A blood pressure monitoring device, storage medium and equipment in motion

By using the PPG waveform in the static state in the motion state to construct a blood pressure monitoring model, and using the characteristic parameter change relationship correction model in the transition state, the problem of low blood pressure monitoring accuracy in the motion state is solved, and more accurate blood pressure monitoring is achieved.

CN114052689BActive Publication Date: 2024-12-03SHANDONG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111487919.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-07
Publication Date
2024-12-03
Estimated Expiration
2041-12-07

AI Technical Summary

Technical Problem

The prior art has poor blood pressure monitoring accuracy in the state of exercise, especially because mechanical disturbances caused by human movement have a great interference with physiological signal acquisition, making it difficult to accurately extract useful information.

Method used

The PPG waveform acquisition module, static model construction module, motion feature parameter extraction module and blood pressure monitoring model correction module are used to construct the blood pressure monitoring model through PPG waveforms in the stationary state, and the characteristic parameter change relationship correction model in the transition state is used to improve motion Blood pressure monitoring accuracy in state.

Benefits of technology

The blood pressure monitoring accuracy in the exercise state is improved, the characteristic parameter estimation errors caused by motion noise are reduced, the accuracy of blood pressure prediction is enhanced, and the reliability of monitoring results is improved by eliminating individualized differences and environmental interference.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114052689B_ABST
    Figure CN114052689B_ABST
Patent Text Reader

Abstract

The present invention provides a blood pressure monitoring device, storage medium and equipment under motion state. The device includes a PPG waveform acquisition module for acquiring the PPG waveform of the subject from the static state to the motion state; a static model construction module for constructing a static blood pressure monitoring model based on the static PPG waveform; a motion characteristic parameter extraction module for directly extracting the characteristic parameters of the motion state PPG waveform, and judging one by one whether the noise coefficient corresponding to each characteristic parameter in the motion state is less than the corresponding noise threshold and the characteristic parameter value is within the corresponding preset threshold range, if so, the corresponding characteristic parameter remains unchanged; otherwise, according to the pre-constructed functional relationship between the characteristic parameters in the motion state and the characteristic parameters in the static state, the corresponding characteristic parameters in the motion state are updated; a blood pressure monitoring model correction module for using the corrected blood pressure monitoring model to monitor the blood pressure of the subject in the motion state.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of blood pressure monitoring devices, and in particular, relates to a blood pressure monitoring device, a storage medium and equipment in a motion state. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] At present, blood pressure measurement methods are mainly divided into two categories: non-continuous measurement and continuous measurement. Non-continuous measurement mainly includes blood pressure measurement devices used in hospitals and homes, such as electronic sphygmomanometers and mercury sphygmomanometers. Its principle is mainly based on the oscillometric method or Korotkoff sound method. Although it can measure blood pressure values ​​more accurately, the cuff needs to be inflated and deflated frequently, and it can only measure the blood pressure value at a certain moment, and it is impossible to perform continuous dynamic measurement of blood pressure. Continuous dynamic blood pressure measurement is divided into two categories: non-invasive and invasive. It has an early warning effect for potential critically ill patients with hypertension. It can remind patients to seek medical treatment in time by displaying abnormal blood pressure values, and try to avoid the serious harm caused by acute cardiovascular diseases. Invasive continuous blood pressure measurement is accompanied by serious complications, such as local infection and thrombosis. In addition, it is expensive and difficult to operate.

[0004] Non-invasive continuous blood pressure measurement methods mainly include arterial tension method, volume compensation method, and photoplethysmography (PPG) method, and the PPG-based method has gradually become a hot topic in the field of continuous blood pressure measurement. The pulse wave transit time (PTT) method or pulse wave velocity (PWV) method, which is more studied and more mature, requires another ECG signal and searches for characteristic points to obtain the functional relationship between PTT or PWV and blood pressure. However, this method requires simultaneous measurement of pulse wave signals and ECG signals, which is inconvenient to operate and the equipment is cumbersome and difficult to carry.

[0005] At present, the continuous blood pressure monitoring method based on PPG usually requires the subject to remain still for monitoring, and the blood pressure monitoring model in the static state is relatively complete. However, since the blood pressure in the human body in motion is higher than that in the static state, it is more harmful to the human body. Compared with the blood pressure in the static state, the blood pressure in the motion state has more clinical reference value. When the human body is in motion, various mechanical disturbances interfere greatly with the acquisition of physiological signals, and it is difficult to extract characteristic parameters from the PPG waveform. The inventors found that despite the use of signal processing methods such as filtering or feature extraction, the useful information that can be extracted is still very small, resulting in generally poor accuracy of blood pressure monitoring in motion. Summary of the invention

[0006] In order to solve the technical problems existing in the above-mentioned background technology, the present invention provides a blood pressure monitoring device, a storage medium and a device in a motion state, which can improve the accuracy of blood pressure monitoring in a motion state.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] A first aspect of the present invention provides a blood pressure monitoring device in a motion state, comprising:

[0009] A PPG waveform acquisition module, which is used to acquire the PPG waveform of the subject from a static state to a moving state; wherein the PPG waveform includes static state, transition state and moving state PPG waveforms;

[0010] A resting model building module, which is used to build a resting blood pressure monitoring model based on a resting PPG waveform;

[0011] The motion characteristic parameter extraction module is used to directly extract the characteristic parameters of the PPG waveform in the motion state, and judge one by one whether the noise coefficient corresponding to each characteristic parameter in the motion state is less than the corresponding noise threshold and the characteristic parameter value is within the corresponding preset threshold range. If so, the corresponding characteristic parameter remains unchanged; otherwise, according to the pre-established functional relationship between the characteristic parameters in the motion state and the characteristic parameters in the static state, the corresponding characteristic parameters in the motion state are updated;

[0012] The blood pressure monitoring model correction module is used to obtain a blood pressure difference by subtracting the resting reference blood pressure from the resting blood pressure calculated value obtained by using the resting blood pressure monitoring model, and then use the functional relationship between the blood pressure difference and each characteristic parameter of the PPG waveform from the resting state to the exercise state to correct the resting blood pressure monitoring model, and finally use the corrected blood pressure monitoring model to monitor the blood pressure of the subjects in the exercise state.

[0013] As an implementation manner, in the motion feature parameter extraction module, the pre-constructed functional relationship between the feature parameters in the motion state and the feature parameters in the static state is:

[0014] x i-motion =x i-static +h i (Δa, Δp, Δq);

[0015] Among them, x i-motion is the characteristic parameter in motion state, x i-static is the characteristic parameter in the static state; h i (Δa, Δp, Δq) are the characteristic parameters x in the transition state i The relationship between the change in the characteristic parameters relative to the static state and the parameters describing the intensity of movement in the transition state, Δa, Δp and Δq.

[0016] As an implementation mode, in the motion feature parameter extraction module, the motion intensity description parameter Δa=apresent -a static , Δp=p present -p static , Δq=q present -q static ;

[0017] Among them, a present 、p present and q present They represent the motion acceleration parameter a, the number of zero-crossing points of the first-order differential pulse wave signal, and the number of zero-crossing points of the second-order differential pulse wave signal respectively in the current state; a static 、p static and q static They respectively represent the motion acceleration parameter in a stationary state, the number of zero-crossing points of the first-order differential pulse wave signal, and the number of zero-crossing points of the second-order differential pulse wave signal.

[0018] As an embodiment, the blood pressure monitoring device in the exercise state further includes:

[0019] The individual difference elimination module is used to eliminate the individual differences of the PPG waveform.

[0020] As an implementation method, the individualized differences include wearing tightness and position correctness as well as quantification of skin characteristics.

[0021] As an implementation method, in the process of quantifying skin characteristics, the light source intensity adjustment value is calculated according to the model of skin characteristic coefficient and light source intensity; the adjusted light source intensity is I=I s *(a+b*ε), where a and b are matching coefficients, ε is the skin feature coefficient of the subject, I0 is the standard output light intensity, I1 is the standard output light intensity; I s is the intensity of the selected standard light source.

[0022] As an implementation mode, in the stationary model construction module, the stationary PPG waveform is also subjected to signal preprocessing, and then the feature parameters of the preprocessed stationary PPG waveform are extracted, and finally, the feature parameters whose Pearson correlation coefficient is greater than the set threshold are obtained through correlation analysis, and a corresponding feature parameter set is formed.

[0023] As an implementation mode, in the motion feature parameter extraction module, the noise coefficient β=c*k1+d*k2; c, and d are preset noise matching coefficients, k1 is the number of zero crossing points of the first-order differential signal, and k2 is the number of zero crossing points of the second-order differential signal.

[0024] A second aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps:

[0025] Collecting the PPG waveform of the subject from a static state to a moving state; wherein the PPG waveform includes static state, transition state and moving state PPG waveforms;

[0026] A resting blood pressure monitoring model was constructed based on the resting PPG waveform;

[0027] Directly extract the characteristic parameters of the PPG waveform in the motion state, and judge one by one whether the noise coefficient corresponding to each characteristic parameter in the motion state is less than the corresponding noise threshold and the characteristic parameter value is within the corresponding preset threshold range. If so, the corresponding characteristic parameter remains unchanged; otherwise, according to the pre-established functional relationship between the characteristic parameters in the motion state and the characteristic parameters in the static state, update the corresponding characteristic parameters in the motion state;

[0028] The resting reference blood pressure is subtracted from the resting blood pressure calculated using the resting blood pressure monitoring model to obtain the blood pressure difference. The resting blood pressure monitoring model is then corrected using the functional relationship between the blood pressure difference and each characteristic parameter of the PPG waveform from the resting state to the exercise state. Finally, the corrected blood pressure monitoring model is used to monitor the blood pressure of subjects in the exercise state.

[0029] A third aspect of the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the program:

[0030] Collecting the PPG waveform of the subject from a static state to a moving state; wherein the PPG waveform includes static state, transition state and moving state PPG waveforms;

[0031] A resting blood pressure monitoring model was constructed based on the resting PPG waveform;

[0032] Directly extract the characteristic parameters of the PPG waveform in the motion state, and judge one by one whether the noise coefficient corresponding to each characteristic parameter in the motion state is less than the corresponding noise threshold and the characteristic parameter value is within the corresponding preset threshold range. If so, the corresponding characteristic parameter remains unchanged; otherwise, according to the pre-established functional relationship between the characteristic parameters in the motion state and the characteristic parameters in the static state, update the corresponding characteristic parameters in the motion state;

[0033] The resting reference blood pressure is subtracted from the resting blood pressure calculated using the resting blood pressure monitoring model to obtain the blood pressure difference. The resting blood pressure monitoring model is then corrected using the functional relationship between the blood pressure difference and each characteristic parameter of the PPG waveform from the resting state to the exercise state. Finally, the corrected blood pressure monitoring model is used to monitor the blood pressure of subjects in the exercise state.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] (1) The present invention provides a judgment standard for signal quality under motion state, so that the characteristic parameters under motion state can be obtained more accurately, and the high-correlation characteristic parameters that cannot be accurately extracted under motion state are supplemented. The supplementation method is to find the difference between the characteristic parameters that cannot be accurately extracted due to motion noise and the corresponding characteristic parameters in the static state in the transition state, and establish a functional relationship with the description parameters of the intensity of exercise. In this way, the characteristic parameters that cannot be accurately extracted due to motion noise can be estimated through this functional relationship, so that the blood pressure prediction accuracy under motion state is improved.

[0036] (2) The present invention proposes a specific method to eliminate individual differences, including introducing a pressure sensor to make the force uniform and moderate, reducing ambient light interference, and reducing the impact of different skin characteristics on the measurement results by adjusting the incident light intensity.

[0037] (3) The continuous blood pressure monitoring model of the present invention has low requirements on system computing power, can be applied to small embedded systems, and is suitable for wearable portable devices. The present invention verifies the blood pressure change trend of subjects in a state of tension or emotional excitement.

[0038] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0040] Figure 1 This is a schematic diagram of eliminating individual differences in wearing styles according to an embodiment of the present invention;

[0041] FIG2( a ) is an original pulse wave signal of an embodiment of the present invention;

[0042] FIG2( b ) is a first-order differential pulse wave signal according to an embodiment of the present invention;

[0043] FIG2( c ) is a second-order differential pulse wave signal according to an embodiment of the present invention;

[0044] FIG3( a ) is a characteristic parameter corresponding to the original pulse wave signal of an embodiment of the present invention;

[0045] FIG3( b ) is a characteristic parameter corresponding to a first-order differential pulse wave signal according to an embodiment of the present invention;

[0046] Figure 4is a schematic diagram of a blood pressure monitoring device under exercise conditions according to an embodiment of the present invention;

[0047] FIG5(a) is a correlation verification between the SBP (Systolic Blood Pressure) calculated without using the present invention and the SBP measured by the electronic sphygmomanometer;

[0048] FIG5( b ) is a correlation verification between the SBP calculated by the present invention and the SBP measured by the electronic sphygmomanometer;

[0049] FIG5(c) is a correlation verification between DBP (diastolic pressure, i.e., low pressure in blood pressure, D is the abbreviation of diastolic period in English, BP stands for BLOOD PRESSURE, i.e., the abbreviation of blood pressure in English) calculated without using the present invention and DBP measured by an electronic sphygmomanometer;

[0050] FIG5( d ) is a correlation verification between the DBP calculated by the present invention and the DBP measured by the electronic sphygmomanometer;

[0051] FIG6( a ) shows the error range and proportion of SBP calculated without using the present invention;

[0052] FIG6( b ) is the error range and proportion of SBP calculated using the present invention;

[0053] FIG6( c ) shows the error range and proportion of DBP calculated without using the present invention;

[0054] FIG6( d ) shows the DBP error range and proportion calculated using the present invention. DETAILED DESCRIPTION

[0055] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0056] It should be noted that the following detailed descriptions are all illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0057] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0058] Under normal circumstances, a person's transition from static to dynamic or from dynamic to static is not two distinct states, but there is a transition state in the middle. If the human body's physiological signals are continuously monitored, it will be found that the transition state contains the signal change rules from the static state to the dynamic state. If these change rules are found, the physiological signals of the dynamic state can be obtained through the physiological signals of the static state and the transition state, thereby making up for the defect of poor accuracy in blood pressure monitoring due to the inability to obtain sufficiently accurate signals due to the large interference of the dynamic state.

[0059] Embodiment 1

[0060] like Figure 4 As shown, this embodiment provides a blood pressure monitoring device in a motion state, which includes a PPG waveform acquisition module, a static model construction module, a motion feature parameter extraction module and a blood pressure monitoring model correction module.

[0061] Due to the individual differences among subjects, in order to improve the accuracy of the blood pressure finally monitored by the blood pressure monitoring device under exercise, the blood pressure monitoring device under exercise also includes an individual difference elimination module, wherein the individual difference elimination module is used to eliminate the individual differences of the PPG waveform.

[0062] Specifically, the individual differences include wearing tightness and position accuracy as well as quantification of skin characteristics.

[0063] Individual differences among subjects, such as tightness of wearing, uneven wearing position causing interference from ambient light, skin color, differences in subcutaneous tissue composition, and different levels of skin hair, need to be eliminated. The main measures are as follows:

[0064] On the one hand, if Figure 1 As shown, thin-film pressure sensors are embedded at the four diagonals of the side of the wearable device that fits the skin. The pressure values ​​at each position are p1, p2, p3, and p4, respectively. The pressure threshold range at each position is (p imin , p imax )(1≤i≤4), when p1, p2, p3, p4 all meet the pressure threshold range of their respective positions (p imin ,p imax ), proceed to the next step, otherwise the user will be prompted to adjust the wearing tightness.

[0065] For example: the individual difference elimination module includes a wearing normalization module, and the wearing normalization module includes a pressure sensor and a microprocessor. The pressure sensor is arranged at the four diagonals of the side of the blood pressure monitoring device that is in contact with the skin in the exercise state, and is used to monitor the pressure value at each diagonal and transmit it to the microprocessor. The microprocessor is used to determine whether the pressure value at each diagonal meets the pressure threshold range of the corresponding position. If so, it is determined that the wearing tightness meets the requirements.

[0066] And the microprocessor is used to measure the difference Δp of the pressure values ​​at any two diagonal points m,n , let Δp m,n =p m -p n , (1≤m,n≤4,m≠n), if Δp m,n Satisfy the threshold range (Δp min , Δp max ), it is considered that the wearing position is correct and the wearing tightness meets the requirements, and measurement can be performed.

[0067] On the other hand, the skin characteristics of different people are quantified to obtain the skin characteristic coefficient ε, which is defined by the following method: When the standard incident light intensity is I s When the standard output light intensity is I0. However, due to individual differences, when the standard incident light intensity is I s When the standard output light intensity is I1, then The skin characteristic coefficient of the current subject is calculated based on the model of skin characteristic coefficient and light source intensity to obtain the light source intensity adjustment value. The adjusted light source intensity is I = I s *(a+b*ε), where a and b are matching coefficients, ε is the skin feature coefficient of the subject, I s is the intensity of the selected standard light source.

[0068] In a specific implementation, a PPG waveform acquisition module is used to acquire the PPG waveform of a subject from a static state to a motion state; wherein the PPG waveform includes static state, transition state and motion state PPG waveforms.

[0069] An electronic sphygmomanometer is used to collect the reference blood pressure value of the subject in a static state, and a photoelectric sensor device is used to collect the PPG waveform of the subject in a static state; a photoelectric sensor device is used to collect the PPG waveform of the subject in a transition state and a motion state.

[0070] In a specific implementation, a resting model construction module is used to construct a resting blood pressure monitoring model based on a resting PPG waveform.

[0071] In the static model building module, a static blood pressure monitoring model is constructed according to the relationship between the characteristic parameters of the static PPG waveform and the static reference blood pressure value.

[0072] In the stationary model construction module, the stationary PPG waveform is also subjected to signal preprocessing, and then the characteristic parameters of the preprocessed stationary PPG waveform are extracted. Finally, the characteristic parameters whose Pearson correlation coefficient is greater than the set threshold are obtained through correlation analysis, and a corresponding characteristic parameter set is formed.

[0073] Specifically, the PPG waveform in the static state is preprocessed to obtain the original pulse wave signal M, the first-order differential pulse wave signal N, and the second-order differential pulse wave signal Q, respectively. Figure 2(a) to Figure 2(c) As shown. Using the local maximum peak search algorithm, feature points of the three types of signals M, N, and Q are identified and extracted respectively. Based on these feature points, the corresponding feature parameters are calculated. After correlation analysis, feature parameters with a Pearson correlation coefficient greater than a set threshold (for example, 0.6) are obtained, and a feature parameter set X = (x 1-static ,x 2-static …x n-static ), where n-static represents the number of characteristic points in the static state. Find the relationship between these characteristic parameters and the static reference blood pressure value, and establish the blood pressure monitoring model BP in the static state static =f(x1, x2, ...x n ).

[0074] In this embodiment, the signal preprocessing methods used include low-pass filtering with different window lengths, sliding mean filtering, sliding median filtering and sliding weighted filtering.

[0075] Sliding mean filtering algorithm: N is the window length value;

[0076] Sliding median filtering algorithm: y(k)=Med{x(k), x(k+1), ...x(k+N-1)}, 0≤k≤N, N is the window length value.

[0077] In this step, the main feature points that need to be identified are: the main wave peak point (a), the dicrotic wave peak point (b), the trough point (c), and the trough point (d) of the descending gorge of the original pulse wave signal as shown in Figure 3(a); the main wave peak point (e) and various zero-crossing points (f1, f2, f3, f4) of the first-order differential pulse wave signal as shown in Figure 3(b); the peak point (g) of the second-order difference and various zero-crossing points (h1, h2, h3, h4).

[0078] The Pearson correlation coefficient is calculated as follows:

[0079]

[0080] Where X represents the reference blood pressure BP, Y represents a characteristic parameter that needs to be analyzed for correlation, Cov(X,Y) represents the covariance, Var[X] represents the variance of X, and Var[Y] represents the variance of Y.

[0081] The characteristic parameters with Pearson correlation coefficient greater than 0.6 are listed as follows:

[0082] The cardiac cycle of the original PPG signal is T1; the diastolic time is T2; the proportion of diastolic time is T2 / T1; the time difference between the peak point of the main wave of the original PPG signal and the peak point of the second-order difference is T3; the time difference between the peak point of the main wave and the peak point of the dicrotic wave is T4; the systolic time is T5; the proportion of systolic time is T5 / T1; the maximum slope of the rising branch of the main wave of the original signal MS (corresponding to the value of the main wave peak point e of the first-order differential PPG signal); the height of the main wave peak point is H1; the height of the dicrotic wave peak point is H2; the height of the descending isthmus is H3; the height of the trough point is H4; the height of the main wave peak point of the first-order differential pulse wave signal is H5; the area under the systolic curve is S1; the area under the diastolic curve is S2; the cardiovascular characteristic quantity K, in, Among them, H i is the height value of all points in a cardiac cycle.

[0083] In a specific implementation, the motion characteristic parameter extraction module is used to directly extract the characteristic parameters of the PPG waveform in the motion state, and judge one by one whether the noise coefficient corresponding to each characteristic parameter in the motion state is less than the corresponding noise threshold and the characteristic parameter value is within the corresponding preset threshold range. If so, the corresponding characteristic parameter remains unchanged; otherwise, the corresponding characteristic parameter in the motion state is updated according to the pre-constructed functional relationship between the characteristic parameters in the motion state and the characteristic parameters in the static state.

[0084] Among them, the feature parameters in the motion state can be extracted by using a feature parameter extraction method such as a feature parameter extraction method in a static state; or feature points can be directly identified and feature parameters calculated.

[0085] The quality of the PPG signal collected under the motion state is evaluated, and the characteristic parameters directly extracted under the motion state are: X = (x 1-motion ,x 2-motion …x n-motion ).

[0086] Among them, for the signal quality, the noise coefficient and motion feature parameter threshold are introduced for judgment.

[0087] The noise coefficient calculation formula is β = c*k1 + d*k2, where c and d are matching coefficients. k1 is the number of zero-crossing points of the first-order differential signal, and k2 is the number of zero-crossing points of the second-order differential signal. For the characteristic parameter x in motioni (1≤i≤n), β is obtained according to the above noise coefficient calculation formula, and β is set according to experience in the motion state relative to the characteristic parameter x i The threshold value is V ith ; define x i The value of v i , v is set based on experience i The numerical threshold range in the motion state is (th vimin , th vimax ). If (1) the noise coefficient β is less than the threshold V ith And satisfy (2)v i In the threshold range (th vimin , th vimax ), then the signal quality is considered to be i It meets the requirements and it is considered that this characteristic parameter x i The extraction of is accurate. Otherwise, the signal quality is considered to be i It does not meet the requirements and needs to update the feature parameter x i :

[0088] First, the motion acceleration parameter a is measured by a three-axis accelerometer, and the number of zero-crossing points of the first-order differential pulse wave signal N is set to p, and the number of zero-crossing points of the second-order differential pulse wave signal Q is set to q. Now define three motion intensity description parameters Δa=a present -a static , Δp=p present -p static , Δq=q present -q static Among them, a present 、p present and q present They represent the motion acceleration parameter a in the current state, the zero-crossing number parameter p of the first-order differential pulse wave signal N, and the zero-crossing number parameter q of the second-order differential pulse wave signal Q; a static 、p static and q static They respectively represent the motion acceleration parameter a in a stationary state, the zero-crossing point number parameter p of the first-order differential pulse wave signal N, and the zero-crossing point number parameter q of the second-order differential pulse wave signal Q.

[0089] Secondly, when the value of the motion acceleration parameter a is va≤va1, the subject is considered to be in a stationary state; when va1<va<va2, the subject is considered to be in a transition state; when va≥va2, the subject is considered to be in a moving state.

[0090] For a characteristic parameter x to be updated in the motion state i(1≤i≤n), x in static state, transition state and motion state i Respectively represented as x i-static 、x i-transition and x i-motion , then x in the transition state i-transition =x i-static +h i (Δa, Δp, Δq), where h i (Δa, Δp, Δq) = x i-transition -x i-static is the characteristic parameter x obtained in the transition state i The change relative to the static state (x i-transition -x i-static ) and the parameters describing the intensity of motion during the transition state, Δa, Δp, Δq. i When (Δa, Δp, Δq) is applied to the motion state, this relationship represents the x obtained in the motion state. i The change relative to the static state (x i-motion -x i-static ) and the relationship between the parameters describing the intensity of the movement in the motion state Δa, Δp, Δq. In summary, the inaccurate feature parameters x are extracted in the motion state. j-motion You can calculate x like this: i-motion =x i-static +h i (Δa, Δp, Δq).

[0091] In summary, the characteristic parameter set X = (x 1-motion ,x 2-motion …x n-motion ) in the feature parameter x to be updated i After the above steps, the update is performed to achieve the goal of completing n feature parameters.

[0092] In the specific implementation, the blood pressure monitoring model correction module is used to first obtain the resting reference blood pressure, then subtract it from the resting blood pressure calculated value obtained by the resting blood pressure monitoring model to obtain the blood pressure difference, and then use the functional relationship between the blood pressure difference and each characteristic parameter of the PPG waveform from the resting state to the exercise state to correct the resting blood pressure monitoring model, and finally use the corrected blood pressure monitoring model to monitor the blood pressure of the subject in the exercise state. The specific implementation process is as follows.

[0093] At rest, the reference blood pressure (BP static-ref , that is, the blood pressure measured by the electronic sphygmomanometer at rest) and the use of model BP static =f(x1, x2, ...x n ) Calculated blood pressure BP static-calThe difference is ΔBP=BP static-ref -BP static-cal . Find ΔBP and each characteristic parameter x i The functional relationship between them is ΔBP=g(x1,x2,...x n ). The modified blood pressure monitoring model for exercise status is: BP motion =BP static + ΔBP = f(x1, x2, ... x n )+g(x1,x2,...x n ).

[0094] In order to obtain the set X=(x 1-motion ,x 2-motion …x n-motion ) and extract as many accurate feature parameters x as possible under the motion state j-motion (1≤j≤m, m is unknown and its size is determined by the signal quality of each person). It is necessary to adjust the window length of the sliding mean filter and sliding median filter algorithms and the cutoff frequency of the low-pass filter. The specific measures are to increase the window length of the sliding mean filter and sliding median filter algorithms and reduce the cutoff frequency of the low-pass filter. (If it is necessary to specify the specific values, you can increase 2 window length values ​​and reduce the cutoff frequency of 2Hz based on experience).

[0095] When actually measuring blood pressure, use the blood pressure monitoring model:

[0096] BP motion =BP static + ΔBP = f(x1, x2, ... x n )+g(x1,x2,...x n ) to measure.

[0097] Figure 5(a)-Figure 5(d) This is a consistency test and correlation analysis of the blood pressure calculated using the blood pressure model algorithm and the blood pressure measured by the electronic sphygmomanometer. By comparing Figure 5(a) and Figure 5(b), it can be found that the correlation coefficient of Figure 5(b) is higher, so the accuracy of calculating SBP using the present invention is higher; by comparing Figure 5(c) and Figure 5(d), it can be found that the correlation coefficient of B-2 is higher, so the accuracy of calculating DBP using the present invention is higher.

[0098] Figure 6(a)-Figure 6(d)It is a schematic diagram of the range of the difference between the blood pressure calculated using the blood pressure model algorithm and the blood pressure measured by the electronic sphygmomanometer. By comparing Figure 6(a) and Figure 6(b), it can be found that the difference of SBP in Figure 6(b) accounts for a higher proportion in 0-5mmHg and 5-10mmHg, so the accuracy of SBP calculated using the present invention is better; by comparing Figure 6(c) and Figure 6(d), it can be found that the difference of DBP in Figure 6(d) accounts for a higher proportion in 0-5mmHg and a smaller proportion in 10-15mmHg, so the accuracy of DBP calculated using the present invention is better.

[0099] Embodiment 2

[0100] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the following steps are implemented:

[0101] Collecting the PPG waveform of the subject from a static state to a moving state; wherein the PPG waveform includes static state, transition state and moving state PPG waveforms;

[0102] A resting blood pressure monitoring model was constructed based on the resting PPG waveform;

[0103] Directly extract the characteristic parameters of the PPG waveform in the motion state, and judge one by one whether the noise coefficient corresponding to each characteristic parameter in the motion state is less than the corresponding noise threshold and the characteristic parameter value is within the corresponding preset threshold range. If so, the corresponding characteristic parameter remains unchanged; otherwise, according to the pre-established functional relationship between the characteristic parameters in the motion state and the characteristic parameters in the static state, update the corresponding characteristic parameters in the motion state;

[0104] The resting reference blood pressure is subtracted from the resting blood pressure calculated using the resting blood pressure monitoring model to obtain the blood pressure difference. The resting blood pressure monitoring model is then corrected using the functional relationship between the blood pressure difference and each characteristic parameter of the PPG waveform from the resting state to the exercise state. Finally, the corrected blood pressure monitoring model is used to monitor the blood pressure of subjects in the exercise state.

[0105] Embodiment 3

[0106] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the program:

[0107] Collecting the PPG waveform of the subject from a static state to a moving state; wherein the PPG waveform includes static state, transition state and moving state PPG waveforms;

[0108] A resting blood pressure monitoring model was constructed based on the resting PPG waveform;

[0109] Directly extract the characteristic parameters of the PPG waveform in the motion state, and judge one by one whether the noise coefficient corresponding to each characteristic parameter in the motion state is less than the corresponding noise threshold and the characteristic parameter value is within the corresponding preset threshold range. If so, the corresponding characteristic parameter remains unchanged; otherwise, according to the pre-established functional relationship between the characteristic parameters in the motion state and the characteristic parameters in the static state, update the corresponding characteristic parameters in the motion state;

[0110] The resting reference blood pressure is subtracted from the resting blood pressure calculated using the resting blood pressure monitoring model to obtain the blood pressure difference. The resting blood pressure monitoring model is then corrected using the functional relationship between the blood pressure difference and each characteristic parameter of the PPG waveform from the resting state to the exercise state. Finally, the corrected blood pressure monitoring model is used to monitor the blood pressure of subjects in the exercise state.

[0111] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0112] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A blood pressure monitoring device for exercise, characterized in that: include: A PPG waveform acquisition module, which is used to acquire the PPG waveform of the subject from a static state to a moving state; wherein the PPG waveform includes static state, transition state and moving state PPG waveforms; A resting model building module, which is used to build a resting blood pressure monitoring model based on a resting PPG waveform; The motion characteristic parameter extraction module is used to directly extract the characteristic parameters of the PPG waveform in the motion state, and judge one by one whether the noise coefficient corresponding to each characteristic parameter in the motion state is less than the corresponding noise threshold and the characteristic parameter value is within the corresponding preset threshold range. If so, the corresponding characteristic parameter remains unchanged; otherwise, according to the pre-established functional relationship between the characteristic parameters in the motion state and the characteristic parameters in the static state, the corresponding characteristic parameters in the motion state are updated; A blood pressure monitoring model correction module, which is used to obtain a blood pressure difference by subtracting a resting reference blood pressure from a resting blood pressure calculated value obtained by using a resting blood pressure monitoring model, and then correct the resting blood pressure monitoring model using a functional relationship between the blood pressure difference and each characteristic parameter of the PPG waveform from a resting state to an exercise state, and finally use the corrected blood pressure monitoring model to monitor the blood pressure of the subject in an exercise state; In the motion feature parameter extraction module, the pre-constructed functional relationship between the feature parameters in the motion state and the feature parameters in the static state is: i-motion =x i-static +h i (Δa, Δp, Δq); where, x i-motion is the characteristic parameter in motion state, x i-static is the characteristic parameter in the static state; h i (Δa, Δp, Δq) are the characteristic parameters x in the transition state i The relationship between the change of the characteristic parameters relative to the static state and the parameters describing the intensity of the motion in the transition state, Δa, Δp and Δq; In the motion feature parameter extraction module, the motion intensity description parameter Δa=a present -a static , Δp=p present -p static , Δq=q present -q static ; Among them, a present 、p present and q present They represent the motion acceleration parameter a, the number of zero-crossing points of the first-order differential pulse wave signal, and the number of zero-crossing points of the second-order differential pulse wave signal respectively in the current state; a static 、p static and q static They respectively represent the motion acceleration parameter in a stationary state, the number of zero-crossing points of the first-order differential pulse wave signal, and the number of zero-crossing points of the second-order differential pulse wave signal; In the motion feature parameter extraction module, the noise coefficient = c*k1+d*k2; c, and d are preset noise matching coefficients, k1 is the number of zero-crossing points of the first-order differential signal, and k2 is the number of zero-crossing points of the second-order differential signal.

2. The blood pressure monitoring device during exercise as claimed in claim 1, characterized in that: The blood pressure monitoring device in the exercise state further comprises: The individual difference elimination module is used to eliminate the individual differences of the PPG waveform.

3. The blood pressure monitoring device during exercise as claimed in claim 2, characterized in that: The individual differences include wearing tightness and positioning accuracy as well as quantification of skin characteristics.

4. The blood pressure monitoring device during exercise as claimed in claim 3, characterized in that: In the process of quantifying skin characteristics, the light source intensity adjustment value is calculated according to the model of skin characteristic coefficient and light source intensity; the adjusted light source intensity is I = I s *(a+b*ε), where a and b are matching coefficients, ε is the skin feature coefficient of the subject, I0 is the standard output light intensity, I1 is the standard output light intensity; I s is the intensity of the selected standard light source.

5. The blood pressure monitoring device during exercise as claimed in claim 1, characterized in that: In the stationary model construction module, the stationary PPG waveform is also subjected to signal preprocessing, and then the characteristic parameters of the preprocessed stationary PPG waveform are extracted. Finally, the characteristic parameters whose Pearson correlation coefficient is greater than the set threshold are obtained through correlation analysis, and a corresponding characteristic parameter set is formed.

6. A computer-readable storage medium of the blood pressure monitoring device during exercise as claimed in claim 1, wherein a computer program is stored thereon, characterized in that: When the program is executed by the processor, the following steps are implemented: Collecting the PPG waveform of the subject from a static state to a moving state; wherein the PPG waveform includes static state, transition state and moving state PPG waveforms; A resting blood pressure monitoring model was constructed based on the resting PPG waveform; Directly extract the characteristic parameters of the PPG waveform in the motion state, and judge one by one whether the noise coefficient corresponding to each characteristic parameter in the motion state is less than the corresponding noise threshold and the characteristic parameter value is within the corresponding preset threshold range. If so, the corresponding characteristic parameter remains unchanged; otherwise, according to the pre-established functional relationship between the characteristic parameters in the motion state and the characteristic parameters in the static state, update the corresponding characteristic parameters in the motion state; The resting reference blood pressure is subtracted from the resting blood pressure calculated using the resting blood pressure monitoring model to obtain the blood pressure difference. The resting blood pressure monitoring model is then corrected using the functional relationship between the blood pressure difference and each characteristic parameter of the PPG waveform from the resting state to the exercise state. Finally, the corrected blood pressure monitoring model is used to monitor the blood pressure of subjects in the exercise state.

7. A computer device for a blood pressure monitoring device during exercise as claimed in claim 1, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the following steps are implemented: Collecting the PPG waveform of the subject from a static state to a moving state; wherein the PPG waveform includes static state, transition state and moving state PPG waveforms; A resting blood pressure monitoring model was constructed based on the resting PPG waveform; Directly extract the characteristic parameters of the PPG waveform in the motion state, and judge one by one whether the noise coefficient corresponding to each characteristic parameter in the motion state is less than the corresponding noise threshold and the characteristic parameter value is within the corresponding preset threshold range. If so, the corresponding characteristic parameter remains unchanged; otherwise, according to the pre-established functional relationship between the characteristic parameters in the motion state and the characteristic parameters in the static state, update the corresponding characteristic parameters in the motion state; The resting reference blood pressure is subtracted from the resting blood pressure calculated using the resting blood pressure monitoring model to obtain the blood pressure difference. The resting blood pressure monitoring model is then corrected using the functional relationship between the blood pressure difference and each characteristic parameter of the PPG waveform from the resting state to the exercise state. Finally, the corrected blood pressure monitoring model is used to monitor the blood pressure of subjects in the exercise state.