∈ [0,200], variable x₄Unit: mmHg, span: x₄∈ [0,200], for variable x₅, x₅=0 represents Male, x₅=1 represents women.

S2: random acquisition m group somatic data, described data are with (x₁, x₂, x₃, x₄, x₅, y) describe, based on described m group data Solving to obtain and respectively affect coefficient in described forecast model, wherein y represents the disease state of gathered person.

In the present embodiment, the chaos APSO algorithm that uses when respectively affecting coefficient in solving forecast model solves, Refer to Fig. 2, each method flow diagram affecting coefficient solving in forecast model provided for the present embodiment, this method for solving bag Include following steps:

S201: gather m group somatic data, with (x_i1, x_i2, x_i3, x_i4, x_i5, y_i) describe, i=1,2 ..., m；

The quantity certain condition to be met of institute's collecting sample data, to ensure that the model solving acquisition affects the accurate of coefficient Degree.M >=5000 in the present embodiment, m is positive integer.

Heart rate value x_i1Unit is beat/min, span [0,200], body temperature value x_i2Unit is DEG C, span [0,50], Shrink pressure value x_i3Unit is mmHg, span [0,200], diastolic blood pressure values x_i4Unit is mmHg, span [0,200], x_i5 =0 represents male, x_i5=1 represents women；y_iRepresent the disease state of gathered person, when gathered person's P P >=50% Time, y_i=1, otherwise y_i=0.

S202: setup control parameter, defines fitness function, specifically includes step:

S2020: setup control parameter, including setting population size N, maximum iteration time K_max, wherein N, K_maxIt is big In the positive integer of zero；

S2021: variable declarations, including: current iteration number of times k, current optimal value P of i-th particle_besti, the currently overall situation Optimal value G_best, inertia weight ω, Studying factors c1 and c2, random number ξ, η, wherein k is the positive integer more than zero；

S2022: particle encodes, and affects factor beta to each₀、β₁、β₂、β₃、β₄、β₅Encode, the coding bag of i-th particle Include position encoded and velocity encoded cine, position encoded for β_i=(β_i0, β_i1, β_i2, β_i3, β_i4, β_i5), velocity encoded cine is v_i=(v_i0, v_i1, v_i2, v_i3, v_i4, v_i5)；

S2023: definition fitness function f (β), wherein β=(β₀, β₁, β₂, β₃, β₄, β₅) it is particle position.

After carrying out particle coding, calculate inertia weight ω, Studying factors c1 and c2, and random number ξ, η, then define Fitness function f (β).

S203: initialize, specifically includes: make k=0, it is thus achieved that the initial position β of i-th particle_i ⁽⁰⁾=(β_i0 ⁽⁰⁾, β_i1 ⁽⁰⁾, β_i2 ⁽⁰⁾, β_i3 ⁽⁰⁾, β_i4 ⁽⁰⁾, β_i5 ⁽⁰⁾), initial velocity v_i=(v_i0 ⁽⁰⁾, v_i1 ⁽⁰⁾, v_i2 ⁽⁰⁾, v_i3 ⁽⁰⁾, v_i4 ⁽⁰⁾, v_i5 ⁽⁰⁾), wherein β_ij ⁽⁰⁾、v_ij ⁽⁰⁾It is the random number of [-1000,1000], j=1,2,3,4,5.

S204: run iteration, for all i=1,2 ..., N, update speed and the position of i-th particle, for kth time Iteration, if f is (β_i ^(k))>P_besti, then P is made_besti=f (β_i ^(k)), if max{P_best1, P_best2..., P_bestN}>G_best, then make G_best=max{P_best1, P_best2..., P_bestN}；

S205: work as k=K_maxTime, export G_bestAnd the particle position β=(β of correspondence₀, β₁, β₂, β₃, β₄, β₅)。

When iterations reaches maximum iteration time, the G that output obtains_bestAnd the particle position β=(β of correspondence₀, β₁, β₂, β₃, β₄, β₅), then solve and obtain forecast model respectively affects coefficient.

S3: the health data of personnel to be measured is substituted in described forecast model, calculate and obtain predicting the outcome of personnel to be measured.

Solving based on the healthy data gathered to obtain forecast model affects factor beta₀、β₁、β₂、β₃、β₄、β₅ After, will affect coefficient substitute in forecast model, then the expression formula of this forecast model determines, when being predicted, by personnel to be measured Health data, including heart rate, body temperature, contraction pressure, diastolic pressure, sex etc., information data substitutes in forecast model expression formula, then Acquisition personnel to be measured can be calculated and suffer from grippal predicting the outcome.

Therefore, influenza Forecasting Methodology described in the present embodiment, with the health data of human body, including heart rate, body temperature, receipts Contractive pressure, diastolic pressure and sex set up influenza forecast model, then gather the data that some is individual, based on gather Sample data uses chaos APSO algorithm to solve to obtain and respectively affect coefficient in forecast model, obtains forecast model Expression formula, and then by this forecast model, personnel to be measured are suffered from influenza and make a prediction.The present embodiment influenza is pre- Survey method, suffers from influenza according to human heart rate, body temperature, contraction pressure, diastolic pressure and sex etc. to human body and makes prediction, obtain P, enables people to popularity flu and makes prevention in time.

It is further preferred that in the present embodiment use chaos APSO algorithm solving model affect coefficient Time, in above-mentioned steps S204, when a particle search a to locally optimal solution, all particles are by this optimal solution Attract, the vicinity of easy rapid aggregation to this locally optimal solution, thus earliness occurs, then appearance will be difficult to more preferably Fitness value, it is difficult to search global optimum.In consideration of it, in the present embodiment method for solving, described step S204 also includes:

Iteration secondary for kth, calculating population's fitness:

${\mathrm{\&sigma;}}^2=\frac{1}{N}\underset{i=1}{\overset{N}{\&Sigma;}}{\left(\frac{f\left({{\mathrm{\&beta;}}_i}^{\left(k\right)}\right)-\frac{1}{N}\underset{i=1}{\overset{N}{\&Sigma;}}f\left({{\mathrm{\&beta;}}_i}^{\left(k\right)}\right)}{f}\right)}^2,$

f(β_i ^(k)) representing the i-th particle fitness when iterations k, f is fitness evaluation value, is specifically described as:

$f=max\{1,max\{|f\left({{\mathrm{\&beta;}}_i}^{\left(k\right)}\right)-\frac{1}{N}\underset{i=1}{\overset{N}{\&Sigma;}}f\left({{\mathrm{\&beta;}}_i}^{\left(k\right)}\right)|\left\}\right\};$

Judge whether σ²＜ σ² _minIf otherwise running+1 iteration of kth, σ² _minFor default minimum population fitness, for just Number.

Work as σ²＜ σ² _min, it is determined that population enters Premature Convergence state；Work as σ²≥σ² _min, show that population does not enters into early Ripe convergence state, now can carry out k+1 iteration.Established standards judges whether population enters Premature Convergence state, it is simple to search Rope is to globally optimal solution, to improve the accuracy of solving result.

When judging that population enters Premature Convergence state, in order to strengthen the search performance of globally optimal solution, through less Iterations finds optimal solution, can choose the higher excellent particle of adaptability and carry out small chaotic disturbance behaviour in running iteration Make, and the particle that fitness is the highest is carried out random disturbance, to increase the ability of particle search globally optimal solution.

Specifically, when judging σ in above-mentioned steps²＜ σ² _min, then judge that population enters Premature Convergence state, to adaptation Spending the highest particle and carry out random disturbance, concrete grammar comprises the following steps:

One original chaotic vector of S300: stochastic generation, is described as z₀=(z₀₀,z₀₁,z₀₂,z₀₃,z₀₄,z₀₅), wherein z₀₀、 z₀₁、z₀₂、z₀₃、z₀₄、z₀₅Span be [0,1]；

S301: chaos iteration generates Q chaos vector, and the l vector description is z_l=(z₁₀,z_l1,z_l2,z_l3,z_l4, z_l5), l=1,2, ..., Q, 0 < Q < N, Q is positive integer；

S302: produce Q particle, the l particle is described as β_l=(β_l0,β_l1,β_l2,β_l3,β_l4,β_l5)；

S303: the particle that fitness is the highest is carried out random disturbance, is described as:

${z_l}^{\&prime;}=|\frac{e^{\frac{l}{Q}}-1}{e-1}-\frac{1}{l}{|}^{\frac{3}{2}}{z}^{*}+(1-|\frac{e^{\frac{l}{Q}}-1}{e-1}-\frac{1}{l}{|}^{\frac{3}{2}})z_l;$

Wherein, z_l' for applying the chaos vector that (β 0, β 1, β 2, β 3, β 4, β 5) is corresponding after random disturbance, z* be optimum Value β *=(β₀*, β₁*, β₂*, β₃*, β₄*, β₅*) the corresponding vector that [0,1] is formed afterwards, z it are mapped to_lFor the chaos after iteration l time Vector, l is chaos iteration number of times, and Q is maximum chaos iteration number of times, and z* is specifically described as:

$z^{*}=\frac{{\mathrm{\&beta;}}^{*}-{\mathrm{\&beta;}}_{min}}{{\mathrm{\&beta;}}_{\max }-{\mathrm{\&beta;}}_{min}};$

β_maxFor the maximum of particle coding, value is 1000 here；β_minFor the minima of particle coding, value here For-1000.

Next iteration is carried out after the particle that fitness is the highest is carried out random disturbance.

Therefore, in the present embodiment method for solving, judge whether population enters Premature Convergence state by established standards, Judge, when population enters Premature Convergence state, to choose the higher excellent particle of adaptability and carry out small chaotic disturbance operation, and The particle that fitness is the highest is carried out random disturbance, and to increase the ability of particle search globally optimal solution, make acquisition solves knot Fruit is more accurate.

Accordingly, the embodiment of the present invention also provides for a kind of influenza prediction means, refer to Fig. 3, and device includes:

Information acquisition module 400, including:

Heart rate sensor 401, for gathering the heart rate information of personnel to be measured；

Pressure transducer 402, for gathering contraction pressure and the diastolic pressure of personnel to be measured；

Body temperature trans 403, for gathering the body temperature of personnel to be measured；

User operation module 410, for receiving the gender information of the personnel to be measured of user's input；

Data analysis module 420, for according to the health data of the personnel to be measured gathered, health data include heart rate, Body temperature, contraction pressure, diastolic pressure and sex, pass through set up influenza forecast model calculating acquisition personnel to be measured and suffer from popular The probability that sexuality emits, described influenza forecast model describes with equation below:

$P=\frac{e^{{\mathrm{\&beta;}}_0+{\mathrm{\&beta;}}_1{x}_1+{\mathrm{\&beta;}}_2{x}_2+{\mathrm{\&beta;}}_3{x}_3+{\mathrm{\&beta;}}_4{x}_4+{\mathrm{\&beta;}}_5{x}_5}}{1+e^{{\mathrm{\&beta;}}_0+{\mathrm{\&beta;}}_1{x}_1+{\mathrm{\&beta;}}_2{x}_2+{\mathrm{\&beta;}}_3{x}_3+{\mathrm{\&beta;}}_4{x}_4+{\mathrm{\&beta;}}_5{x}_5}},$

Wherein, P is for suffering from grippal probability, β₀For constant, β₁For Heart rate influences coefficient, β₂For temperature influence coefficient, β₃Coefficient, β is affected for shrinking pressure₄Coefficient, β is affected for diastolic pressure₅For Effect of gender coefficient, variable x₁For heart rate value, variable x₂For Body temperature value, variable x₃For shrinking pressure value, variable x₄For diastolic blood pressure values, variable x₅For sex.

The embodiment of the present invention provide influenza prediction means, including information acquisition module, user operation module and Data analysis module, wherein information acquisition module includes heart rate sensor, pressure transducer and body temperature trans, is grasped by user Make module and can input the information data of personnel to be measured.The present embodiment influenza prediction means, builds with the health data of human body Vertical influenza forecast model, the health data being based on includes heart rate, body temperature, contraction pressure, diastolic pressure and sex, is treating When survey personnel detect, obtain the heart rate of personnel to be measured by information acquisition module, shrink the information such as pressure, diastolic pressure, body temperature, and defeated Entering sex, data analysis module provides personnel to be measured based on the forecast model calculating set up and suffers from grippal probability.

Therefore, influenza prediction means of the present invention, according to human heart rate, body temperature, contraction pressure, diastolic pressure and sex etc. Human body is suffered from influenza make prediction, obtain P, enable people to popularity flu and make prevention in time.

In the present embodiment, described user operation module is additionally operable to output and display predicts the outcome.

Described user operation module is additionally operable to receive and record the personal information of the personnel to be measured of user's input, including surname Name, sex, age and detection record.When user needs the detection signal inquiring about related personnel, user operation mould can be passed through Block is searched accordingly.

Above a kind of influenza Forecasting Methodology provided by the present invention and device are described in detail.Herein Applying specific case to be set forth principle and the embodiment of the present invention, the explanation of above example is only intended to help Understand method and the core concept thereof of the present invention.It should be pointed out that, for those skilled in the art, do not taking off On the premise of the principle of the invention, it is also possible to the present invention is carried out some improvement and modification, these improve and modification also falls into this In invention scope of the claims.

Claims (8)

1. an influenza Forecasting Methodology, it is characterised in that including:

Set up influenza forecast model with the health data of human body, described health data include heart rate, body temperature, contraction pressure, Diastolic pressure and sex, described influenza forecast model uses equation below to describe:

$P=\frac{e^{{\mathrm{\&beta;}}_0+{\mathrm{\&beta;}}_1{x}_1+{\mathrm{\&beta;}}_2{x}_2+{\mathrm{\&beta;}}_3{x}_3+{\mathrm{\&beta;}}_4{x}_4+{\mathrm{\&beta;}}_5{x}_5}}{1+e^{{\mathrm{\&beta;}}_0+{\mathrm{\&beta;}}_1{x}_1+{\mathrm{\&beta;}}_2{x}_2+{\mathrm{\&beta;}}_3{x}_3+{\mathrm{\&beta;}}_4{x}_4+{\mathrm{\&beta;}}_5{x}_5}};$

Wherein, P is for suffering from grippal probability, β₀For constant, β₁For Heart rate influences coefficient, β₂For temperature influence coefficient, β₃For Shrink pressure and affect coefficient, β₄Coefficient, β is affected for diastolic pressure₅For Effect of gender coefficient, variable x₁For heart rate value, variable x₂For body temperature Value, variable x₃For shrinking pressure value, variable x₄For diastolic blood pressure values, variable x₅For sex；

Random acquisition m group somatic data, described data are with (x₁, x₂, x₃, x₄, x₅, y) describe, solve based on described m group data and obtain Described forecast model respectively affects coefficient, wherein y represents the disease state of gathered person；

The health data of personnel to be measured is substituted in described forecast model, calculate and obtain predicting the outcome of personnel to be measured.

2. the method for claim 1, it is characterised in that use chaos APSO algorithm to solve acquisition described pre- Surveying respectively affects coefficient in model.

3. the method for claim 1, it is characterised in that described random acquisition m group somatic data, described data are with (x₁, x₂, x₃, x₄, x₅, y) describe, based on described m group data solve acquisition described forecast model in respectively affect coefficient, including:

S201: gather m group somatic data, with (x_i1, x_i2, x_i3, x_i4, x_i5, y_i) describe, i=1,2 ..., m；

S202: setup control parameter, including setting population size N, maximum iteration time K_max, wherein N, K_maxIt is more than zero Positive integer；

Variable declarations, including: current iteration number of times k, current optimal value P of i-th particle_besti, current global optimum G_best, Inertia weight ω, Studying factors c1 and c2, random number ξ, η, wherein k is the positive integer more than zero；

Particle encodes, and affects factor beta to each₀、β₁、β₂、β₃、β₄、β₅Encoding, the coding of i-th particle includes position encoded And velocity encoded cine, position encoded for β_i=(β_i0, β_i1, β_i2, β_i3, β_i4, β_i5), velocity encoded cine is v_i=(v_i0, v_i1, v_i2, v_i3, v_i4, v_i5)；

Definition fitness function f (β), wherein β=(β₀, β₁, β₂, β₃, β₄, β₅) it is particle position；

S203: initialize, specifically includes: make k=0, it is thus achieved that the initial position β of i-th particle_i ⁽⁰⁾=(β_i0 ⁽⁰⁾, β_i1 ⁽⁰⁾, β_i2 ⁽⁰⁾, β_i3 ⁽⁰⁾, β_i4 ⁽⁰⁾, β_i5 ⁽⁰⁾), initial velocity v_i=(v_i0 ⁽⁰⁾, v_i1 ⁽⁰⁾, v_i2 ⁽⁰⁾, v_i3 ⁽⁰⁾, v_i4 ⁽⁰⁾, v_i5 ⁽⁰⁾), wherein β_ij ⁽⁰⁾、 v_ij ⁽⁰⁾It is the random number of [-1000,1000], j=1,2,3,4,5；

S204: run iteration, for all i=1,2 ..., N, update speed and the position of i-th particle, for kth time repeatedly In generation, if f is (β_i ^(k))>P_besti, then P is made_besti=f (β_i ^(k)), if max{P_best1, P_best2..., P_bestN}>G_best, then make G_best=max{P_best1, P_best2..., P_bestN}；

S205: work as k=K_maxTime, export G_bestAnd the particle position β=β of correspondence₀, β₁, β₂, β₃, β₄, β₅)。

4. method as claimed in claim 3, it is characterised in that described step S204 also includes: for kth time iteration, calculate Population's fitness:

${\mathrm{\&sigma;}}^2=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{\left(\frac{f\left({{\mathrm{\&beta;}}_i}^{\left(k\right)}\right)-\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}f\left({{\mathrm{\&beta;}}_i}^{\left(k\right)}\right)}{f}\right)}^2,$

f(β_i ^(k)) representing the i-th particle fitness when iterations k, f is fitness evaluation value, is specifically described as:

$f=max\{1,max\{|f\left({{\mathrm{\&beta;}}_i}^{\left(k\right)}\right)-\frac{1}{N}\underset{i=1}{\overset{N}{\&Sigma;}}f\left({{\mathrm{\&beta;}}_i}^{\left(k\right)}\right)|\left\}\right\};$

Judge whether σ²＜ σ² _minIf otherwise running+1 iteration of kth, σ² _minFor default minimum population fitness, for positive number.

5. method as claimed in claim 4, it is characterised in that if σ²＜ σ² _min, then judge that population enters Premature Convergence shape State, carries out random disturbance to the particle that fitness is the highest, specifically includes:

One original chaotic vector of S300: stochastic generation, is described as z₀=(z₀₀,z₀₁,z₀₂,z₀₃,z₀₄,z₀₅), wherein z₀₀、z₀₁、 z₀₂、z₀₃、z₀₄、z₀₅Span be [0,1]；

S301: chaos iteration generates Q chaos vector, and the l vector description is z_l=(z₁₀,z_l1,z_l2,z_l3,z_l4,z_l5), l= 1,2, ..., Q, 0 < Q < N, Q is positive integer；

S302: produce Q particle, the l particle is described as β_l=(β_l0,β_l1,β_l2,β_l3,β_l4,β_l5)；

S303: the particle that fitness is the highest is carried out random disturbance, is described as:

${z_l}^{\&prime;}=|\frac{e^{\frac{l}{Q}}-1}{e-1}-\frac{1}{l}{|}^{\frac{3}{2}}{z}^{*}+(1-|\frac{e^{\frac{l}{Q}}-1}{e-1}-\frac{1}{l}{|}^{\frac{3}{2}})z_l;$

Wherein, z_l' for applying the chaos vector that (β 0, β 1, β 2, β 3, β 4, β 5) is corresponding after random disturbance, z* be optimal value β *= (β₀*, β₁*, β₂*, β₃*, β₄*, β₅*) the corresponding vector that [0,1] is formed afterwards, z it are mapped to_lFor the chaos vector after iteration l time, l For chaos iteration number of times, Q is maximum chaos iteration number of times, and z* is specifically described as:

$z^{*}=\frac{{\mathrm{\&beta;}}^{*}-{\mathrm{\&beta;}}_{min}}{{\mathrm{\&beta;}}_{\max }-{\mathrm{\&beta;}}_{min}};$

β_maxFor the maximum of particle coding, β_minMinima for particle coding.

6. an influenza prediction means, it is characterised in that including:

Information acquisition module, including:

Heart rate sensor, for gathering the heart rate information of personnel to be measured；

Pressure transducer, for gathering contraction pressure and the diastolic pressure of personnel to be measured；

Body temperature trans, for gathering the body temperature of personnel to be measured；

User operation module, for receiving the gender information of the personnel to be measured of user's input；

Data analysis module, for the health data according to the personnel to be measured gathered, health data includes heart rate, body temperature, receipts Contractive pressure, diastolic pressure and sex, pass through set up influenza forecast model calculating acquisition personnel to be measured and suffer from influenza Probability, described influenza forecast model with equation below describe:

$P=\frac{e^{{\mathrm{\&beta;}}_0+{\mathrm{\&beta;}}_1{x}_1+{\mathrm{\&beta;}}_2{x}_2+{\mathrm{\&beta;}}_3{x}_3+{\mathrm{\&beta;}}_4{x}_4+{\mathrm{\&beta;}}_5{x}_5}}{1+e^{{\mathrm{\&beta;}}_0+{\mathrm{\&beta;}}_1{x}_1+{\mathrm{\&beta;}}_2{x}_2+{\mathrm{\&beta;}}_3{x}_3+{\mathrm{\&beta;}}_4{x}_4+{\mathrm{\&beta;}}_5{x}_5}},$

Wherein, P is for suffering from grippal probability, β₀For constant, β₁For Heart rate influences coefficient, β₂For temperature influence coefficient, β₃For Shrink pressure and affect coefficient, β₄Coefficient, β is affected for diastolic pressure₅For Effect of gender coefficient, variable x₁For heart rate value, variable x₂For body temperature Value, variable x₃For shrinking pressure value, variable x₄For diastolic blood pressure values, variable x₅For sex.

7. device as claimed in claim 6, it is characterised in that described user operation module is additionally operable to output and display prediction knot Really.

8. device as claimed in claim 6, it is characterised in that described user operation module is additionally operable to reception and record user is defeated The personal information of the personnel to be measured entered, including name, sex, age and detection record.