Method and apparatus for predicting blood pressure, electronic device, and readable storage medium

Abstract

A method and apparatus for predicting blood pressure, an electronic device, and a readable storage medium, relating to the technical field of medical treatment. The method for predicting blood pressure comprises: synchronously acquiring physiological signals, wherein the physiological signals comprise an electrocardiographic signal, a pulse signal, and a heart sound signal; performing feature extraction on the physiological signals to obtain signal features, wherein the signal features comprise an pre-ejection period duration and a pulse transmission time; and inputting the signal features into a pre-trained blood pressure prediction model, so that the blood pressure prediction model outputs a blood pressure prediction result, thereby improving the prediction accuracy of blood pressure.

Description

Blood pressure prediction method and device, electronic equipment and readable storage medium

[0001] The present application claims priority to the Chinese patent application No. 202410453505.7, filed on April 15, 2024, and entitled "Blood pressure prediction method and device, electronic equipment and readable storage medium", the whole content of which is incorporated herein by reference. TECHNICAL FIELD

[0002] The present application relates to the technical field of medicine, in particular to a blood pressure prediction method and device, electronic equipment and readable storage medium. BACKGROUND

[0003] The cuffless blood pressure measurement device mainly estimates blood pressure through pulse wave transmission time. Pulse wave transmission time (PWTT) is defined as the time required for cardiac ejection to reach the distal point at the same time. The method of obtaining pulse wave transmission time is usually to synchronously collect electrocardiogram signal and pulse wave signal, taking the R wave peak of electrocardiogram signal as the starting point and taking the pulse wave feature point as the ending point. The time difference of this segment is the pulse wave transmission time.

[0004] However, the R wave peak is not the time when the heart starts to contract. There is a preparation time before the heart starts to contract, which is called the preejection period (PEP). Due to the existence of the preejection period, the blood pressure result estimated based on the pulse wave transmission time is unreliable.

[0005] The disclosure of the above background art content is only used to assist in understanding the inventive concept and technical solutions of the present application. It does not necessarily belong to the prior art of the present patent application, nor does it necessarily give technical teaching. It is provided as general background information and does not necessarily constitute prior art. SUMMARY

[0006] The main purpose of the present application is to provide a blood pressure prediction method and device, electronic equipment and readable storage medium, aiming to solve the technical problem of how to improve the prediction accuracy of blood pressure on the server.

[0007] To achieve the above purpose, the present application provides a blood pressure prediction method, which comprises:

[0008] synchronously collecting physiological signals, wherein the physiological signals include electrocardiogram signals, pulse signals and heart sound signals;

[0009] performing feature extraction on the physiological signals to obtain signal features, wherein the signal features include preejection period duration and pulse wave transmission time;

[0010] inputting the signal features into a pre-trained blood pressure prediction model, so that the blood pressure prediction model outputs a blood pressure prediction result.

[0011] Optionally, the step of extracting features from the physiological signals comprises:

[0012] determining a QRS peak of the electrocardiogram signal, determining a heart sound peak of the heart sound signal, determining a time difference between the two peaks, and taking the time difference as a pre-ejection period duration;

[0013] acquiring a pulse transit time based on the pulse signal and the electrocardiogram signal.

[0014] Optionally, the signal features further comprise a systolic duration ratio and a diastolic duration ratio, and the step of extracting features from the physiological signals further comprises:

[0015] taking a systolic duration corresponding to the pulse signal as a first systolic duration, and taking a diastolic duration corresponding to the pulse signal as a first diastolic duration;

[0016] segmenting the heart sound signal to obtain a systolic duration and a diastolic duration corresponding to the heart sound signal;

[0017] taking a systolic duration corresponding to the heart sound signal as a second systolic duration, and taking a diastolic duration corresponding to the heart sound signal as a second diastolic duration;

[0018] taking a ratio of the first systolic duration and the second systolic duration as a systolic duration ratio;

[0019] taking a ratio of the first diastolic duration and the second diastolic duration as a diastolic duration ratio.

[0020] Optionally, the step of segmenting the heart sound signal to obtain a systolic duration and a diastolic duration corresponding to the heart sound signal comprises:

[0021] acquiring a first threshold and a second threshold corresponding to the heart sound signal;

[0022] determining an upper envelope of the heart sound signal, and filtering the upper envelope to obtain a heart sound envelope signal;

[0023] taking a duration window in which a signal value of the heart sound envelope signal is greater than the first threshold as a first duration window;

[0024] a start time of the first duration window as a first start time, and an end time of the first duration window as a first end time;

[0025] a duration window of the heart sound envelope signal with a signal value greater than the second threshold value as a second duration window;

[0026] a start time of the second duration window as a second start time, and an end time of the first duration window as a second end time;

[0027] a time difference between the first end time and the second start time as a systolic duration corresponding to the heart sound signal;

[0028] a time difference between the first start time and the second end time as a diastolic duration corresponding to the heart sound signal.

[0029] Optionally, before the step of obtaining the first threshold value and the second threshold value corresponding to the heart sound signal, the method further comprises:

[0030] determining a QRS peak point of the electrocardiogram signal, and taking a time corresponding to the QRS peak point as a first time;

[0031] determining a trough point of the pulse signal, and taking a time corresponding to the trough point as a second time;

[0032] determining a first zero-crossing point and a second zero-crossing point of the pulse signal, determining a third time corresponding to the first zero-crossing point, and determining a fourth time corresponding to the second zero-crossing point;

[0033] determining a first threshold value corresponding to the first time and the second time, and determining a second threshold value corresponding to the third time and the fourth time.

[0034] Optionally, the step of determining the first threshold value corresponding to the first time and the second time comprises:

[0035] determining a first peak value of the heart sound envelope signal between the first time and the second time;

[0036] adjusting the first peak value based on a preset adjustment coefficient to obtain the first threshold value, wherein the first threshold value is less than the first peak value.

[0037] Optionally, the step of determining the second threshold value corresponding to the third time and the fourth time comprises:

[0038] determining a second peak value of the heart sound envelope signal between the third time and the fourth time;

[0039] adjust the second peak value based on a preset adjustment coefficient to obtain a second threshold value, wherein the second threshold value is less than the second peak value.

[0040] In addition, to achieve the above object, the present application also provides a blood pressure prediction device, which comprises:

[0041] a physiological signal acquisition module, configured to synchronously acquire physiological signals, wherein the physiological signals comprise electrocardiogram signals, pulse signals and heart sound signals;

[0042] a feature extraction module, configured to perform feature extraction on the physiological signals to obtain signal features, wherein the signal features comprise pre-ejection duration and pulse transit time;

[0043] a blood pressure prediction module, configured to input the signal features into a pre-trained blood pressure prediction model, so that the blood pressure prediction model outputs a blood pressure prediction result.

[0044] The present application also provides an electronic device, which is a physical device, and comprises at least one processor and a memory in communication connection with the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the steps of the blood pressure prediction method.

[0045] The present application also provides a readable storage medium, which is a computer readable storage medium, and stores a program for implementing a blood pressure prediction method, and the program is executed by a processor to implement the steps of the blood pressure prediction method.

[0046] The present application also provides a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the steps of the blood pressure prediction method.

[0047] In the present application, physiological signals are synchronously acquired, wherein the physiological signals comprise electrocardiogram signals, pulse signals and heart sound signals; feature extraction is performed on the physiological signals to obtain signal features, wherein the signal features comprise pre-ejection duration and pulse transit time; and the signal features are input into a pre-trained blood pressure prediction model, so that the blood pressure prediction model outputs a blood pressure prediction result. In this way, compared with the prior art, the present application embodiment predicts the blood pressure of a subject based on fused signal features such as pre-ejection duration and pulse transit time, takes into account the influence of pre-ejection duration on blood pressure, and thus improves the prediction accuracy of blood pressure. BRIEF DESCRIPTION OF DRAWINGS

[0048] The accompanying drawings, which are incorporated herein and constitute part of the specification, illustrate embodiments consistent with the application and, together with the description, serve to explain the principles of the application.

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings are only some embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0050] Fig. 1 is a flowchart of a first embodiment of the blood pressure prediction method of the present application;

[0051] Fig. 2 is a schematic diagram of the overall architecture of the wearable device system of the present application;

[0052] Fig. 3 is a schematic diagram of the structure of the wearable device of the present application;

[0053] Fig. 4 is a schematic diagram of the prediction process of the blood pressure prediction method of the present application;

[0054] Fig. 5 is another schematic diagram of the prediction process of the blood pressure prediction method of the present application;

[0055] Fig. 6 is a schematic diagram of the physiological signals of the blood pressure prediction method of the present application;

[0056] Fig. 7 is a schematic diagram of the device modules of the blood pressure prediction device of the present application;

[0057] Fig. 8 is a schematic diagram of the device structure of the hardware operating environment involved in the blood pressure prediction device in the embodiments of the present application.

[0058] The purposes, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0059] In order to make the above-mentioned purposes, features and advantages of the present application more apparent and easy to understand, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0060] Embodiment One

[0061] Traditional blood pressure measurement methods are mostly based on the oscillographic method of cuff-type electronic sphygmomanometer. However, the inflation and deflation of the cuff during the measurement process can cause great discomfort to the user, and the measurement device is large in size and not portable. In recent years, the emergence of cuffless blood pressure measurement devices has solved the discomfort of the cuff and greatly reduced the size. With the increasing number of users of wearable devices in recent years, people's demand for measuring blood pressure by wearable devices has also increased, especially some cuffless measurement devices are integrated into smart watches, which improves the popularity of users and the real-time measurement, but the accuracy of the current cuffless blood pressure measurement device needs to be improved.

[0062] The cuffless blood pressure measurement device mainly estimates blood pressure through pulse wave transit time. Pulse wave transit time (PWTT) is defined as the time required for the same time cardiac ejection to reach the distal point. The method of obtaining pulse wave transit time is to synchronously collect electrocardiogram signal and pulse wave signal, taking the R wave peak of electrocardiogram signal as the starting point and taking the pulse wave feature point as the end point. The time difference of this segment is the pulse wave transit time.

[0063] However, in fact, the R wave peak is not the time when the heart starts to contract. There is a preparation time before the heart starts to contract, which is called the preejection period (PEP). Due to the existence of the preejection period, the blood pressure result estimated based on the pulse wave transit time is unreliable.

[0064] Based on the above problems, the blood pressure prediction method of the first embodiment is proposed. Please refer to FIG. 1, the blood pressure prediction method comprises:

[0065] Step S10, synchronously collecting physiological signals, wherein the physiological signals include electrocardiogram signals, pulse signals and heart sound signals;

[0066] In this embodiment, the blood pressure prediction method is applied to a wearable device, which can be a smart watch, a smart bracelet, etc.

[0067] For example, referring to FIGS. 2-3, the wearable device comprises a housing 1 and a sensor arranged on the housing, the sensor being used for collecting physiological signals. The sensor comprises a first sensor for collecting heart sound signals, a second sensor for collecting pulse signals, and a third sensor for collecting electrocardiogram signals. The first sensor can be a VPU (Voice Pick Up, bone conduction) sensor 250, the second sensor can be an optical pulse sensor 240, and the third sensor can be an electrode sensor. Specifically, the electrode sensor can include three electrodes, a first electrode 210 and a third electrode 230 for forming a loop for electrocardiogram signal collection, and a second electrode 220 for providing a reference point to eliminate the potential difference between the body and the wearable device and improve the signal-to-noise ratio of the electrocardiogram signal collection.

[0068] Further, referring to FIGS. 2-3, the wearable device can further include a device switch 110 for switching the wearable device on and off, a device processor 120 for executing program codes in the memory to perform various functions of the wearable device, a time calibration module 130 for displaying real-time time and data synchronization acquisition calibration, an interaction module 140 for collecting personalized information of a user, signals generated in response to a user performing a blood pressure measurement operation, and simple blood pressure measurement use instructions, the signals generated in response to the user performing the blood pressure operation including a blood pressure measurement start signal, and the use instructions including a blood pressure measurement signal acquisition position and a user acquisition posture, a physiological signal acquisition module 150 for collecting physiological signals related to blood pressure measurement of a user and acceleration and gyroscope signals, the physiological signals related to blood pressure measurement including electrocardiogram signals, pulse signals, and heart sound signals, a data processing module 160 for processing physiological signal segments collected by the physiological signal acquisition module in real time, the data processing steps mainly including signal denoising, signal quality assessment, and feature extraction, a wireless communication module 180 for transmitting physiological signal data, real user blood pressure values, and user personalized information collected by the wearable device to a server or a terminal through a wireless module for construction of a blood pressure measurement database, and a blood pressure measurement module 170 for analyzing physiological signal data and / or personalized information of a user, ultimately performing blood pressure prediction, returning a blood pressure prediction result, and determining a current blood pressure level.

[0069] In addition, referring to FIG. 3, the physiological signal acquisition module further includes a 6-axis signal acquisition module, and acceleration and gyroscope signals are mainly acquired by the 6-axis signal acquisition module, which is mainly composed of a 6-axis sensor 260 integrated inside the device. Before blood pressure measurement is performed, the 6-axis signal acquisition module can calculate current Euler angles to assist a user in positioning an acquisition position and a posture, and during blood pressure measurement, the 6-axis signal acquisition module can monitor movement of a user's arm, in combination with electrocardiogram signals, heart sound signals, and pulse signals, to denoise physiological signals and improve signal-to-noise ratios of the signals during acquisition.

[0070] It can be understood that the structure shown in the embodiments of the present application does not constitute a specific limitation on the wearable device. It can have more or fewer components described above, or combine certain components, or split certain components, or different component arrangements. The various components described above can be implemented in hardware, software, or a combination of hardware and software, including one or more signal processing or application-specific integrated circuits.

[0071] Further, the physiological signals are collected at the same time, and the relationship between the electrocardiogram signal, the pulse signal, the heart sound signal and time is determined to ensure strict synchronization of the physiological signals. The time when the data is recorded can be determined according to the real-time time after calibration.

[0072] In step S20, the physiological signal is subjected to feature extraction to obtain a signal feature, wherein the signal feature includes a pre-ejection period duration and a pulse transit time.

[0073] Further, before the signal feature of the physiological signal is extracted, the physiological signal is subjected to signal processing, such as filtering, noise reduction, segmentation and the like, to improve the accuracy of the extracted signal feature. The signal feature is obtained based on the processed physiological signal.

[0074] In an embodiment, the step of extracting the feature of the physiological signal includes:

[0075] In step S201, the physiological signal is subjected to signal preprocessing to obtain a preprocessed physiological signal, wherein the signal preprocessing includes one or more of filtering, normalization and noise reduction.

[0076] In this embodiment, preferably, the signal preprocessing includes normalization, filtering and noise reduction. Optionally, the collected physiological signal is subjected to normalization, and further subjected to band-pass filtering of the electrocardiogram signal, the pulse signal and the heart sound signal through a pre-designed FIR band-pass filter. The filtered data and the acceleration signal are subjected to secondary processing through adaptive filtering to filter out noise caused by motion artifacts, thereby obtaining clean and high-quality physiological signals.

[0077] In step S202, the signal duration of the preprocessed physiological signal is determined.

[0078] In step S203, if the signal duration is greater than a preset duration, the preprocessed physiological signal is subjected to signal segmentation processing to obtain a plurality of sub-physiological signals.

[0079] It should be noted that the pre-processed physiological signal is subjected to signal segmentation processing to obtain a plurality of sub-physiological signals, which can be subjected to segmentation processing in a time-overlapping manner, for example, the signal duration is 12 seconds, and the signal is segmented in a time length of 3 seconds, and four sub-physiological signals of 0-3 seconds, 3-6 seconds, 6-9 seconds, and 9-12 seconds can be obtained. The physiological signal can also be segmented in a time-overlapping manner, and each signal is segmented into a signal segment with a fixed time length (Sig_t), and the overlap length is any value in the range (0-Sig_t-1). For example, the signal duration is 9 seconds, the signal is segmented in a fixed time length of 3 seconds, and the overlap length is 1 second, and four sub-physiological signals of 0-3 seconds, 2-5 seconds, 4-7 seconds, and 6-9 seconds can be obtained.

[0080] Further, after the pre-processed physiological signal is subjected to signal segmentation processing to obtain a plurality of sub-physiological signals, the signal quality of each sub-physiological signal can be evaluated, the sub-physiological signal with poor signal quality can be deleted, the signal features can be extracted based on the sub-physiological signal with good signal quality, and the effectiveness of the extracted signal features can be ensured.

[0081] As one of the embodiments, the signal quality of each sub-physiological signal can be evaluated by calculating the RR interval of the electrocardiogram signal, the KSQI index, the peak interval of the pulse signal, the peak interval standard deviation, the number of zero-crossing points, the SSQI coefficient of the heart sound signal, the signal mean value, and the like in each signal segment of the sub-physiological signal, and the signal segment with noise interference can be excluded. These features are input into a classification model such as an SVM (Support Vector Machine) model to perform coarse classification on each signal, such as dividing the sub-physiological signal into usable and unusable sub-physiological signals, and extracting signal features based on the usable sub-physiological signal.

[0082] In step S204, the sub-signal features of the sub-physiological signal are extracted.

[0083] In step S205, the mean signal feature of all the sub-signal features is determined, and the mean signal feature is used as the signal feature of the psychological signal.

[0084] If the signal duration of the pre-processed physiological signal is greater than a preset duration, such as 5 seconds, 6 seconds, 7 seconds, or the like, the pre-processed physiological signal can be subjected to signal segmentation processing, and the sub-signal features of each sub-physiological signal are extracted.

[0085] It can be understood that the preprocessed physiological signals are subjected to signal segmentation processing, that is, the preprocessed electrocardiogram signals, the preprocessed pulse signals and the preprocessed heart sound signals are subjected to segmentation processing. The sub-signal features of each segment of the sub-physiological signals are extracted, and each sub-signal feature constitutes an n-dimensional feature sequence X(t), n is the number of features, the mean value of each dimension feature of all X(t) is calculated, and the mean value signal feature, that is, the final signal feature, is obtained. For example, assuming that the physiological signal includes electrocardiogram signals, pulse signals and heart sound signals, three segments of sub-physiological signals are obtained after segmentation processing, which are sub-physiological signal 1, sub-physiological signal 2 and sub-physiological signal 3. The sub-physiological signal 1 includes sub-electrocardiogram signal 1, sub-heart sound signal 1 and sub-pulse signal 1, the sub-physiological signal 2 includes sub-electrocardiogram signal 2, sub-heart sound signal 2 and sub-pulse signal 2, and the sub-physiological signal 3 includes sub-electrocardiogram signal 3, sub-heart sound signal 3 and sub-pulse signal 3. The signal features of the sub-physiological signal 1 are extracted, which are denoted as sub-signal features 1 【1_1, 1_2, 1_3】. The signal features of the sub-physiological signal 1 are extracted, which are denoted as sub-signal features 2 【2_1, 2_2, 2_3】. The signal features of the sub-physiological signal 3 are extracted, which are denoted as sub-signal features 3 【3_1, 3_2, 3_3】. The final signal features obtained are

(1_1+2_1+3_1) / 3, (1_2+2_2+3_2) / 3, (1_3+2_3+3_3) / 3

[0086] In addition, further, after the sub-signal features of each sub-physiological signal are extracted, all the sub-signal features can be subjected to outlier processing, and the abnormal features are deleted. Alternatively, for the signal features of each segment of the sub-physiological signals, the outlier feature sequence processing is performed in accordance with the box plot principle, and an exemplary processing flow can be as follows: S1, the feature values of each segment of the sub-physiological signal features constitute an n-dimensional feature sequence X(t), n is the number of features; S2, the upper edge and the lower edge of each dimension feature are set, for example, the upper quartile Q1i, the lower quartile Q3i, the interquartile range IQRi of each dimension feature are calculated, the upper edge Q1i-1.5IQRi and the lower edge Q3i+1.5IQRi are taken, i=1:n; S3, the abnormal data of each dimension feature outside the upper and lower edges are filtered out; S4, the feature sequence S(t) after removing the outliers is obtained. S5, the mean value of the screened feature sequence is calculated to obtain the mean value signal feature S-.

[0087] The signal features include but are not limited to the pre-ejection period duration and the pulse transit time, such as the pulse arrival time (PAT), the pulse signal and the heart sound signal systolic duration ratio, the pulse signal and the heart sound signal diastolic duration ratio, the time domain feature, the frequency domain feature, the time-frequency feature, the statistical feature and the like.

[0088] The time domain features include, but are not limited to, the following features: R-R interval (RR) of the electrocardiogram signal, R-R standard deviation (SDNN), root mean square difference (RMSSD), etc.; PP interval (PP) of the pulse signal, half-amplitude pulse width (PW50), systolic time, diastolic time, rise time, fastest rise area, peak height, rise slope, etc.; first zero-crossing time of the VPG signal (a signal obtained by first-order differential processing of the pulse signal), last inflection point time, peak value and first zero-crossing slope, peak slope, peak area, etc.; APG signal (a signal obtained by second-order differential processing of the pulse signal) minimum point time, peak point and minimum point slope, first zero-crossing and minimum point slope; first heart sound duration, second heart sound duration, systolic duration, diastolic duration, etc. in the heart sound signal.

[0089] The frequency domain features include, but are not limited to, the following features: power spectral density of the electrocardiogram signal; first component frequency and its amplitude of the pulse signal, second component frequency and its amplitude, third component frequency and its amplitude; S1 main component frequency of the heart sound signal, S2 main component frequency, etc.

[0090] The time-frequency features include, but are not limited to, the following features: wavelet coefficients, Hilbert Huang transform coefficients, mel-frequency cepstral coefficients, linear prediction coefficient features, etc.

[0091] The statistical features include, but are not limited to, the following features: kurtosis factor, skewness factor, standard deviation of the feature sequence, etc.

[0092] In step S30, the signal features are input into the pre-trained blood pressure prediction model, so that the blood pressure prediction model outputs a blood pressure prediction result.

[0093] The trained blood pressure model can be a blood pressure prediction model trained based on a database, the database including at least real blood pressure data and signal feature data corresponding to each real blood pressure data. The database can be obtained by pre-collecting real blood pressure data and signal feature data.

[0094] Further, in order to improve the prediction accuracy of the blood pressure prediction model and realize personalized prediction of blood pressure, real blood pressure values and physiological signals of the detection object can be collected multiple times in advance to establish an individual data set, the individual data set is used as a database for training the prediction model, and the training of the blood pressure prediction model is completed.

[0095] As one of the embodiments, the pre-training process of the blood pressure prediction model can be: S1, selecting features from the multi-category feature library through the existing database; S2, calculating mutual information between each two features, where p(x) is the probability of occurrence of x, p(y) is the probability of occurrence of y, and p(x, y) is the probability of simultaneous occurrence of x and y, i.e., the joint probability. The higher the mutual information, the higher the dependence between the two features. Remove features below the mutual information threshold to obtain a new feature subset; S3, calculating the correlation coefficient between the new feature subset and the blood pressure, the higher the correlation coefficient, the higher the linear correlation between the feature and the blood pressure; S4, sorting the feature subset according to the correlation coefficient from high to low to obtain the sorted feature subset; S5, dividing the sorted feature subset into a training set and a test set, with a ratio of 8:2; S6, for the training set, using ten-fold cross-validation and backward feature selection to select the number of features to obtain the feature subset S with the lowest RMSE (Root Mean Squared Error); S7, using the final feature subset to train a multiple linear regression model: where BP is a specific blood pressure value, S is the optimal feature subset, Ki is the fitting coefficient of the multiple linear regression model, and n is the dimension of the optimal feature subset.

[0096] In addition, the age, gender, height, weight, BMI (Body Mass Index) index, and other personalized features of the tester can also be collected, and the signal feature data is trained to obtain a pre-trained personalized blood pressure prediction model. After extracting the signal features based on the physiological signals, the age, gender, height, weight, BMI (Body Mass Index) index, and other personalized features of the tester are further detected, and the signal features and personalized features are input into the pre-trained personalized blood pressure prediction model to output the blood pressure prediction result. Considering the influence of age, gender, height, weight, BMI (Body Mass Index) index, and other factors on blood pressure, the accuracy of blood pressure prediction can be further improved.

[0097] Further, after obtaining the blood pressure prediction result, the blood pressure prediction result can be output on the wearable device, and the blood pressure prediction result can also be uploaded to a terminal or a server in communication connection with the wearable device for the user to view. The terminal and the server can realize the user's historical blood pressure data viewing, which is helpful for doctors to diagnose. According to the historical data, a personalized blood pressure measurement scheme is customized to help users better understand their own health status and further give more scientific medical advice to users.

[0098] The physiological signals are synchronously collected in the embodiment, wherein the physiological signals include electrocardiogram signals, pulse signals and heart sound signals; feature extraction is performed on the physiological signals to obtain signal features, wherein the signal features include pre-ejection period duration and pulse transit time; and the signal features are input into a pre-trained blood pressure prediction model, so that the blood pressure prediction model outputs a blood pressure prediction result. Compared with the prior art of estimating blood pressure based on pulse transit time, the embodiment predicts the blood pressure of a detector based on fused signal features such as pre-ejection period duration and pulse transit time, considers the influence of pre-ejection period duration on blood pressure, and thus improves the prediction accuracy of blood pressure.

[0099] To facilitate understanding of the technical concept or technical principle of the present application, a specific embodiment is listed:

[0100] Referring to FIGS. 3-4, the blood pressure prediction process in the present embodiment is as follows:

[0101] The wristwatch device is started through a device switch, and the wristwatch device can be connected to a server and a mobile phone APP. After time calibration, the user can input personalized information such as height, weight, etc. The user is prompted to configure the mode, and after the user wears the wristwatch device according to the prompt information, the wristwatch device starts blood pressure measurement, collects physiological signals including electrocardiogram signals, heart sound signals and pulse signals, and also collects acceleration and angular velocity information of the user to calculate the current Euler angle of the user, assisting the user to position the collection position and posture. During blood pressure measurement, the acceleration and angular velocity information of the user can be collected to monitor the movement of the user's arm, and in combination with electrocardiogram signals, heart sound signals and pulse signals, the physiological signals are denoised, the collected physiological signals are processed, such as filtering (which can specifically include band-pass filtering and adaptive filtering), signal quality evaluation, etc. The signal quality evaluation result is used for feature extraction of physiological signals with high signal quality, the extracted signal features are input into a blood pressure prediction model for blood pressure prediction, and the blood pressure prediction result is displayed and uploaded.

[0102] It should be noted that the above specific embodiments are only used to understand the present application and do not constitute a limitation on the blood pressure prediction process and the application device of the present application. Further simple transformations based on the technical concept are within the protection scope of the present application.

[0103] Embodiment Two

[0104] Based on the first embodiment of the present application, in another embodiment of the present application, the same or similar contents as the above embodiment one can be referred to the above introduction, and will not be described in detail hereinafter. On this basis, the step of performing feature extraction on the physiological signals includes:

[0105] Step A10, determining a QRS peak of the electrocardiogram signal, determining a heart sound peak of the phonocardiogram signal, determining a time difference between the two peaks, and taking the time difference as the pre-ejection period duration;

[0106] It should be noted that if the physiological signal is segmented, the feature extraction of the physiological signal in this embodiment can actually be the feature extraction of each sub-physiological signal, and the electrocardiogram signal, the phonocardiogram signal and the pulse signal are also included in the electrocardiogram signal, the phonocardiogram signal and the pulse signal of the sub-physiological signal.

[0107] It can be understood that, as shown in FIG. 6, the physiological signal includes an electrocardiogram (ECG) signal, a phonocardiogram (PCG) signal and a pulse signal, the signal peak of the electrocardiogram signal is determined, the signal peak of the phonocardiogram signal is determined, specifically, the signal peak of the electrocardiogram signal belonging to the same heartbeat cycle is determined, the signal peak of the phonocardiogram signal is determined, the time corresponding to the two signal peaks is obtained, and the time difference between the two is taken as the pre-ejection period duration. If the physiological signal includes multiple heartbeat cycles, the pre-ejection period duration corresponding to each heartbeat cycle can be determined, and the average of all pre-ejection period durations is taken as the final pre-ejection period duration.

[0108] Step A20, obtaining a pulse transit time based on the pulse signal and the electrocardiogram signal.

[0109] Taking the R wave peak of the electrocardiogram signal as the starting point and the pulse wave feature point as the ending point, the time difference in this segment is the pulse transit time, so that one or more pulse wave feature points can be selected to obtain corresponding one or more pulse transit times, that is, the pulse transit time includes one or more, and the user can set the selection rule of the pulse wave feature point according to the actual situation. The number of characteristics of the pulse transit time is not specifically limited in this embodiment.

[0110] Similarly, based on the synchronous acquisition of the electrocardiogram signal and the pulse signal, one or more pulse arrival times (PAT) are determined, and the one or more pulse arrival times can also be taken as the extracted signal characteristics.

[0111] In this embodiment, the signal characteristics such as the pre-ejection period duration and the pulse transit time are extracted, which provides an effective data basis for predicting the blood pressure value.

[0112] In a possible implementation, the signal characteristics further include a systolic duration ratio and a diastolic duration ratio, and the step of extracting the features of the physiological signal further includes:

[0113] Step B10, taking the systolic duration corresponding to the pulse signal as a first systolic duration, and taking the diastolic duration corresponding to the pulse signal as a first diastolic duration;

[0114] The pulse signal has different signal characteristics in the systolic period and the diastolic period. Based on this, the systolic duration and the diastolic duration are extracted from the pulse signal. The specific extraction manner can adopt the prior art, and will not be described herein again.

[0115] Step B20, performing heart sound segmentation on the heart sound signal to obtain a systolic duration corresponding to the heart sound signal and a diastolic duration corresponding to the heart sound signal, taking the systolic duration corresponding to the heart sound signal as a second systolic duration, and taking the diastolic duration corresponding to the heart sound signal as a second diastolic duration;

[0116] The heart sound segmentation of the heart sound signal specifically distinguishes the exact positions of the first heart sound (S1), the systolic period, the second heart sound (S2) and the diastolic period in each cardiac cycle. In each cardiac cycle, the first heart sound (S1) is caused by the blood pressure change and vibration caused by the closure of the mitral valve and the tricuspid valve; the second heart sound (S2) is caused by the closure and vibration of the aortic valve and the pulmonary valve. The systolic period is the interval between S1 and S2, and the diastolic period is from S2 to the beginning of the next cardiac cycle S1. The correctness of the heart sound signal segmentation directly affects the analysis effect of the subsequent heart sound signal. The systolic duration is the time length between S1 and S2, and the diastolic duration is the time length between S2 and the next cardiac cycle S1.

[0117] Step B30, taking the ratio of the first systolic duration and the second systolic duration as a systolic duration ratio;

[0118] Step B40, taking the ratio of the first diastolic duration and the second diastolic duration as a diastolic duration ratio.

[0119] In this embodiment, after the heart sound segmentation of the heart sound signal, the second systolic duration and the second diastolic duration based on the heart sound signal are obtained, and the first systolic duration and the first diastolic duration of the pulse signal are obtained. The ratio between the two is also taken as a signal feature, so as to comprehensively measure the diastolic duration and the systolic duration of the heart sound signal and the pulse signal, and the systolic duration ratio and the diastolic duration ratio can reflect whether the peripheral artery is diseased, and further improve the accuracy of blood pressure prediction.

[0120] In a possible implementation, the step of performing heart sound segmentation on the heart sound signal to obtain a systolic duration corresponding to the heart sound signal and a diastolic duration corresponding to the heart sound signal comprises:

[0121] Step C10, obtaining a first threshold value and a second threshold value corresponding to the heart sound signal;

[0122] Step C20, determining an upper envelope line of the heart sound signal, and performing filtering processing on the upper envelope line to obtain a heart sound envelope signal;

[0123] The filtering processing can be baseline removal and moving average filtering processing.

[0124] Step C30, taking a duration window in which the signal value of the heart sound envelope signal is greater than the first threshold value as a first duration window;

[0125] Step C40, taking a starting time of the first duration window as a first starting time, and taking an ending time of the first duration window as a first ending time;

[0126] Step C50, taking a duration window in which the signal value of the heart sound envelope signal is greater than the second threshold value as a second duration window;

[0127] Step C60, taking a starting time of the second duration window as a second starting time, and taking an ending time of the first duration window as a second ending time;

[0128] Step C70, taking a time difference between the first ending time and the second starting time as a systolic duration corresponding to the heart sound signal;

[0129] Step C80, taking a time difference between the first starting time and the second ending time as a diastolic duration corresponding to the heart sound signal.

[0130] It can be understood that for each heartbeat cycle, there is a corresponding first duration window and second duration window, the first duration window is recorded as S1(t), wherein t is the tth heartbeat cycle (or cardiac cycle), the second duration window is recorded as S2(t), the first start time of S1(t) is recorded as S1start(t), the first end time of S1(t) is recorded as S1end(t), the second start time of S2(t) is recorded as S2start(t), and the second end time of S2(t) is recorded as S2end(t). The systolic duration corresponding to the tth heartbeat cycle is the time difference between S1end(t) and S2start(t), and the diastolic duration is the time difference between S1start(t+1) and S2end(t). Based on this, if the heart sound signal includes heart sound signals of multiple heartbeat cycles, then for each heartbeat cycle, there is a systolic duration and a diastolic duration. The mean duration of the systolic durations corresponding to all heartbeat cycles can be taken as the final second systolic duration, and similarly, the mean duration of the diastolic durations corresponding to all heartbeat cycles can be taken as the final diastolic duration.

[0131] In this embodiment, the heart sound signal is segmented by threshold comparison, which reduces the resource consumption of heart sound segmentation.

[0132] In a possible implementation, before the step of acquiring the first threshold and the second threshold corresponding to the heart sound signal, the method further includes:

[0133] Step D10, determining the QRS wave peak point of the electrocardiogram signal, and taking the time corresponding to the QRS wave peak point as the first time;

[0134] Step D20, determining the wave trough point of the pulse signal, and taking the time corresponding to the wave trough point as the second time;

[0135] Step D30, determining the first zero-crossing point and the second zero-crossing point of the pulse signal, determining the third time corresponding to the first zero-crossing point, and determining the fourth time corresponding to the second zero-crossing point;

[0136] Step D40, determining the first threshold corresponding to the first time and the second time, and determining the second threshold corresponding to the third time and the fourth time.

[0137] It can be understood that, for each heartbeat cycle of the electrocardiosignal and the pulse signal, the first time of the electrocardiosignal peak point in the heartbeat cycle is determined, the peak point can be specifically a QRS wave peak point of the electrocardiosignal, the second time of the trough point of the pulse signal, and the third time and the fourth time of the first zero-crossing point and the second zero-crossing point of the pulse signal, and then the first threshold and the second threshold of the heartbeat cycle are determined, the heart sound segmentation of the heart sound signal of the heartbeat cycle is performed based on the first threshold and the second threshold of the heartbeat cycle, and the second systolic duration and the second diastolic duration of the heart sound signal of the heartbeat cycle are obtained, so that the heart sound segmentation of the heart sound signal is performed based on each heartbeat cycle, and the accuracy of the heart sound segmentation of the heart sound signal is improved.

[0138] In a possible implementation, the step of determining the first threshold corresponding to the first time and the second time comprises:

[0139] Step E10, determining a first peak value of the heart sound envelope signal between the first time and the second time;

[0140] Step E20, adjusting the first peak value based on a preset adjustment coefficient to obtain a first threshold, wherein the first threshold is less than the first peak value.

[0141] The preset adjustment coefficient can be any coefficient set in advance, such as 0.1, 0.15, 0.2, etc. The first peak value is adjusted based on the preset adjustment coefficient, and the first threshold can be obtained by multiplying the first peak value by the preset adjustment coefficient.

[0142] In this embodiment, the first threshold is determined based on the first peak value between the first time and the second time. The first threshold is related to the first peak value, and different peak values correspond to different thresholds, so that intelligent adjustment of the threshold can be realized, and the segmentation accuracy of the heart sound segmentation is improved.

[0143] In a possible implementation, the step of determining the second threshold corresponding to the third time and the fourth time comprises:

[0144] Step F10, determining a second peak value of the heart sound envelope signal between the third time and the fourth time;

[0145] Step F30, adjusting the second peak value based on a preset adjustment coefficient to obtain a second threshold, wherein the second threshold is less than the second peak value.

[0146] The preset adjustment coefficient can be the same or different from the adjustment coefficient described above, and the second peak value is adjusted based on the preset adjustment coefficient, or the second threshold value is obtained by multiplying the second peak value by the preset adjustment coefficient. The second threshold value is related to the second peak value, and different peak values correspond to different threshold values, so that intelligent adjustment of the threshold value can be realized, and the segmentation accuracy of heart sound segmentation is improved.

[0147] To help understand the technical concept or technical principle of the present application, a specific embodiment is listed:

[0148] In the specific embodiment, the heart sound segmentation process is:

[0149] S1, identify the QRS wave peak point of the electrocardiogram signal to obtain a peak point position sequence R(t);

[0150] S2, identify the pulse signal trough point and two zero-crossing points to obtain a trough point position sequence V(t), a first zero-crossing point position sequence Cross_zero1(t), and a second zero-crossing point position sequence Cross_zero2(t);

[0151] S3, calculate the upper envelope line of the heart sound signal, and perform baseline removal and sliding average filtering on the upper envelope line to eliminate baseline drift and noise interference;

[0152] S4, calculate the peak value of the heart sound envelope signal between the R(t) and V(t) sequences, set the threshold value to 10% of the peak value, and take the time difference between the envelope start point sequence S1start(t) and the end point sequence S1end(t) greater than the threshold value as the S1 duration sequence S1(t);

[0153] S5, calculate the peak value of the heart sound envelope signal between the Cross_zero1(t) and Cross_zero2(t) sequences, and set the threshold value to 10% of the peak value. The time difference between the envelope start point sequence S2start(t) and the end point sequence S2end(t) greater than the threshold value is taken as the S2 duration sequence S2(t);

[0154] S6, calculate the systolic duration sequence Sys(t) using the time difference between S1end(t) and S2start(t), and calculate the diastolic duration sequence Dia(t) using S1start(t+1) and S2end(t).

[0155] It should be noted that the above specific embodiment is only used to understand the present application and does not constitute a limitation on the heart sound segmentation process of the present application. More forms of simple transformation based on this technical concept are within the protection scope of the present application.

[0156] Embodiment three

[0157] The embodiment of the present application also provides a blood pressure prediction device, please refer to Figure 7, the server comprises a plurality of core boards, and the device comprises:

[0158] A physiological signal acquisition module 10 is configured to synchronously acquire physiological signals, wherein the physiological signals include electrocardiogram signals, pulse signals and heart sound signals;

[0159] A feature extraction module 20 is configured to extract features of the physiological signals to obtain signal features, wherein the signal features include pre-ejection period duration and pulse transit time;

[0160] A blood pressure prediction module 30 is configured to input the signal features into a pre-trained blood pressure prediction model, so that the blood pressure prediction model outputs a blood pressure prediction result.

[0161] The feature extraction module is further configured to:

[0162] determine a QRS peak value of the electrocardiogram signals, determine a heart sound peak value of the heart sound signals, determine a time difference between the two peak values, and take the time difference as the pre-ejection period duration;

[0163] acquire the pulse transit time based on the pulse signals and the electrocardiogram signals.

[0164] The signal features further include a systolic duration ratio and a diastolic duration ratio, and the feature extraction module is further configured to:

[0165] take a systolic duration corresponding to the pulse signals as a first systolic duration, and take a diastolic duration corresponding to the pulse signals as a first diastolic duration;

[0166] perform heart sound segmentation on the heart sound signals to obtain a systolic duration corresponding to the heart sound signals and a diastolic duration corresponding to the heart sound signals, take the systolic duration corresponding to the heart sound signals as a second systolic duration, and take the diastolic duration corresponding to the heart sound signals as a second diastolic duration;

[0167] take a ratio of the first systolic duration and the second systolic duration as the systolic duration ratio;

[0168] take a ratio of the first diastolic duration and the second diastolic duration as the diastolic duration ratio.

[0169] The feature extraction module is further configured to:

[0170] acquire a first threshold value and a second threshold value corresponding to the heart sound signals;

[0171] determine an upper envelope line of the heart sound signal, filter the upper envelope line to obtain a heart sound envelope signal;

[0172] determine a first duration window in which a signal value of the heart sound envelope signal is greater than the first threshold value, determine a first start time as a start time of the first duration window, and determine a first end time as an end time of the first duration window;

[0173] determine a second duration window in which the signal value of the heart sound envelope signal is greater than the second threshold value, determine a second start time as a start time of the second duration window, and determine a second end time as an end time of the first duration window;

[0174] determine a time difference between the first end time and the second start time as a systolic duration corresponding to the heart sound signal;

[0175] determine a time difference between the first start time and the second end time as a diastolic duration corresponding to the heart sound signal.

[0176] The feature extraction module is further configured to:

[0177] determine a QRS peak point of the electrocardiogram signal, and determine a first time corresponding to the QRS peak point;

[0178] determine a wave trough point of the pulse signal, and determine a second time corresponding to the wave trough point;

[0179] determine a first zero-crossing point and a second zero-crossing point of the pulse signal, determine a third time corresponding to the first zero-crossing point, and determine a fourth time corresponding to the second zero-crossing point;

[0180] determine a first threshold value corresponding to the first time and the second time, and determine a second threshold value corresponding to the third time and the fourth time.

[0181] The feature extraction module is further configured to:

[0182] determine a first peak value of the heart sound envelope signal between the first time and the second time;

[0183] adjust the first peak value based on a preset adjustment coefficient to obtain a first threshold value, where the first threshold value is less than the first peak value.

[0184] The feature extraction module is further configured to:

[0185] determine a second peak value of the heart sound envelope signal between the third time and the fourth time;

[0186] adjust the second peak value based on a preset adjustment coefficient to obtain a second threshold value, wherein the second threshold value is less than the second peak value.

[0187] In addition, the physiological signal acquisition module is further configured to:

[0188] acceleration and gyroscope signals during user measurement.

[0189] The physiological signal acquisition module comprises an electrocardiogram measurement module, a pulse measurement module, a heart sound signal measurement module and a 6-axis signal acquisition module.

[0190] The electrocardiogram signal is mainly acquired by the electrocardiogram measurement module, which mainly comprises a high-impedance chip and three dry electrodes with good conductivity. The pulse signal is mainly acquired by the pulse measurement module, which mainly comprises a multi-wavelength LED (light-emitting diode), a photoelectric sensor and a filter amplification circuit. The final pulse signal is fused by the multi-wavelength LED. In order to ensure the volume of the cuffless blood pressure measurement watch, the high-impedance characteristics of the simulated electrocardiogram signal acquisition, the fusion of the pulse signal, the filter amplification circuit and the synchronization of the two physiological signals are realized by the integrated AFE chip.

[0191] The heart sound signal is mainly acquired by the heart sound signal measurement module, which mainly comprises a digital VPU sensor. The VPU sensor is attached to the inside of the watch shell by a strict rigid connection method. The amplification multiple of the VPU sensor is adjusted to prevent signal overflow during testing, so as to obtain complete heart sound signals.

[0192] The acceleration and gyroscope signals are mainly acquired by the 6-axis signal acquisition module, which mainly comprises a 6-axis sensor integrated inside the watch. Before blood pressure measurement, the 6-axis signal acquisition module can calculate the current Euler angle to assist the user in positioning the acquisition position and posture. During blood pressure measurement, the 6-axis signal acquisition module can monitor the movement of the user's arm, combine with the electrocardiogram signal, heart sound signal and pulse signal, and perform physiological signal denoising to improve the signal-to-noise ratio of the above signals during acquisition.

[0193] In addition, the blood pressure prediction device further comprises:

[0194] A watch shell for placing sensors used for physiological signal acquisition and a watch control system;

[0195] A time calibration module for displaying real-time time and data synchronization acquisition calibration;

[0196] An interaction module is used to collect the user's personalized information, respond to the signals generated by the user during the blood pressure measurement operation, and provide simple blood pressure measurement watch usage instructions. The signals generated during the blood pressure operation include a blood pressure measurement signal. The watch usage prompts include blood pressure measurement watch signal collection positions and user collection postures.

[0197] A data processing module is used to process the physiological signal segments collected by the physiological signal measurement module in real time. The data processing steps mainly include signal noise reduction, signal quality assessment, and feature extraction.

[0198] A wireless communication module is used to transmit the collected physiological signal data, real user blood pressure values, and user personalized information to a server or a mobile terminal through a wireless module for the construction of a blood pressure measurement watch database.

[0199] The watch shell exterior includes a watch switch and three dry electrodes for ECG monitoring. The watch switch mainly controls the switching on and off of the watch and the retrieval of the internal menu of the watch. The first electrode and the third electrode are used to form a loop for ECG signal collection, and the second electrode provides a reference point to eliminate the potential difference between the body and the watch, thereby improving the signal-to-noise ratio of the ECG signal collection. The watch shell interior includes a rigidly connected VPU sensor and a photoplethysmographic pulse sensor for the measurement of heart sound signals and photoplethysmographic pulse signals.

[0200] The time calibration module is used to generate a real-time clock to control the sensors to start collecting physiological signals at the same time, determine the relationship between the ECG signal, pulse signal, and heart sound signal and time, and ensure the strict synchronization of the physiological signals. The recorded data is from the time when the collection is started.

[0201] The interaction module includes an input module, a display module, and a prompt module, and their functions are realized by the internal MCU (Microcontroller Unit) of the watch and the watch screen. The input module 141 in the interaction module is used to input personalized information into the internal watch through the screen of the watch before the blood pressure measurement. The information includes not only personal information such as gender, height, age, and weight, but also information such as whether to take antihypertensive drugs and the name of the antihypertensive drugs.

[0202] The display module in the interaction module is used to display the watch usage method and the measurement steps. After the user completes the personalized information input step, the user can see the way to wear the watch and the posture to maintain during the measurement by clicking the blood pressure measurement function on the screen. After completing the usage tutorial, the test phase is entered, and the watch can prompt the adjustment of the arm position through the prompt module to ensure the quality of physiological signal collection.

[0203] The prompting module in the interaction module comprises a voice chip and a linear vibration motor, and is used for prompting a user operation flow, an arm placement position and a measurement posture, and ensuring blood pressure measurement accuracy.

[0204] The data processing module comprises a signal noise reduction module and a signal quality evaluation module.

[0205] The signal noise reduction module is used for performing normalization operation on the collected physiological signals, and further performing band-pass filtering on electrocardiogram signals, pulse signals and heart sound signals through a pre-designed FIR (Finite Impulse Response) band-pass filter. The filtered data and acceleration signals are further processed through adaptive filtering for a second time, noise generated due to motion artifacts is filtered out, and clean high-quality physiological signals are obtained.

[0206] The signal quality evaluation module is used for dividing each signal into a signal segment with a fixed time length (Sig_t), and the time length is generally > 5s, and the overlapping length is an arbitrary value in the range (0-Sig_t-1). In each signal segment, the RR interval of the electrocardiogram signal, the KSQI index, the peak interval of the pulse signal, the peak interval standard deviation, the number of zero-crossing points, the SSQI coefficient of the heart sound signal, the signal mean value and other characteristics are calculated, and the change time of the acceleration signal amplitude is combined, and each signal is coarsely classified through an SVM model, and the signal segment disturbed by noise is excluded.

[0207] The wireless communication module is used for sending the collected physiological signals, the input personalized information and the actually measured blood pressure value to a terminal and uploading to a cloud server for data collection.

[0208] The blood pressure prediction device provided by the application can solve the technical problem of how to improve the prediction accuracy of blood pressure.

[0209] Embodiment Four

[0210] The electronic device provided by the embodiment of the application comprises at least one processor and a memory in communication connection with the at least one processor, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the blood pressure prediction method in the above-mentioned embodiment one.

[0211] Reference is made below to FIG. 8, which shows a structural schematic diagram of an electronic device suitable for implementing embodiments of the present disclosure. The electronic device in embodiments of the present disclosure can be a wearable device or the like. The electronic device shown in FIG. 8 is merely an example and should not impose any limitation on the functions and use range of embodiments of the present disclosure.

[0212] As shown in FIG. 8, the electronic device can include a processing apparatus 1001 (e.g., a central processor, a graphics processor, or the like) that can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM 1002) or loaded from a storage apparatus into a random access memory (RAM 1004). Various programs and data required for operation of the electronic device are also stored in the RAM 1004. The processing apparatus 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. An input / output (I / O) interface is also connected to the bus 1005.

[0213] Generally, the following systems can be connected to the I / O interface 1006: an input apparatus 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, or the like; an output apparatus 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, or the like; a storage apparatus 1003 including, for example, a magnetic tape, a hard disk, or the like; and a communication apparatus 1009. The communication apparatus 1009 can allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although an electronic device having various systems is shown in the figure, it should be understood that all the shown systems are not required to be implemented or possessed. More or fewer systems can be alternatively implemented or possessed.

[0214] In particular, according to embodiments of the present disclosure, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network through the communication apparatus, or installed from the storage apparatus 1003, or installed from the ROM 1002. When the computer program is executed by the processing apparatus 1001, the above-mentioned functions defined in the methods of embodiments of the present disclosure are performed.

[0215] The electronic device provided by the present application adopts the blood pressure prediction method in the above-mentioned embodiments, and can solve the technical problem of how to improve the prediction accuracy of blood pressure. Compared with the prior art, the electronic device provided by the embodiments of the present application has the same beneficial effects as the blood pressure prediction method provided by the above-mentioned embodiments, and other technical features in the electronic device are the same as the features disclosed in the previous embodiment method, which will not be repeated here.

[0216] It should be understood that portions of the present disclosure can be implemented in hardware, software, firmware, or combinations thereof. In the description of the above-described embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0217] The above merely provides a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0218] Example five

[0219] The embodiment of the present application provides a computer readable storage medium having computer readable program instructions stored thereon, and the computer readable program instructions are used for executing the blood pressure prediction method in the above-described embodiment one.

[0220] The computer readable storage medium provided by the embodiment of the present application may, for example, be a U disk, but is not limited to an electric, magnetic, optical, electromagnetic, infrared, or semiconductor system, system or device, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electric connection with one or more conductive lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the embodiment, the computer readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer readable storage medium can be transmitted by any appropriate medium, including but not limited to: electric wire, optical cable, RF (radio frequency), etc., or any suitable combination of the above.

[0221] The above computer readable storage medium can be contained in an electronic device; or can exist separately without being assembled into an electronic device.

[0222] The above computer readable storage medium carries one or more programs, when the one or more programs are executed by an electronic device, the electronic device synchronously acquires a physiological signal, wherein the physiological signal includes an electrocardiogram signal, a pulse signal and a heart sound signal; extracts features of the physiological signal to obtain signal features, wherein the signal features include a pre-ejection period duration and a pulse transit time; inputs the signal features into a pre-trained blood pressure prediction model, so that the blood pressure prediction model outputs a blood pressure prediction result.

[0223] Computer program code for carrying out operations of the present disclosure can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0224] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0225] The modules involved in the embodiments of the present disclosure can be implemented in the manner of software or hardware. In some cases, the name of the module does not constitute a limitation on the module itself.

[0226] The readable storage medium provided by the present application is a computer readable storage medium, which stores computer readable program instructions for executing the blood pressure prediction method described above, and can solve the technical problem of how to improve the prediction accuracy of blood pressure. Compared with the prior art, the computer readable storage medium provided by the embodiment of the present application has the same beneficial effects as the blood pressure prediction method provided by the first embodiment or the second embodiment, which will not be repeated here.

[0227] Embodiment six

[0228] The embodiment of the present application further provides a computer program product comprising a computer program, which, when executed by a processor, implements the steps of the blood pressure prediction method as described above.

[0229] The computer program product provided by the present application can solve the technical problem of how to improve the prediction accuracy of blood pressure. Compared with the prior art, the beneficial effects of the computer program product provided by the embodiment of the present application are the same as those of the blood pressure prediction method provided by the embodiment one or the embodiment two, which will not be repeated here.

[0230] The above is only the preferred embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation using the content of the present application specification and drawings, or direct or indirect application in other related technical fields, are also included in the patent processing scope of the present application.

Claims

1. A blood pressure prediction method, characterized by, The blood pressure prediction method comprises: synchronously collecting physiological signals, wherein the physiological signals comprise electrocardiogram signals, pulse signals and heart sound signals; extracting features of the physiological signals to obtain signal features, wherein the signal features comprise pre-ejection period duration and pulse transit time; inputting the signal features into a pre-trained blood pressure prediction model to obtain a blood pressure prediction result output by the blood pressure prediction model.

2. The blood pressure prediction method of claim 1, wherein, The step of extracting features of the physiological signals comprises: determining a QRS peak value of the electrocardiogram signals, determining a heart sound peak value of the heart sound signals, determining a time difference between the two peak values, and taking the time difference as the pre-ejection period duration; acquiring the pulse transit time based on the pulse signals and the electrocardiogram signals.

3. The blood pressure prediction method of claim 1, wherein, The signal features further comprise systolic duration ratio and diastolic duration ratio, and the step of extracting features of the physiological signals further comprises: taking a systolic duration corresponding to the pulse signals as a first systolic duration, and taking a diastolic duration corresponding to the pulse signals as a first diastolic duration; segmenting the heart sound signals to obtain a systolic duration and a diastolic duration corresponding to the heart sound signals; taking a systolic duration corresponding to the heart sound signals as a second systolic duration, and taking a diastolic duration corresponding to the heart sound signals as a second diastolic duration; taking a ratio of the first systolic duration and the second systolic duration as the systolic duration ratio; taking a ratio of the first diastolic duration and the second diastolic duration as the diastolic duration ratio.

4. The blood pressure prediction method of claim 3, wherein, The step of segmenting the heart sound signals to obtain a systolic duration and a diastolic duration corresponding to the heart sound signals comprises: acquiring a first threshold value and a second threshold value corresponding to the heart sound signals; determining an upper envelope of the heart sound signals, filtering the upper envelope to obtain a heart sound envelope signal; taking a duration window in which a signal value of the heart sound envelope signal is greater than the first threshold value as a first duration window; taking a starting time of the first duration window as a first starting time, and taking an ending time of the first duration window as a first ending time; taking a duration window in which a signal value of the heart sound envelope signal is greater than the second threshold value as a second duration window; taking a starting time of the second duration window as a second starting time, and taking an ending time of the first duration window as a second ending time; taking a time difference between the first ending time and the second starting time as the systolic duration corresponding to the heart sound signals; taking a time difference between the first starting time and the second ending time as the diastolic duration corresponding to the heart sound signals.

5. The blood pressure prediction method of claim 4, wherein, Before the step of acquiring the first threshold value and the second threshold value corresponding to the heart sound signals, the method further comprises: determining a QRS wave peak point of the electrocardiogram signals, and taking a time corresponding to the QRS wave peak point as a first time; determine a trough point of the pulse signal, and take a second time corresponding to the trough point as the second time; determine a first zero-crossing point and a second zero-crossing point of the pulse signal, determine a third time corresponding to the first zero-crossing point, and determine a fourth time corresponding to the second zero-crossing point; determine a first threshold value corresponding to the first time and the second time, and determine a second threshold value corresponding to the third time and the fourth time.

6. The blood pressure prediction method of claim 5, wherein, The step of determining the first threshold value corresponding to the first time and the second time comprises: determining a first peak value of the heart sound envelope signal between the first time and the second time; adjusting the first peak value based on a preset adjustment coefficient to obtain the first threshold value, wherein the first threshold value is less than the first peak value.

7. The blood pressure prediction method of claim 5, wherein, The step of determining the second threshold value corresponding to the third time and the fourth time comprises: determining a second peak value of the heart sound envelope signal between the third time and the fourth time; adjusting the second peak value based on a preset adjustment coefficient to obtain the second threshold value, wherein the second threshold value is less than the second peak value.

8. A blood pressure prediction device, characterized by, The blood pressure prediction device comprises: a physiological signal acquisition module configured to synchronously acquire physiological signals, wherein the physiological signals comprise electrocardiogram signals, pulse signals, and heart sound signals; a feature extraction module configured to perform feature extraction on the physiological signals to obtain signal features, wherein the signal features comprise pre-ejection period duration and pulse transit time; a blood pressure prediction module configured to input the signal features into a pre-trained blood pressure prediction model, so that the blood pressure prediction model outputs a blood pressure prediction result.

9. An electronic device, comprising: The electronic device comprises: at least one processor; and a memory connected in communication with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the steps of the blood pressure prediction method according to any one of claims 1 to 7.

10. A readable storage medium, characterized by, The readable storage medium is a computer readable storage medium, and the computer readable storage medium stores a program for implementing a blood pressure prediction method. The program for implementing the blood pressure prediction method is executed by a processor to implement the steps of the blood pressure prediction method according to any one of claims 1 to 7.

Images