CN102894982B - Non-invasive detecting method for blood viscosity based on pulse wave - Google Patents

Abstract

The invention discloses a non-invasive detecting method for blood viscosity based on pulse wave, and belongs to the fields of biomechanics and hemorheology. The method is characterized by comprising the following steps of: detecting the pressure pulse wave of a radial artery by using a pressure sensor to realize non-invasive detection of the blood viscosity, establishing a relation among the pulse wave of the radial artery, pressure gradient and blood flow, and solving an ordinary differential equation to obtain a numerical solution of the blood flow at different moments in a cardiac cycle; and establishing a relation between the blood flow and the blood viscosity, and thus obtaining the blood viscosity. Meanwhile, by the method, personalized calculation can be realized, and the pain of patients can be furthest relieved; and the detecting method is simple and convenient, is beneficial to promoting universal and popular development of blood viscosity detection, and finally realizes early prediction and clinical diagnosis functions of certain hemodynamic diseases.

Description

Blood viscosity noinvasive detection method based on pulse wave

Technical field

The invention belongs to biomechanics technology, relate to computer program, circuit design etc., the noinvasive that can be used for blood viscosity detects.

Background technology

When blood flows in blood vessel, the internal friction of its inner each intermolecular generation is called blood viscosity.Blood viscosity is one of important indicator of reflection fluid dynamics character and coagulating property.Measure blood viscosity and have extremely important physiology and pathology sense to studying formation, development and the prevention of some cardiovascular and cerebrovascular disease.Current clinical general employing Testing in vitro mode, utilizes respectively the viscosity of the device measuring Newton type fluids such as capillary viscosimeter, rotating cylinder viscometer and non-Newtonian fluid.These class methods have its limitation: because the improper meeting of blood sampling causes blood viscosity to measure obvious error; The dilution that interpolation anticoagulant can cause blood constituent is with change and because measuring method is different, and the flow regime of blood is different, the error of the principle causing etc.Therefore,, for the deficiency existing in traditional measurement method, this patent has proposed the blood viscosity noinvasive detection method based on the non-linear pulse wave propagate theory of radial artery.

The method is set up disparate modules, and the multi-parts such as the information exchange excess pressure sensor that progressively can detect, A/D converter, memorizer, CPU are changed and processed, and the noinvasive of finally realizing viscosity detects.In module process of establishing, first the equation of motion based on non-linear pulse wave theory and fluid momentum conservation, set up the barometric gradient relational model of pulse wave velocity and blood pressure waveform figure, barometric gradient and blood flow relational model, blood flow and blood viscosity relational model.Then, set up ordinary differential equation long-pending about blood flow, vessel cross-sections, barometric gradient, by Euler's method solving equation, obtain in a cardiac cycle not blood flow value in the same time, then realize the calculating of blood viscosity.Finally by error analysis, judge whether result of calculation reaches convergence precision.If convergence, Output rusults; Otherwise will carry out iterative computation, and repeat blood flow and blood viscosity and calculate process.In addition, this method has also been set up the relation of the personalizing parameters such as human vas radius and individual height, age, shoulder breadth, be intended to set up the relation of model physical constants and measurable parameter, by measurement or result of calculation, replace physical constants, thereby realize the personalization detection of this method.

Current noinvasive detection method, is mainly to utilize pressure transducer to detect radial artery pulse wave, estimates one and detects the good waveform of effect and choose the detection that this waveform is correlated with.Etection theory is mainly calculating blood pressure waveform temporal evolution parameter K value, based on cardiovascular model theory, carries out correlation computations derivation, finally obtains viscosity number.Its maximum defect is, does not obtain in a cardiac cycle blood pressure pulsatile change in time.Different blood pressure waveforms and numerical value will represent the physiological and pathological state of different human body, and the method has only been considered the difference of blood pressure waveform, does not consider the different situations of blood pressure values.This is also an emphasis of the present invention.

Summary of the invention

The object of the invention is, avoids the existing wound detection means that has, and the noinvasive of realizing blood viscosity detects.The blood viscosity detection method that the present invention proposes and the fundamental difference of other noinvasive detection methods are, theoretical based on pulse wave propagate, by a plurality of pressure pulse waveforms, obtain an average pressure oscillogram, set up the barometric gradient of this mean blood pressure oscillogram and carry out hemodynamic calculating and analysis.

The invention is characterized in, in computer, realize according to the following steps successively:

Step (1), computer initialization:

Input: the radial artery pressure pulse wave detecting with pressure transducer or cuff sphygmomanometer and measured's pulse pressure value;

Measured's personal information, at least comprises sex, age, height, body weight and shoulder breadth;

Step (2), after computer is demarcated the radial artery pressure pulse wave of input, obtains the average pulse pressure P of measured in a cardiac cycle _m;

Step (3), computer is calculated as follows the spread speed c (t) of radial artery pressure pulse wave:

$c\left(t\right)=\sqrt{\frac{P\left(t\right)(1+b*\ln \frac{P\left(t\right)}{\mathrm{Pm}})}{\mathrm{\ρb}}},$ Wherein:

P (t), is pulse pressure value corresponding to each sampled point in a cardiac cycle,

ρ, is density of blood, ρ=1.05*10 ^-3kg/cm ³,

B, is pulse waveform parameter, is calculated as follows:

$b=\frac{a{\left(\frac{0.81}{k+0.48}\right)}^2-1}{\mathrm{\π}},$

$k=\frac{P_m-P_d}{P_s-P_d},$ Wherein:

K, is shape factor,

P _d, be diastolic pressure, P _sfor systolic pressure, be measured value,

A is the parameter relevant with velocity profile, a=0.57;

Step (4), computer is calculated as follows barometric gradient F (t):

$F\left(t\right)=\frac{1}{\mathrm{\ρc}\left(t\right)}\frac{\mathrm{dP}\left(t\right)}{\mathrm{dt}},$ Wherein:

be the derivative of pulse pressure value to the time in a cardiac cycle;

Step (5), separates following blood flow ordinary differential equation, obtains blood flow Q (t):

$\frac{\mathrm{dQ}\left(t\right)}{\mathrm{dt}}+\mathrm{\λ}\left(t\right)Q\left(t\right)+\mathrm{\ϵ}\left(t\right)Q^2\left(t\right)=A\left(t\right)F\left(t\right),$ Wherein:

$\mathrm{\λ}\left(t\right)=\frac{8\mathrm{\α\γ}}{R^2\left(t\right)}-\frac{\mathrm{\β}[{\mathrm{\β}}_1^2{\left({\mathrm{\α}}_2{\mathrm{\β}}_{2m}\right)}^4-1]}{P_m-[{\mathrm{\β}}_1^2{\left({\mathrm{\α}}_2{\mathrm{\β}}_{2m}\right)}^4-1](P\left(t\right)-P_m)}\frac{\mathrm{dP}\left(t\right)}{\mathrm{dt}},$

wherein:

λ (t), is the time dependent first power coefficient of blood flow,

for the semi-cone angle under naturalness in blood vessel,

β ₁, be the ratio of the length when body length and the blood vessel stress balance of blood vessel under normal physiological conditions, β ₁=0.57,

β _2m, vessel radius vessel radius ratio when not being out of shape during for blood vessel stress balance, β _2m=1.105,

α ₂, be human body correction factor under physiological condition,

γ, is blood motion viscosity, is the ratio of hemodynamics viscosity and density of blood, and when first calculating, dynamic viscosity is progressive viscosity, mu ₀, under physiological condition, desired viscosity when shear rate is enough large, μ ₀=3.6522081, be setting value,

α, β, β ₀, be human body correction factor under physiological condition, wherein:

α, the non-linear pulse wave propagate coefficient while relating to the interior blood flow temporal evolution of a cardiac cycle during for calculating λ (t),

β, β ₀non-linear pulse wave propagate coefficient while relating to the interior vessel radius temporal evolution of a cardiac cycle during for calculating ε (t),

α=0.57, β=1.173, β ₀=-1.5, β is relevant with vessel radius elastic dilatation, β ₀relevant with vessel radius elastic shrinkage, ε (t), is the time dependent secondary power coefficient of blood flow,

A (t), is the time dependent function of vessel radius in a cardiac cycle, describes the dynamic process of blood vessel elasticity shrinkage expansion in a cardiac cycle, with following formula, represents:

A (t)=π R ²(t), wherein:

R (t), reacts vessel radius over time,

$R\left(t\right)=R_m*\sqrt{1+b*\ln \frac{P\left(t\right)}{P_m}},$ Wherein:

R _m, be testee's physiological parameter, vessel radius value representation when not being out of shape,

$R_m=0.042+0.000625(1+0.36\sqrt{G})\sqrt{\mathrm{hl}},$ Wherein:

G, is testee's age,

H, is testee's height,

L is testee's shoulder breadth;

Step (6), is calculated as follows a mean blood flow Q in cardiac cycle _m, waveform parameter ξ, shear rate D (t) and flow factor S:

$Q_m=\frac{1}{T}{\∫}_0^{t}Q\left(t\right)\mathrm{dt}=\frac{1}{N}{\mathrm{\Σ}}_{n=1}^{n}Q\left(n\right),$ Wherein:

Q (t), the blood flow measuring during for moment t,

N is a sampling number in cardiac cycle T, is numerically equal to a sampling instant number in cardiac cycle T, t=1, and 2 ... t _n, be setting value,

$\mathrm{\ξ}=\frac{k-K_0}{K_0},$ Wherein:

K ₀, be the physiological standard value of shape factor K, K ₀=0.33,

$D\left(t\right)=\frac{4\mathrm{aQ}\left(t\right)}{\mathrm{\π}{R}^3\left(t\right)},$

$S=\frac{K_1-K_Q}{K_1},$ Wherein:

K ₁, be peak value blood flow Q _pwith mean blood flow Q _mthe normal physiologic values of ratio, K ₁=6.35,

K _q, be peak value blood flow Q _pwith mean blood flow Q _mthe value of calculation of ratio;

Step (7), is calculated as follows a blood viscosity μ in cardiac cycle:

μ=μ ₀(1+g ₁d ^-1(t)+g ₂d ^-2(t)) (1+g ₃ξ+g ₄ξ ²) (1+g ₅s+g ₆s ²), wherein:

G ₁, for to D ^-1(t) correction factor, g ₁=8.16,

G ₂, for to D ^-2(t) correction factor, g ₂=47.52,

G ₃, be the correction factor to ξ, g ₃=0.7506,

G ₄, for to ξ ²correction factor, g ₄=-0.03756,

G ₅, be the correction factor to S, g ₅=0.0089,

G ₆, for to S ²correction factor, g ₆=0.00184;

Step (8), the error of blood viscosity μ is judged:

Set: iterations is secondary at least,

Calculate the relative μ of μ ₀relative error:

If: make μ=μ ₀,

: return to step (1), gather the radial artery pressure pulse wave of next cardiac cycle, re-execute step (2)-step (7), recalculate μ, until till.

The present invention is based on the hardware devices such as pressure transducer, A/D converter and gather after pulse wave information, by computer, complete correlation computations, finally obtain blood viscosity result.Required hardware device is simple, easy to operate, easily detects, and completes this computer procedures required time lower than 2 minutes simultaneously, farthest reduces time cost.Acquired results meets within the scope of biological value 4-8, belongs to clinical acceptable value.In addition, same measured is carried out to clinical have wound detection and noinvasive simultaneously and detect, acquired results relative error, in 20%, belongs within the scope of acceptable error.Therefore, this invention has result of calculation more accurately, also has very strong Practical significance simultaneously.

Accompanying drawing explanation

Fig. 1 is schematic flow sheet of the present invention.

Fig. 2 is barometric gradient of the present invention and flow rate calculation subprogram schematic flow sheet.

The specific embodiment

First, utilize pressure transducer, cuff sphygmomanometer to detect radial artery pressure pulse wave and measured's pressure value, and import computer into and demarcate.Meanwhile, input testee personal information, completes computer initialization process.

Then, by barometric gradient and blood flow relational model, the numerical integration form of deriving complete flow equation is as follows:

$\frac{\mathrm{dQ}\left(t\right)}{\mathrm{dt}}+\mathrm{\λ}\left(t\right)Q\left(t\right)+\mathrm{\ϵ}\left(t\right)Q^2\left(t\right)=A\left(t\right)F\left(t\right),$

Wherein flow parameter λ (t) and ε (t) are calculated by following formula respectively:

$\mathrm{\λ}\left(t\right)=\frac{8\mathrm{\α\γ}}{R^2\left(t\right)}-\frac{\mathrm{\β}[{\mathrm{\β}}_1^2{\left({\mathrm{\α}}_2{\mathrm{\β}}_{2m}\right)}^4-1]}{P_m-[{\mathrm{\β}}_1^2{\left({\mathrm{\α}}_2{\mathrm{\β}}_{2m}\right)}^4-1](P\left(t\right)-P_m)}\frac{\mathrm{dP}\left(t\right)}{\mathrm{dt}},$

The 3rd step, calculates shape factor, shear rate, three parameters of the blood flow factor by improving one's methods, and calculation method of parameters is as follows:

$\mathrm{\ξ}=\frac{k-K_0}{K_0},$

$D\left(t\right)=\frac{4\mathrm{aQ}\left(t\right)}{\mathrm{\π}{R}^3\left(t\right)},$

$S=\frac{K_1-K_Q}{K_1},$

The 4th step, the equation of motion based on non-linear pulse wave theory and fluid momentum conservation, set up barometric gradient relational model, barometric gradient and the blood flow relational model of pulse wave velocity and blood pressure waveform figure, and finally obtain blood viscosity solving model, as follows:

μ=μ ₀(1+g ₁D ^-1(t)+g ₂D ^-2(t))(1+g ₃ξ+g ₄ξ ²)(1+g ₅S+g ₆S ²)

Finally, by solving after viscosity number in parameter substitution viscosity solving equation group, carry out error judgement.Herein, error is decided to be relative error, and at least twice iterative process, relative error <5%, stops calculating.If result in error allowed band, Output rusults; Otherwise, realize iterative process.Error calculation method is as follows:

With reference to schematic flow sheet, the concrete implementation step of the present invention comprises:

Step 1: computer initialization.Comprise input measured name, sex, age, height, body weight, shoulder breadth, systolic pressure, diastolic pressure.Gather, show, store pulse waveform, obtain average waveform and demarcate.Input initial viscosity value u ₀, calculate shape factor K value, viscosity coefficient ξ value, velocity of wave propagation c value.

Step 2: calculate pulse wave velocity.

Step 3: calculate the first moment barometric gradient value.

Step 4: calculate the first blood flow value storage constantly, then calculate barometric gradient and the blood flow value in whole cycle from the 2nd moment, comprise the following steps:

1) as 1≤t≤t _ntime, calculate calculating pressure gradient F (t), t=t ₀=0 constantly as timing starting point, does not measure;

2) calculate calculated flow rate parameter lambda (t), ε (t);

3) calculate each flow value Q (t) constantly, and storage;

4) t=t+1 cycle calculations;

Step 5: calculate average discharge Qm, peak flow Qp value, average shear rate Dm, peak value shear rate Dp, viscosity parameter S value and other related parameter values.

Step 6: calculate the blood viscosity value μ in a cardiac cycle.

Step 7: carry out error judgement.If result in error allowed band, direct Output rusults; Otherwise, return to step (1), gather the radial artery pressure pulse wave of next cardiac cycle, re-execute step (2)-step (7), recalculate μ, until till, wherein:

Claims (1)

1. the blood viscosity noinvasive detection method based on pulse wave, is characterized in that realizing according to the following steps successively in computer:

Step (1), computer initialization:

Input: the radial artery pressure pulse wave detecting with pressure transducer or cuff sphygmomanometer and measured's pulse pressure value;

Measured's personal information, at least comprises sex, age, height, body weight and shoulder breadth;

Step (2), after computer is demarcated the radial artery pressure pulse wave of input, obtains the average pulse pressure P of measured in a cardiac cycle _m;

Step (3), computer is calculated as follows the spread speed c (t) of radial artery pressure pulse wave:

wherein:

P (t), is pulse pressure value corresponding to each sampled point in a cardiac cycle,

ρ, is density of blood, ρ=1.05*10 ^-3kg/cm ³,

B, is pulse waveform parameter, is calculated as follows:

wherein:

K, is shape factor,

P _d, be diastolic pressure, P _sfor systolic pressure, be measured value,

A is the parameter relevant with velocity profile, a=0.57;

Step (4), computer is calculated as follows barometric gradient F (t):

wherein:

be the derivative of pulse pressure value to the time in a cardiac cycle;

Step (5), separates following blood flow ordinary differential equation, obtains blood flow Q (t):

wherein:

wherein:

λ (t), is the time dependent first power coefficient of blood flow,

for the semi-cone angle under naturalness in blood vessel,

β ₁, be the ratio of the length when body length and the blood vessel stress balance of blood vessel under normal physiological conditions, β ₁=0.57,

β _2m, vessel radius vessel radius ratio when not being out of shape during for blood vessel stress balance, β _2m=1.105,

α ₂, be human body correction factor under physiological condition,

γ, is blood motion viscosity, is the ratio of hemodynamics viscosity and density of blood, and when first calculating, dynamic viscosity is progressive viscosity, mu ₀, under physiological condition, desired viscosity when shear rate is enough large, μ ₀=3.6522081, be setting value,

α, β, β ₀, be human body correction factor under physiological condition, wherein:

α, the non-linear pulse wave propagate coefficient while relating to the interior blood flow temporal evolution of a cardiac cycle during for calculating λ (t),

β, β ₀non-linear pulse wave propagate coefficient while relating to the interior vessel radius temporal evolution of a cardiac cycle during for calculating ε (t), α=0.57, β=1.173, β ₀=-1.5, β is relevant with vessel radius elastic dilatation, β ₀relevant with vessel radius elastic shrinkage, ε (t), is the time dependent secondary power coefficient of blood flow,

A (t), is the time dependent function of vessel radius in a cardiac cycle, describes the dynamic process of blood vessel elasticity shrinkage expansion in a cardiac cycle, with following formula, represents:

A (t)=π R ²(t), wherein:

R (t), reacts vessel radius over time,

wherein:

R _m, be testee's physiological parameter, vessel radius value representation when not being out of shape,

wherein:

G, is testee's age,

H, is testee's height,

L is testee's shoulder breadth;

Step (6), is calculated as follows a mean blood flow Q in cardiac cycle _m, waveform parameter ξ, shear rate D (t) and flow factor S:

wherein:

Q (t), the blood flow measuring during for moment t,

N is a sampling number in cardiac cycle T, is numerically equal to a sampling instant number in cardiac cycle T, t=1, and 2 ... t _n, be setting value,

wherein:

K ₀, be the physiological standard value of shape factor K, K ₀=0.33,

wherein:

K ₁, be peak value blood flow Q _pwith mean blood flow Q _mthe normal physiologic values of ratio, K ₁=6.35,

K _q, be peak value blood flow Q _pwith mean blood flow Q _mthe value of calculation of ratio;

Step (7), is calculated as follows a blood viscosity μ in cardiac cycle:

μ=μ ₀(1+g ₁d ^-1(t)+g ₂d ^-2(t)) (1+g ₃ξ+g ₄ξ ²) (1+g ₅s+g ₆s ²), wherein:

G ₁, for to D ^-1(t) correction factor, g ₁=8.16,

G ₂, for to D ^-2(t) correction factor, g ₂=47.52,

G ₃, be the correction factor to ξ, g ₃=0.7506,

G ₄, for to ξ ²correction factor, g ₄=-0.03756,

G ₅, be the correction factor to S, g ₅=0.0089,

G ₆, for to S ²correction factor, g ₆=0.00184;

Step (8), the error of blood viscosity μ is judged:

Set: iterations is secondary at least,

Calculate the relative μ of μ ₀relative error:

If: make μ=μ ₀,

: return to step (1), gather the radial artery pressure pulse wave of next cardiac cycle, re-execute step (2)-step (7), recalculate μ, until till.