CN106510669A - Wrist-belt-free blood pressure measuring system - Google Patents

Abstract

The invention provides a cuffless blood pressure measurement system. The cuffless blood pressure measurement system includes: a fitting unit, which is used to perform function fitting on the reference blood pressure data and pulse wave transit time; an online calculation unit for the quality of physiological signal acquisition, which is used to online evaluate the physiological signal waveform in daily work and living conditions The wave form quality that obtains below obtains wave form quality factor; Pulse wave transit time online calculating unit, is used for calculating pulse wave transit time from physiological signal waveform, and described pulse wave transit time is a sequence; Kalman filtering unit, is used for The blood pressure or the pulse wave transit time is subjected to Kalman filtering; and the blood pressure calculation unit is used to calculate the blood pressure according to the pulse wave transit time and fitting parameters, and the blood pressure is a sequence. The invention can reduce the influence of noise on the accuracy of blood pressure monitoring through the physiological signal waveform, and improve the measurement accuracy of non-cuff, dynamic and continuous blood pressure.

Description

Cuffless Blood Pressure Measurement System

technical field

The invention relates to the field of blood pressure monitoring sensor measurement, in particular to a cuffless blood pressure measurement system.

Background technique

With the rapid development of social economy, people pay more and more attention to health status; especially with the growth of aging population and the increase of chronic disease population, the number of hypertensive patients is increasing, the existing arm or wrist ambulatory blood pressure monitoring equipment Due to the need to inflate and deflate, it has a great impact on the monitored person (especially not suitable for dynamic continuous blood pressure monitoring during sleep), cannot realize beat-by-beat and continuous monitoring of arterial blood pressure, and has shortcomings such as high power consumption.

Indirect beat-to-beat blood pressure based on pulse wave transit time is a commonly used continuous and ambulatory blood pressure monitoring method, but because the pulse wave transit time is calculated based on physiological signals such as ECG, blood oxygen, impedance and body surface pressure pulse waves , since work during the day, sports in life, and turning over at night will affect the acquisition quality of physiological signal waveforms such as ECG, blood oxygen, impedance, and pressure pulse waves, which in turn will affect the measurement results of pulse wave transit time, and ultimately affect Measurement accuracy of cuffless, continuous, ambulatory blood pressure.

Contents of the invention

(1) Technical problems to be solved

The object of the present invention is to provide a cuffless, Continuous, dynamic, beat-by-beat blood pressure measurement system.

(2) Technical solution

The present invention provides a cuffless blood pressure measurement system, which includes: a fitting unit for performing function fitting on reference blood pressure data and pulse wave transit time; an online calculation unit for physiological signal acquisition quality online for evaluating physiological The waveform quality of the signal waveform obtained under the daily work and living conditions is obtained to obtain the waveform quality factor; the pulse wave transit time online calculation unit is used to calculate the pulse wave transit time from the physiological signal waveform, and the pulse wave transit time is a sequence; A Kalman filter unit is used to perform Kalman filter on blood pressure or pulse wave transit time; and a blood pressure calculation unit is used to calculate blood pressure according to pulse wave transit time and fitting parameters, the blood pressure being a sequence.

(3) Beneficial effects

It can be seen from the above technical solutions that the method and system for improving the accuracy of cuffless blood pressure measurement of the present invention have at least one of the following beneficial effects:

(1) Online evaluation of the influence of body movement and other factors on the quality of physiological signal waveform acquisition, and real-time grasp of the strength of noise during the measurement process;

(2) When the quality of the physiological signal waveform is poor, the Kalman filter method is used to estimate the blood pressure value at the current point by using the blood pressure value calculated from the pulse wave in the past, thereby reducing the noise and affecting the blood pressure monitoring accuracy through the physiological signal waveform The influence of the blood pressure can improve the measurement accuracy of cuffless, dynamic and continuous blood pressure.

Description of drawings

Fig. 1 is the component unit of cuffless blood pressure measurement system of the present invention;

Fig. 2 is a flow chart of the cuffless blood pressure measurement system of the present invention;

Fig. 3 is another flowchart of the cuffless blood pressure measurement system of the present invention;

Fig. 4 is a schematic diagram of pulse wave transit time calculation in the present invention;

Figure 5 is a comparison of the error distribution between the new method and the traditional method for cuffless continuous systolic and diastolic blood pressure measurement.

detailed description

In order to make the technical solutions and advantages of the present invention clearer and easier to understand, the present invention will be further described in detail below in combination with specific implementation examples and with reference to the accompanying drawings.

Please refer to Fig. 1, an embodiment of the present invention provides a cuffless, continuous, dynamic blood pressure measurement system, including a fitting unit 1, a physiological signal acquisition quality online calculation unit 2, a pulse wave transit time online calculation unit 3, a Kalman Filter unit 4 , blood pressure calculation unit 5 . Fitting unit 1 is used for function fitting of reference blood pressure data and pulse wave transit time; physiological signal acquisition quality online calculation unit 2 is used for online evaluation of physiological signal waveforms such as ECG, blood oxygen, impedance and body surface pressure pulse wave. The waveform quality obtained under daily work and living conditions is evaluated online due to the influence of factors such as body movement on the quality of physiological signal waveform acquisition, and the waveform quality factor is obtained; the pulse wave transmission time online calculation unit 3 is used to calculate the pulse wave from the physiological signal waveform Transmission time, the pulse wave transmission time is a sequence; the Kalman filter unit 4 is used to perform Kalman filtering on the blood pressure sequence or the pulse wave transmission time series; the blood pressure calculation unit 5 is used to calculate according to the pulse wave transmission time and fitting parameters Blood pressure, which is a sequence.

Specifically, the fitting unit includes:

The first fitting subunit is used to measure the physiological signal wave and the reference blood pressure value N times, and calculate the corresponding pulse wave transmission time based on the physiological signal wave obtained by each measurement;

The second fitting subunit is used to fit the pulse wave transit time and the corresponding reference blood pressure value to obtain the functional relationship between the two.

In the Kalman filtering unit, Kalman filtering is performed on the blood pressure sequence based on the waveform quality factor, and in a working process:

When the quality factor of the physiological signal waveform is higher than the preset first threshold, it is divided into the following two situations:

When the waveform quality factor is higher than the preset second threshold, increase the weight of the current blood pressure value, estimate the blood pressure value, and obtain the filtered blood pressure value;

When the waveform quality factor is lower than the preset second threshold, the weight of the previous blood pressure value is increased, and the blood pressure value is estimated to obtain the filtered blood pressure value;

When the quality factor of the physiological signal waveform is lower than the first threshold, the update of the Kalman filtering process is not performed, and the previous blood pressure value is used as the current blood pressure value;

The Kalman filter unit works continuously to obtain continuous high-precision blood pressure sequences.

Or, in the Kalman filtering unit, Kalman filtering is performed on the pulse wave transmission time series based on the waveform quality factor, in a working process,

When the quality factor of the physiological signal waveform is higher than the preset first threshold, it is divided into the following two situations:

When the waveform quality factor is higher than the preset second threshold, increase the weight of the current pulse wave transmission time, estimate the pulse wave transmission time, and obtain the pulse wave transmission time after filtering;

When the waveform quality factor is lower than the preset second threshold, increase the weight of the previous pulse wave transmission time, estimate the pulse wave transmission time, and obtain the pulse wave transmission time after filtering;

When the physiological signal waveform quality factor is lower than the first threshold, the Kalman filtering process is not updated, and the previous pulse wave transmission time is used as the current pulse wave transmission time;

The Kalman filtering unit works continuously to obtain a continuous high-precision pulse wave transmission time series.

Please refer to Figure 2, in this embodiment, the specific working process of the cuffless blood pressure measurement system is:

During calibration:

First, the user performs N measurements, and obtains physiological signal waves and reference blood pressure in each measurement;

Secondly, for each measured physiological signal wave, the pulse wave transit time online calculation unit 3 calculates its corresponding pulse wave transit time;

Again, the pulse wave transit time and the corresponding reference blood pressure are fitted by the fitting unit 1 to obtain the functional relationship between the two;

During the measurement:

Firstly, the physiological signal is measured by the physiological signal acquisition quality online calculation unit 2 to obtain a physiological signal wave, and the waveform quality factor of the physiological signal wave is calculated;

Secondly, for the physiological signal wave, the pulse wave transit time online calculation unit calculates its corresponding pulse wave transit time, and continuously measures to obtain the corresponding pulse wave transit time sequence;

Again, the blood pressure calculation unit uses the functional relationship to calculate the actual blood pressure sequence corresponding to the pulse wave transmission time series;

Finally, the Kalman filtering unit 4 performs Kalman filtering on the actual blood pressure sequence based on the waveform quality factor. When the waveform quality factor of the physiological signal is higher than the preset threshold, the higher the waveform quality factor, the more dependent the blood pressure estimate is. For the current blood pressure value, the lower the waveform quality factor, the more the estimated value of blood pressure depends on the previous blood pressure value. When the waveform quality factor of the physiological signal is lower than the threshold, the blood pressure value is considered too unreliable, and Kalman filtering is not performed at this time The process is updated to obtain a continuous high-precision blood pressure sequence.

Wherein, the physiological signal includes one or a combination of electrocardiographic signal, blood oxygen signal, impedance signal, and body surface pressure pulse wave signal.

When calculating the waveform quality factor of physiological signal waves, the individual waveform quality factors of physiological signal waveforms such as ECG, blood oxygen, impedance, and body surface pressure pulse waves can be calculated separately, or the two waveform quality factors can be combined based on the calculation of a single waveform quality factor. one or more waveform quality factors to obtain a composite waveform quality factor. The acquisition quality of the physiological signal waveform is high when there is no noise or the noise is small, and the acquisition quality of the physiological signal waveform is low when there is strong noise. For example, when the quality of the physiological signal waveform is good, the physiological signal waveform quality The factor is set to 1. In the case where there is no pure noise of the physiological signal, the quality factor of the physiological signal waveform is set to 0. In other noise environments, the quality factor of the physiological signal waveform is selected from 0 to 1 according to the size of the noise and the quality of the waveform. value between.

Please refer to Figure 4. The pulse wave transit time can be calculated from one or more physiological signals such as ECG, blood oxygen, impedance and body surface pressure pulse waves. One or more of the pressure pulse wave signals may also be obtained from two or more of the blood oxygen signal, the impedance signal and the body surface pressure pulse wave signal. Here, as an illustration, the pulse wave transit time is obtained based on the ECG signal and the blood oxygen signal. , impedance signal and impedance signal, body surface pressure pulse wave signal and body surface pressure pulse wave signal, blood oxygen signal and impedance signal, etc., which are not listed here due to space limitations). Analyze and extract the R wave position of the electrocardiographic signal 5 and the feature point position of the blood oxygen waveform 6 (which can be the peak, the trough, or the place with the largest rising slope, etc.), and then calculate the transmission time 7 of the pulse wave. A schematic diagram of the pulse wave transit time PTT_peak obtained from the electrical R point and the peak point of the oximetry waveform and the pulse wave transit time PTT_foot obtained from the ECG R point and the oximetry waveform valley point. In specific implementation, the pulse wave The transmission time of can also be obtained from other characteristic points of the ECG waveform 5 and the blood oxygen waveform 6 .

Before cuff blood pressure monitoring, it is necessary to use a linear fitting unit combined with reference blood pressure such as systolic blood pressure, diastolic blood pressure and mean blood pressure obtained from invasive blood pressure monitoring sensors or non-invasive blood pressure sensors based on Keshi sound method and oscillometric method The data and the pulse wave transit time were linearly fitted to obtain the corresponding fitted slope and intercept. The specific linear fitting method can be either the least square method or other linear fitting methods, and the pulse wave transit time and the systolic blood pressure, diastolic blood pressure and mean pressure in the reference blood pressure signal are linearly fitted respectively, and three pairs of Fit parameters, consisting of slope and intercept, respectively. When performing linear fitting according to the least square method, such as y=nx+b, where y represents the reference blood pressure value (systolic blood pressure, diastolic blood pressure or mean pressure), x represents the pulse wave transit time, and the pulse wave transit time and y _{systolic blood pressure} Fitting to obtain n _{systolic pressure} and b _{systolic pressure} , fitting of pulse wave transit time and y _{diastolic pressure} to obtain n _{diastolic pressure} and b _{diastolic pressure} , fitting of pulse wave transit time and y _{average pressure} to obtain n _{average pressure} and b _{average pressure} .

Finally, combined with the physiological signal waveform quality factor, the Kalman filter is performed on the cuff-free continuous monitoring systolic blood pressure, diastolic blood pressure and mean pressure obtained based on the pulse wave transit time. When the quality of the physiological signal waveform is high, the higher the waveform quality factor, The more the estimated value of blood pressure depends on the current blood pressure value, the lower the waveform quality factor is, the more the estimated value of blood pressure depends on the previous blood pressure value. When the waveform quality factor of the physiological signal is lower than the threshold, the blood pressure value is considered too unreliable. At this time, the update of the Kalman filtering process is not performed, so as to obtain a continuous high-precision blood pressure sequence. Thereby reducing the impact of occasional random noise on the measurement accuracy of systolic blood pressure, diastolic blood pressure and mean blood pressure without cuff.

Since the blood pressure monitoring means of the present invention is dynamic, continuous, and beat-by-beat, the blood pressure and pulse wave transmission time are time-series data related to the heartbeat cycle, that is, waveform data. The specific Kalman filtering process is as follows:

The Kalman filter is an optimal autoregressive data processing algorithm. For a discrete control process system, the system can be described by a linear stochastic difference equation:

X(n+1)=F*X(n)+B*U(n)+W(n)

Plus the system's measurements:

Y(n)=H*X(n)+V(n)

In the above two formulas, X(n) is the state of the system at time n, and U(n) is the control amount of the system at time n. F and B are system parameters. Y(n) is the measured value at time n, and H is the parameter of the measurement system. W(n) and V(n) represent process and measurement noise, respectively. They are assumed to be Gaussian white noise, and their covariances are Q, R, respectively.

In practical applications, the measured value Y(n) with noise is used to estimate the current state X(n+1) of the system. First, we use the previous state of the system to estimate the current state of the system:

In the formula is the optimal estimate of the system at time n with covariance P _n|n , is the prior estimate of the covariance P _n+1|n system at time n+1,

P _n+1|n ＝F·P _n|n ·F ^T +Q

The system update process is as follows:

G _n+1 ＝P _n+1|n · ^HT ·(H·P _n+1|n · ^HT +R) ^-1

P _n+1|n+1 ＝(IG _n+1 ·H)·P _n+1|n

In the formula, G _n+1 represents the Kalman gain. Combining the measured value and the prior estimate of the state, we can get the optimal estimate of the current state of the system

In cuffless blood pressure measurement and estimation, the blood pressure value BP is approximately negatively linearly correlated with the pulse wave transit time PTT, the pulse wave transit time PTT is the measurement variable, and the blood pressure value BP is the system state variable, the Kalman formula can be described as:

BP _n+1 = BP _n +Wn

PTT _n =αBP _n +β+Vn

Among them, BP _n is the blood pressure value of the nth time, Wn is noise, BP _n+1 is the blood pressure value of the n+1th time, α is a coefficient, β is a constant, Vn is noise, and PTT _n is the pulse wave transit time.

In order to make the estimated value of cuff blood pressure more from the waveform data and PTT data with good signal quality, we use the waveform quality factor SQI to correct the covariance R of the measurement noise,

Where R ₀ is 1, when the waveform quality factor SQI _n is high (close to 1) tends to 1, making the Kalman filtering result more dependent on the current measurement value. When the waveform quality factor SQI _n is low, R _n tends to infinity, making the Kalman filter reduce the Kalman gain, and the Kalman filtering result is more Much depends on the measured value of the previous blood pressure. In addition, when the waveform quality factor SQI _n is lower than a certain threshold (such as 0.5), it can be considered that the calculated value of the current PTT is very unreliable. At this time, the Kalman filter can not be used to update.

The error distributions of the cuffless continuous systolic blood pressure and diastolic blood pressure and the gold standard of invasive blood pressure monitoring obtained by the new method and the traditional method are shown in Fig. 5(a) and Fig. 5(b), respectively. It can be seen from the figure that The new method significantly reduces the error of cuffless continuous systolic blood pressure and diastolic blood pressure, and improves the measurement accuracy of cuffless continuous blood pressure based on pulse wave transit time.

Please refer to Fig. 3, in this embodiment, the cuffless blood pressure measuring system can also measure according to the following working process:

During calibration:

First, the user performs N measurements, and obtains physiological signal waves and reference blood pressure in each measurement;

Secondly, for each measured physiological signal wave, the pulse wave transit time online calculation unit 3 calculates its corresponding pulse wave transit time;

Again, the pulse wave transit time and the corresponding reference blood pressure are fitted by the fitting unit 1 to obtain the functional relationship between the two;

Measurement process:

Firstly, the physiological signal is measured by the physiological signal acquisition quality online calculation unit 2 to obtain a physiological signal wave, and the waveform quality factor of the physiological signal wave is calculated;

Secondly, for the physiological signal wave, the pulse wave transit time online calculation unit 3 calculates its corresponding pulse wave transit time, and continuously measures to obtain the corresponding pulse wave transit time sequence;

Finally, Kalman filtering is performed on the pulse wave transmission time series based on the waveform quality factor by the Kalman filter unit 4. When the waveform quality factor of the physiological signal is higher than the preset threshold, the higher the waveform quality factor, the higher the pulse wave transmission time. The more the estimated value depends on the current pulse wave transit time, the lower the waveform quality factor is, the more the estimated value of the pulse wave transit time depends on the previous pulse wave transit time. When the physiological signal waveform quality factor is lower than the threshold, the pulse wave is considered to be The transmission time of the pulse wave is too unreliable, so the Kalman filtering process is not updated at this time, so as to obtain a continuous high-precision pulse wave transmission time series;

The blood pressure calculation unit 5 uses the functional relationship to calculate the actual blood pressure sequence corresponding to the pulse wave transmission time series.

The present invention aims at the problem of poor accuracy of cuffless continuous blood pressure measurement based on physiological signal waveforms such as electrocardiogram, blood oxygen, impedance and body surface pressure pulse wave to obtain pulse wave transmission time. The cuffless continuous blood pressure signal obtained from the pulse wave transit time is subjected to Kalman filtering. When the quality of the physiological signal waveform is higher than the preset threshold, the higher the waveform quality factor, the more the estimated value of blood pressure depends on the current blood pressure value. The lower the waveform quality factor, the more the estimated value of blood pressure depends on the previous blood pressure value. When the physiological signal waveform quality factor is lower than the threshold, the blood pressure value is considered too unreliable. At this time, the update of the Kalman filtering process is not performed, thereby Obtain continuous high-precision blood pressure sequences, thereby reducing the influence of noise on the accuracy of blood pressure monitoring through physiological signal waveforms, and improving the measurement accuracy of cuffless, dynamic, and continuous blood pressure.

It should be noted that, in the accompanying drawings or in the text of the specification, implementations that are not shown or described are forms known to those of ordinary skill in the art, and are not described in detail. In addition, the above definitions of each element and method are not limited to the various specific structures, shapes or methods mentioned in the embodiments, and those of ordinary skill in the art can easily modify or replace them, for example:

(1) Pulse wave transit time can also be obtained from ECG signal and impedance signal, blood oxygen signal and blood oxygen signal, impedance signal and impedance signal, body surface pressure pulse wave signal and body surface pressure pulse wave signal, blood oxygen signal and impedance Signals, etc. (limited to space, not listed one by one) in the form of combination;

(2) The fitting between pulse wave transit time and invasive blood pressure, non-invasive blood pressure based on Keshi sound, or non-invasive blood pressure based on oscillometric method can be linear fitting, high-order polynomial fitting, logarithmic Fitting or nonlinear fitting such as cubic spline fitting instead;

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. The word "comprising" does not exclude the presence of elements or steps not listed in a claim.

The specific embodiments described above have further described the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above descriptions are only specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (12)

1. one kind is without cuff blood pressure measuring system, it is characterised in that the system includes：

Fitting unit (1), for carrying out Function Fitting with reference to blood pressure data and pulse wave transmission time；

Physiological signal acquisition quality is online computing unit (2), for online evaluation physiological signal waveform in routine work and life The waveform quality obtained under state, obtains the waveform quality factor；

The online computing unit of pulse wave transmission time (3), it is for calculating pulse wave transmission time from physiological signal waveform, described Pulse wave transmission time is a sequence；

Kalman filtering unit (4), for carrying out Kalman filtering to blood pressure or pulse wave transmission time；And

Blood pressure calculation unit (5), for calculating blood pressure according to pulse wave transmission time and fitting parameter, the blood pressure is a sequence Row.

2. measuring system according to claim 1, it is characterised in that the fitting unit includes：

First fitting subelement, measures for n times and physiological signal ripple and refers to pressure value, and based on measuring the life that obtains each time Reason signal wave, calculates its corresponding pulse wave transmission time；And

Second fitting subelement, for being fitted to pulse wave transmission time and corresponding reference pressure value, obtains both Functional relationship.

3. measuring system according to claim 1, it is characterised in that in the Kalman filtering unit, based on waveform Quality factor carries out Kalman filtering to blood pressure sequence, in a course of work：

When the physiological signal waveform quality factor is higher than default first threshold, it is divided into following two situations：

When the waveform quality factor is higher than default Second Threshold, increase the weight of current pressure value, pressure value is estimated, is obtained To filtered pressure value；

When the waveform quality factor is less than default Second Threshold, increase the weight of previous pressure value, pressure value estimated, Obtain filtered pressure value；

When the physiological signal waveform quality factor is less than the first threshold, the renewal of Kalman filtering process is not carried out, in the past One pressure value is used as current pressure value；

Wherein, the Kalman filtering unit continuously works, so as to obtain continuous high accuracy blood pressure sequence.

4. measuring system according to claim 1, it is characterised in that in the Kalman filtering unit, based on waveform Quality factor carries out Kalman filtering to pulse wave transmission time sequence, in a course of work,

When the physiological signal waveform quality factor is higher than default first threshold, it is divided into following two situations：

When the waveform quality factor is higher than default Second Threshold, increase the weight of current pulse ripple transmission time, pulse wave is passed The defeated time estimated, obtains filtered pulse wave transmission time；

When the waveform quality factor is less than default Second Threshold, increase the weight of previous pulse wave transmission time, to pulse wave Transmission time estimated, obtains filtered pulse wave transmission time；

When the physiological signal waveform quality factor is less than the first threshold, the renewal of Kalman filtering process is not carried out, in the past The pulse wave transmission time of one is used as current pulse wave transmission time；

Kalman filtering unit continuously works, so as to obtain continuous high accuracy pulse wave transmission time sequence.

5. measuring system according to claim 1, it is characterised in that the physiological signal includes electrocardiosignal, blood oxygen letter Number, one or more in impedance signal, body surface pressure pulse wave signal of combination.

6. measuring system according to claim 5, it is characterised in that the physiological signal includes electrocardiosignal, blood oxygen letter Number, the one kind in impedance signal, body surface pressure pulse wave signal, the waveform quality factor is the single waveform quality factor；Or Person

The physiological signal includes the various group in electrocardiosignal, blood oxygen signal, impedance signal, body surface pressure pulse wave signal Close, the waveform quality factor be on the basis of the single waveform quality factor is calculated, merge two or more waveform qualities because Son obtains the comprehensive waveform quality factor.

7. measuring system according to claim 1, it is characterised in that obtain quality in line computation list in the physiological signal In unit,

In the case of muting, the physiological signal waveform quality factor is set to into 1；

In the occasion for not having the pure noise of physiological signal, the physiological signal waveform quality factor is set to into 0；

Under other noise circumstances, the physiological signal waveform quality factor takes 0 to 1 according to the size of noise and the quality of waveform quality Between value.

8. measuring system according to claim 1, it is characterised in that in the online computing unit of the pulse wave transmission time In, by one or more acquisitions in electrocardiosignal and blood oxygen signal, impedance signal, body surface pressure pulse wave signal, Huo Zhecong Two or more middle acquisitions in blood oxygen signal, impedance signal and body surface pressure pulse wave signal.

9. measuring system according to claim 1, it is characterised in that the blood pressure obtained without cuff blood pressure measurement is for shrinking Pressure, diastolic pressure and mean pressure.

10. measuring system according to claim 1, it is characterised in that the reference pressure value includes systolic pressure, diastolic pressure And mean pressure, the systolic pressure, diastolic pressure and mean pressure in pulse wave transmission time and the reference pressure value be fitted respectively, Obtain three kinds of fitting function relations.

11. measuring systems according to claim 10, it is characterised in that the reference pressure value is by invasive monitoring of blood pressure side Method, Ke show that sound non-invasive blood pressure monitoring method or oscillographic method non-invasive blood pressure monitoring method are obtained.

12. measuring systems according to claim 3, it is characterised in that in the Kalman filtering unit, it is specific to block Thalmann filter is as follows：

Kalman filter is an optimization autoregression data processing algorithm, for the system of a discrete control process, should System is described with a linear stochastic difference equation：

X (n+1)=F*X (n)+B*U (n)+W (n),

Along with the measured value of system：

Y (n)=H*X (n)+V (n),

In above-mentioned two formula, X (n) is the system mode at n moment, and U (n) is controlled quentity controlled variable of the n moment to system, and F and B is system Parameter, Y (n) is the measured value at n moment, and H is the parameter of measuring system, and W (n) and V (n) represents process and making an uproar of measuring respectively Sound, they are assumed to white Gaussian noise, and their covariance is Q, R respectively；

Using with noisy measured value Y (n) come current state X (n+1) of estimating system, first, with a upper shape of system State carrys out the current state of estimating system：

${\hat{x}}_{n+1|n}=F\·{\hat{x}}_{n|n}+B\·u_n,$

In formulaBe system in moment n, covariance is P_n|nOptimal estimation,It is that covariance is P_n+1|nSystem is in n+1 The prior estimate at moment,

P_n+1|n=F P_n|n·F^T+Q

The renewal process of system is as follows：

G_n+1=P_n+1|n·H^T·(H·P_n+1|n·H^T+R)^-1

${\hat{x}}_{n+1|n+1}={\hat{x}}_{n+1|n}+G_{n+1}\·(y_{n+1}-H\·{\hat{x}}_{n+1|n})$

P_n+1|n+1=(I-G_n+1·H)·P_n+1|n

G in formula_n+1Kalman gain is represented, with reference to measured value and the priori estimates of state, the optimum of system current state is obtained Estimated value

In the blood pressure measurement without cuff with estimation, pressure value BP negative linear correlations approximate with pulse wave transmission time PTT, PTT It is measurand, BP is system state variables, Karman formula can be described as：

BP_n+1=BP_n+ Wn,

PTT_n=α BP_n+ β+Vn,

Wherein, BP_nFor the pressure value of n-th, Wn is noise, BP_n+1For the pressure value of (n+1)th time, α is coefficient, and β is constant, Vn For noise, PTT_nFor pulse wave transmission time,

In order that the estimated value without Tail cuff blood pressure is more from the good Wave data of signal quality and PTT data, using ripple Form quality amount factor S QI is modified to the covariance R of measurement noise,

$R_n=R_0\·\exp ({\mathrm{SQI}}_n^{-2}-1),$

R in formula₀For 1, when waveform quality factor S QI_nGao ShiTend to 1, make the filter result of Kalman more The measured value of current blood pressure is depended on more, when waveform quality factor S QI_nWhen low, R_nTend to infinitely great so that Kalman filtering Device reduces Kalman gain, and the result of Kalman filtering relies more heavily on the measured value of previous blood pressure.