Non-invasive blood pressure estimation method, apparatus, device, storage medium and product

Abstract

The application provides a non-invasive blood pressure estimation method, device, equipment, storage medium and product, and belongs to the technical field of biomedicine, wherein the method comprises the following steps: preprocessing an ECG signal and an ICG signal; generating physiological characteristic indexes, and the physiological characteristic indexes comprise conventional indexes and special indexes; determining the correlation between the conventional indexes and basic peripheral vascular resistance parameters and basic arterial compliance parameters; inputting the conventional indexes into a pre-trained machine learning model to output basic peripheral vascular resistance prediction parameters and basic arterial compliance prediction parameters; dynamically correcting the special indexes to generate target peripheral vascular resistance parameters and target arterial compliance parameters; and calculating and generating a blood pressure estimation result. The non-invasive blood pressure estimation method, device, equipment, storage medium and product provided by the application can improve the accuracy, dynamic nature and stability of non-invasive blood pressure estimation.

Description

Technical Field

[0001] This invention belongs to the field of biomedical technology, and in particular relates to a non-invasive blood pressure estimation method, device, equipment, storage medium and product. Background Technology

[0002] Blood pressure monitoring is a core tool for assessing cardiovascular health. Regular blood pressure measurements can provide early warning of hypertension and various cardiovascular and cerebrovascular diseases. For patients already diagnosed with cardiovascular disease, blood pressure monitoring can also help assess the effectiveness of treatment plans, helping to stabilize blood pressure within the target range in the long term, thereby significantly reducing the risk of serious complications such as myocardial infarction, stroke, and kidney damage. Therefore, accurate and timely blood pressure monitoring is an indispensable part of clinical diagnosis, health management, and disease prevention.

[0003] In existing technologies, blood pressure monitoring can be broadly divided into invasive and non-invasive methods. Non-invasive blood pressure monitoring, due to its ease of operation and lower risk, has gained widespread application. Typically, non-invasive blood pressure monitoring employs blood pressure estimation methods based on PPG (pulse wave) and ECG (electrocardiogram) signals. However, the accuracy of PPG signal acquisition is easily affected, leading to errors in blood pressure estimation results. Furthermore, blood pressure estimation based on PPG and ECG signals cannot accurately reflect the correlation between blood flow dynamics, neural regulation, and vascular characteristics, thus affecting the dynamics and stability of blood pressure estimation results in complex situations. Summary of the Invention

[0004] In view of this, the present invention aims to provide a non-invasive blood pressure estimation method, device, equipment, storage medium, and product to solve the above-mentioned technical problems.

[0005] To achieve the above objectives, the technical solution created by this invention is implemented as follows:

[0006] In a first aspect, embodiments of the present invention provide a non-invasive blood pressure estimation method, including:

[0007] Acquire ECG and ICG signals, and preprocess them.

[0008] Physiological characteristic indicators are generated based on the preprocessed ECG and ICG signals, and the physiological characteristic indicators include conventional indicators and special indicators.

[0009] Based on the two-element Windkessel model, the correlation between conventional indicators and baseline peripheral vascular resistance parameters and baseline arterial compliance parameters was determined;

[0010] The conventional indicators are input into a pre-trained machine learning model, which is configured to: based on the input conventional indicators, and according to the correlation between the conventional indicators and the baseline peripheral vascular resistance parameters and the baseline arterial compliance parameters, output the baseline peripheral vascular resistance prediction parameters and the baseline arterial compliance prediction parameters.

[0011] Based on specific indicators, the baseline peripheral vascular resistance prediction parameters and baseline arterial compliance prediction parameters are dynamically corrected to generate target peripheral vascular resistance parameters and target arterial compliance parameters.

[0012] The target peripheral vascular resistance parameter and the target arterial compliance parameter are input into the two-element Windkessel model. Based on the conventional indicators, the target peripheral vascular resistance parameter, and the target arterial compliance parameter, the blood pressure estimation result is calculated and generated.

[0013] Furthermore, the preprocessing of the ECG and ICG signals includes:

[0014] Adaptive denoising preprocessing based on wavelet transform is performed on ICG signals;

[0015] The ECG signal is subjected to denoising preprocessing based on a fourth-order bandpass FIR filter.

[0016] Furthermore, the step of generating physiological characteristic indicators based on the preprocessed ECG signal and the preprocessed ICG signal includes:

[0017] An improved Pan-Tompkins algorithm based on Shannon energy envelope estimation is used to detect R-waves in preprocessed ECG signals and extract the signal features of R-waves in ECG signals.

[0018] The preprocessed ICG signal is subjected to time-frequency analysis based on S-transform, and the signal features of points B, C and X in the ICG signal are extracted.

[0019] Identify the signal characteristics of the R wave in ECG signals and the signal characteristics of points B, C, and X in ICG signals to generate physiological characteristic indicators.

[0020] Furthermore, the conventional indicators include: ventricular instantaneous ejection flow, cardiac output, systolic duration, heart rate, initial end-diastolic pressure, and cardiac cycle duration;

[0021] The mathematical expression for the correlation between the aforementioned conventional indicators and baseline peripheral vascular resistance parameters and baseline arterial compliance parameters is as follows:

[0022] ;

[0023] In the above formula, Based on basic peripheral vascular resistance parameters, As a baseline arterial compliance parameter, This refers to the instantaneous ejection blood flow of the ventricle. This is the preset value for instantaneous arterial pressure. A time variable representing a moment within the cardiac cycle;

[0024] The blood pressure estimation result is calculated and generated based on conventional indicators, target peripheral vascular resistance parameters, and target arterial compliance parameters, using the following formula:

[0025] If, during systole, the ventricular ejection flow is assumed to be constant, then:

[0026] ;

[0027] ;

[0028] During diastole, if the ventricular ejection flow is set to 0, then:

[0029] ;

[0030] In the above formula, This refers to ventricular ejection flow during systole. For cardiac output, The duration of the systolic phase. Heart rate, For systolic pressure, For the target peripheral vascular resistance parameter, For target arterial compliance parameters, This is the initial pressure at the end of diastole. For diastolic blood pressure, The duration of the cardiac cycle, It is a time variable that characterizes a moment within the cardiac cycle.

[0031] Furthermore, the specific indicators include: the standard deviation of the cardiac cycle, the root mean square of the difference between adjacent cardiac cycles, the amplitude at point C, the slope from point B to point C, the area enclosed by points B, C, and X, and the time difference from point B to point C.

[0032] The method of dynamically correcting the baseline peripheral vascular resistance prediction parameters and baseline arterial compliance prediction parameters based on specific indicators to generate target peripheral vascular resistance parameters and target arterial compliance parameters is achieved using the following formula:

[0033] ;

[0034] ;

[0035] In the above formula, norm() is the normalization function. For the target peripheral vascular resistance parameter, Based on the parameters for predicting peripheral vascular resistance, It is the root mean square of the difference between adjacent cardiac cycles. The standard deviation of the cardiac cycle. The amplitude at point C, For target arterial compliance parameters, As a predictive parameter for basic arterial compliance, Let S be the slope from point B to point C. Let X be the area enclosed by points B, C, and X. The time difference between point B and point C. , , , , , and All are weighting coefficients, among which Greater than , Greater than ,and and All less than .

[0036] Furthermore, after calculating and generating the blood pressure estimation result, the non-invasive blood pressure estimation method further includes:

[0037] Obtain historical physiological characteristic indicators and historical blood pressure estimation results for multiple consecutive historical cardiac cycles, and calculate the changes in historical physiological characteristic indicators and historical blood pressure estimation results between adjacent historical cardiac cycles;

[0038] Obtain the physiological characteristic indicators of the previous cardiac cycle and calculate the real-time changes in physiological characteristic indicators between the current cardiac cycle and the previous cardiac cycle;

[0039] The real-time physiological characteristic index changes are input into a pre-trained regression model, which is configured to: based on the input real-time physiological characteristic index changes, and according to the mapping relationship between historical physiological characteristic index changes and historical blood pressure estimation result changes, output the predicted blood pressure estimation result changes.

[0040] Obtain the blood pressure estimation result of the previous cardiac cycle, and correct the blood pressure estimation result of the current cardiac cycle based on the blood pressure estimation result of the previous cardiac cycle and the change in the predicted blood pressure estimation result.

[0041] Secondly, embodiments of the present invention also provide a non-invasive blood pressure estimation device, comprising:

[0042] The signal acquisition module is used to acquire ECG and ICG signals and to preprocess the ECG and ICG signals.

[0043] The indicator generation module is used to generate physiological characteristic indicators based on the preprocessed ECG signal and the preprocessed ICG signal, and the physiological characteristic indicators include conventional indicators and special indicators.

[0044] The determination module is used to determine the correlation between conventional indicators and baseline peripheral vascular resistance parameters and baseline arterial compliance parameters based on the two-element Windkessel model;

[0045] The prediction module is used to input conventional indicators into a pre-trained machine learning model, which is configured to output predicted parameters of basic peripheral vascular resistance and basic arterial compliance based on the input conventional indicators and the correlation between the conventional indicators and the baseline peripheral vascular resistance parameters and the baseline arterial compliance parameters.

[0046] The correction module is used to dynamically correct the baseline peripheral vascular resistance prediction parameters and the baseline arterial compliance prediction parameters based on specific indicators, and generate target peripheral vascular resistance parameters and target arterial compliance parameters.

[0047] The output module is used to input the target peripheral vascular resistance parameters and the target arterial compliance parameters into the two-element Windkessel model, and calculate and generate blood pressure estimation results based on conventional indicators, target peripheral vascular resistance parameters, and target arterial compliance parameters.

[0048] Thirdly, embodiments of the present invention also provide an apparatus, comprising:

[0049] One or more processors;

[0050] Storage device for storing one or more programs;

[0051] When the one or more programs are executed by the one or more processors, the one or more processors implement the non-invasive blood pressure estimation method provided in the above embodiments.

[0052] Fourthly, embodiments of the present invention also provide a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform the non-invasive blood pressure estimation method provided in the above embodiments.

[0053] Fifthly, embodiments of the present invention also provide a computer program product, the computer program product storing at least one piece of program code, the at least one piece of program code being executed by a processor to implement the non-invasive blood pressure estimation method provided in the above embodiments.

[0054] Compared with existing technologies, the non-invasive blood pressure estimation method, device, equipment, storage medium, and product described in this invention have the following advantages:

[0055] This invention discloses a non-invasive blood pressure estimation method, device, equipment, storage medium, and product. It can generate physiological characteristic indicators, including conventional and special indicators, from preprocessed ECG and ICG signals. Furthermore, it can use a machine learning model to predict baseline peripheral vascular resistance and arterial compliance parameters based on these conventional indicators. Subsequently, this invention can dynamically correct these parameters according to the special indicators to obtain target peripheral vascular resistance and arterial compliance parameters. Finally, it generates a blood pressure estimation result using these conventional indicators, target peripheral vascular resistance parameters, and target arterial compliance parameters. Compared to existing technologies, the accuracy of blood pressure estimation is improved because the acquisition accuracy of ICG signals is less affected. Additionally, because this invention can dynamically correct the baseline peripheral vascular resistance and arterial compliance parameters, it can fully reflect the correlation between blood flow dynamics, neural regulation, and vascular characteristics, thus ensuring good dynamics and stability of the blood pressure estimation result even under complex conditions. Attached Figure Description

[0056] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments and descriptions of the invention are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0057] Figure 1 A flowchart of the non-invasive blood pressure estimation method described in Embodiment 1 of this invention is provided;

[0058] Figure 2 A schematic diagram of the structure of the non-invasive blood pressure estimation device described in Embodiment 2 of the present invention;

[0059] Figure 3 A schematic diagram of the structure of the device described in Embodiment 3 of the present invention;

[0060] Figure 4 This invention provides a scatter plot of systolic blood pressure results generated during actual testing using the non-invasive blood pressure estimation method described in this invention.

[0061] Figure 5 This invention provides a scatter plot of diastolic blood pressure results generated during actual testing using the non-invasive blood pressure estimation method described in this invention.

[0062] Figure 6A scatter plot of systolic blood pressure error consistency generated during actual testing using the non-invasive blood pressure estimation method described in this invention;

[0063] Figure 7 A scatter plot showing the consistency of diastolic blood pressure error generated during actual testing of the non-invasive blood pressure estimation method described in this invention.

[0064] Figure 8 The present invention creates a systolic blood pressure error histogram during actual testing using the non-invasive blood pressure estimation method described in this invention;

[0065] Figure 9 The diastolic blood pressure error histogram is generated during actual testing using the non-invasive blood pressure estimation method described in this invention. Detailed Implementation

[0066] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present invention, and not all of the structures.

[0067] Example 1

[0068] Figure 1 The flowchart of the non-invasive blood pressure estimation method provided in Embodiment 1 of the present invention is as follows: Figure 1 As shown, the non-invasive blood pressure estimation method specifically includes the following steps:

[0069] Step 110: Acquire ECG and ICG signals, and preprocess them.

[0070] ECG signals reflect the electrical activity of the heart, while ICG signals reflect the changes in thoracic impedance with each heartbeat. In practical applications, both ECG and ICG signals are commonly used medical monitoring signals. In this embodiment, the combination of ECG and ICG signals will be used to achieve non-invasive blood pressure estimation for the subject.

[0071] To improve data accuracy in subsequent processing, after acquiring the ECG and ICG signals, preprocessing should be performed on them to remove noise. In this embodiment, preprocessing the ECG and ICG signals may include the following steps:

[0072] Adaptive denoising preprocessing based on wavelet transform is performed on ICG signals;

[0073] The ECG signal is subjected to denoising preprocessing based on a fourth-order bandpass FIR filter.

[0074] Specifically, in this embodiment, the adaptive denoising preprocessing of the ICG signal based on wavelet transform can use the Mexican Hat wavelet as the basis function. Because the Mexican Hat wavelet has good time-frequency localization characteristics, it can easily capture abrupt changes in signal features such as points B, C, and X in the ICG signal during subsequent processing. Simultaneously, the ICG signal should be decomposed into 5-scale continuous wavelet components, and different processing strategies should be selected for signals at different scales to facilitate the separation of noise from valid signals.

[0075] Scales 1 and 2 primarily contain high-frequency electromyographic noise, which exhibits rapid and irregular fluctuations that significantly interfere with the identification and extraction of subsequent signal features. Therefore, a hard thresholding process is employed: a threshold of 0.8 times the standard deviation of the wavelet coefficients at this scale is used to directly filter out coefficients with amplitudes less than or equal to the threshold, retaining only high-amplitude coefficients that may contain valid signal edge information.

[0076] For scales 3 and 4, since this portion of the signal contains 50Hz power frequency interference and its harmonics, and easily overlaps with the effective signal spectrum in the ICG signal, an improved threshold function is used to process it to balance noise suppression and signal fidelity. The expression for the improved threshold function is as follows:

[0077] ;

[0078] in, The wavelet decomposition of the ICG signal The first scale Wavelet coefficients, For the first threshold after thresholding The first scale Wavelet coefficients, This represents the sign function used to preserve the sign of the original wavelet coefficients. An adaptive threshold is formed by iterating through all coefficients at the current scale and taking the median of the maximum and minimum values.

[0079] Since the improved threshold function can avoid the discontinuity of the hard threshold at the threshold point, it can reduce the amplitude deviation of signal oscillation and soft threshold, thereby ensuring that the amplitude of the processed signal is consistent with that of the original signal.

[0080] For scale 5, since this part of the signal corresponds to the effective signal frequency of the ICG signal (i.e., 30-45Hz), all coefficients are retained to fully preserve the waveform characteristics of the ICG signal.

[0081] After completing the above processing, the residual noise components of scale 1 and scale 2 after threshold processing should be removed, and the processed coefficients of scale 3, scale 4 and scale 5 should be integrated for reconstruction. Finally, effective suppression of baseline drift, 50Hz power frequency interference and high frequency electromyography noise is achieved, and the denoising preprocessing of ICG signal is completed.

[0082] The noise reduction preprocessing of the ECG signal based on a 4th-order bandpass FIR filter described in this embodiment has the following filter specifications:

[0083] Order: 4th order (to reduce computational complexity while maintaining filtering effectiveness);

[0084] Passband range: 5-30Hz (covering the main energy range of the QRS wave to ensure the integrity of the QRS wave waveform);

[0085] Stopband design: less than 0.5Hz (attenuation ≥60dB, suppressing low-frequency interference such as P wave and T wave) and greater than 40Hz (attenuation ≥60dB, suppressing high-frequency electromyographic noise);

[0086] Window function: Hanning window is used, with a transition band width of less than 2Hz, to reduce the impact of passband ripple on signal distortion.

[0087] The above filters can reduce the interference of P-waves and T-waves in the ECG waveform on the QRS wave extraction process and the noise interference in the ICG waveform, thereby improving the clarity of the signal and the accuracy of feature extraction.

[0088] Step 120: Generate physiological characteristic indicators based on the preprocessed ECG signal and the preprocessed ICG signal, and the physiological characteristic indicators include conventional indicators and special indicators.

[0089] After the ECG and ICG signals are preprocessed, this embodiment will generate physiological characteristic indicators, including conventional and special indicators, based on the preprocessed ECG and ICG signals. This will enable the generation of blood pressure estimation results with good accuracy, dynamism, and stability in subsequent processing.

[0090] Optionally, physiological characteristic indicators can be generated based on the preprocessed ECG and ICG signals, which can be specifically optimized as follows:

[0091] An improved Pan-Tompkins algorithm based on Shannon energy envelope estimation is used to detect R-waves in preprocessed ECG signals and extract the signal features of R-waves in ECG signals.

[0092] The preprocessed ICG signal is subjected to time-frequency analysis based on S-transform, and the signal features of points B, C and X in the ICG signal are extracted.

[0093] Identify the signal characteristics of the R wave in ECG signals and the signal characteristics of points B, C, and X in ICG signals to generate physiological characteristic indicators.

[0094] Among them, the Pan-Tompkins algorithm, which is improved based on Shannon energy envelope estimation, is used to detect R-waves in the preprocessed ECG signal. This can improve the robustness of identifying R-waves of different amplitudes in complex noise environments. Specifically, it includes the following steps:

[0095] First, the slope of the preprocessed ECG signal is analyzed based on the classic Pan-Tompkins algorithm, and a first-order forward differential transform is performed. The mathematical expression is as follows:

[0096] ;

[0097] In the above formula, These are the sampled values ​​of the preprocessed ECG signal. This is the result of the difference.

[0098] The above processing can enhance the steep edge characteristics of the QRS group, further suppress the interference of gently changing T-waves and P-waves, and output a bipolar differential signal, providing the raw input for subsequent energy feature extraction.

[0099] Subsequently, the Shannon energy envelope was extracted from the processed signal, and its mathematical expression is as follows:

[0100] ;

[0101] In the above formula, It is an instantaneous energy sequence. It is the minimum value (usually 10). -8 ), used to avoid logarithmic aberrations, To normalize the difference results so that their values ​​fall within a certain range This eliminates amplitude differences (differences caused by individual differences, lead differences, or changes in electrode contact), thereby making the threshold more stable.

[0102] Through the aforementioned nonlinear transformation, the salient features of the QRS group are selectively amplified, forming an instantaneous energy sequence. The signal is then subjected to a zero-phase low-pass filter (cutoff frequency set to 0.5Hz) to obtain a smooth Shannon energy envelope (SEE). The local peaks within this envelope accurately correspond to the QRS group position. Zero-phase filtering prevents the introduction of group delay / phase drag, ensuring the peak position is not postponed and improving the time accuracy of R-peak localization. Then, a dynamic threshold is set using statistical methods, combined with a local peak search strategy. After each candidate peak is found, a physiological refractory period (e.g., 200-250ms) is set to suppress false detections (preventing the repeated counting of the same QRS edges / high-frequency ripples as multiple peaks). The peak is initially located on the envelope. Then, a refined search is performed near this peak on the original / bandpassed ECG signal (e.g., at the point of maximum rise slope or local maximum amplitude) to accurately extract the R-peak position. Finally, the signal characteristics of the R-wave in the ECG signal are extracted based on the R-peak.

[0103] Furthermore, based on the detected R-peak positions, the signal segment between every two adjacent R-peaks can be defined as a complete cardiac cycle, thus completing the cardiac cycle division. Correspondingly, after completing the cardiac cycle division of the ECG signal, the ICG signal can also be divided into cycles based on the same time interval, so that subsequent processing is carried out within the same cardiac cycle.

[0104] In this embodiment, time-frequency analysis of the preprocessed ICG signal based on S-transform can enhance the local waveform variation characteristics of key feature points in the ICG signal. Compared with traditional wavelet transform, which is a time-scale analysis, its scale is not proportional to frequency, and the transformation process is complex and the results are easily affected by noise, S-transform can better adapt to the characteristics of ICG signal, thereby effectively enhancing the local variation characteristics of the waveform corresponding to key feature points. Therefore, when extracting signal features of points B, C, and X in the ICG signal using S-transform, better accuracy can be achieved.

[0105] Specifically, since the key feature points in the ICG signal mainly include point B (the starting point of left ventricular ejection), point C (the peak point of ventricular systole), and point X (the ending point of left ventricular ejection), this embodiment will explain the specific processing method and effect of the S-transform for the above key feature points.

[0106] When processing point B, the S-transform is used to capture the abrupt change characteristics of the ICG signal at the beginning stage, thereby enhancing the local waveform changes at that point and making its starting marker clearer.

[0107] When processing point C, the sensitivity of the S-transform to the extreme value characteristics of the signal is utilized to highlight the peak impedance change in the ICG signal corresponding to the maximum ventricular contraction state, thereby enhancing the energy accumulation characteristics of point C in the time-frequency domain and improving its detection accuracy as a key indicator for assessing cardiac pump function.

[0108] When processing point X, the characteristic of the termination of the ICG signal trend is identified by S-transform to clearly define this point as the marker of the end of ejaculation.

[0109] The above processing can significantly enhance the features of the ICG signal, such as the initial abrupt change at point B, the peak extreme value at point C, and the trend termination at point X, thereby facilitating the extraction of signal features at points B, C, and X.

[0110] After extracting the signal characteristics of the R-wave in the ECG signal and the signal characteristics of points B, C, and X in the ICG signal, this embodiment will generate physiological characteristic indicators based on these signal characteristics. It should be noted that the process of generating physiological characteristic indicators based on the signal characteristics of the R-wave in the ECG signal and the signal characteristics of points B, C, and X in the ICG signal is a mature existing technology. In practical applications, staff can analyze the above signal characteristics in the time domain, frequency domain, amplitude, and waveform morphology to generate corresponding physiological characteristic indicators. Alternatively, a pre-trained neural network model can be used to replace manual analysis, thereby converting signal characteristics into physiological characteristic indicators.

[0111] Step 130: Based on the two-element Windkessel model, determine the correlation between conventional indicators and baseline peripheral vascular resistance parameters and baseline arterial compliance parameters.

[0112] The two-element Windkessel model is a simplified model for simulating hemodynamics in the cardiovascular system. It approximates the arterial system as a second-order circuit composed of resistance and capacitance, where resistance represents the resistance of the vessel wall to blood flow, and capacitance reflects the elastic compliance of the arterial wall. By describing the dynamic relationship between pressure and flow, this model can effectively simulate the propagation and attenuation of pulse waves in the arterial system. Therefore, this embodiment will use the two-element Windkessel model to determine the correlation between conventional indicators and baseline peripheral vascular resistance parameters and baseline arterial compliance parameters, so as to facilitate the prediction of baseline peripheral vascular resistance parameters and baseline arterial compliance parameters in subsequent processing.

[0113] Specifically, to predict baseline peripheral vascular resistance parameters and baseline arterial compliance parameters, the conventional indicators in this embodiment may include ventricular instantaneous ejection flow. Correspondingly, the mathematical expression for the correlation between conventional indicators and baseline peripheral vascular resistance parameters and baseline arterial compliance parameters is as follows:

[0114] ;

[0115] In the above formula, Based on basic peripheral vascular resistance parameters, As a baseline arterial compliance parameter, This refers to the instantaneous ejection blood flow of the ventricle. This is the preset value for instantaneous arterial pressure. It is a time variable that characterizes a moment within the cardiac cycle.

[0116] It should be noted that the preset instantaneous arterial pressure value described in this embodiment is a value preset based on the normal instantaneous arterial pressure of healthy individuals. When the time variable represents the current systolic phase, the preset instantaneous arterial pressure value can be preset between 90 and 139 (corresponding to normal human systolic pressure). When the time variable represents the current diastolic phase, the preset instantaneous arterial pressure value can be preset between 60 and 89 (corresponding to normal human diastolic pressure). Accordingly, when presetting the instantaneous arterial pressure value within the above range, the operator can also adjust the value based on the actual age of the subject, so that the preset instantaneous arterial pressure value is higher for the subject's actual age.

[0117] For example, when determining the instantaneous ventricular ejection flow rate (VRF) among conventional indicators, the ventricular ejection time can be determined based on the distance from point B to point X. The peak value of the rapid ventricular ejection phase can be determined using the amplitude at point C. Then, the stroke volume is calculated using the common Kubicek formula, and a corresponding flow waveform is generated using the ICG signal to finally obtain the instantaneous ventricular ejection flow rate. In the prior art, the specific methods for determining the instantaneous ventricular ejection flow rate based on the signal characteristics of the ICG signal are well known to those skilled in the art and are not part of the core inventive point of this application; therefore, the specific calculation steps are not described in this application.

[0118] Step 140: Input the conventional indicators into the pre-trained machine learning model. The machine learning model is configured to: based on the input conventional indicators, and according to the correlation between the conventional indicators and the baseline peripheral vascular resistance parameters and the baseline arterial compliance parameters, output the baseline peripheral vascular resistance prediction parameters and the baseline arterial compliance prediction parameters.

[0119] In this embodiment, the random forest model, a common machine learning model, can be used to predict the baseline peripheral vascular resistance and baseline arterial compliance parameters. When used, after the conventional indicators are input into the random forest model, the model predicts the baseline peripheral vascular resistance and baseline arterial compliance parameters based on the correlation between the conventional indicators determined in step 130 and the baseline peripheral vascular resistance and baseline arterial compliance parameters. These predicted parameters are then output as results for use in subsequent steps.

[0120] As an optional implementation of this embodiment, to improve the accuracy of the prediction results of the machine learning model, the physiological characteristic indicators may also include other multidimensional indicators. When training the machine learning model using historical data, other historical multidimensional indicators can be input into the machine learning model simultaneously with historical conventional indicators, so that the machine learning model learns the mapping relationship between other historical multidimensional indicators and historical prediction results (i.e., historical baseline peripheral vascular resistance prediction parameters and historical baseline arterial compliance prediction parameters generated based on historical conventional indicators) during the training process. Therefore, in practical applications, inputting conventional indicators and other multidimensional indicators into the machine learning model simultaneously can obtain more accurate baseline peripheral vascular resistance prediction parameters and baseline arterial compliance prediction parameters.

[0121] As examples, and not limitations, other multidimensional indicators include, but are not limited to, the following: time difference from point R to point B, time difference from point R to point C, time difference from point R to point X, time difference from point C to point X, time at the right intersection of the 75% horizontal line, time at the left intersection of the 75% horizontal line, time difference between the left and right intersections of the 75% horizontal line, time at the right intersection of the 50% horizontal line, time at the left intersection of the 50% horizontal line, time difference between the left and right intersections of the 50% horizontal line, time at the right intersection of the 25% horizontal line, time at the left intersection of the 25% horizontal line, time difference between the left and right intersections of the 25% horizontal line, time of occurrence of the maximum derivative, time of occurrence of the minimum derivative, time difference between the maximum and minimum derivatives, amplitude at point B, amplitude at point X, amplitude difference between points C and B, amplitude difference between points C and X, amplitude of the original signal corresponding to the maximum derivative, amplitude of the original signal corresponding to the minimum derivative, and amplitude of the maximum derivative between B and C. Value, minimum derivative amplitude between C and X, ratio of amplitude at point B to amplitude at point C, ratio of amplitude at point B to amplitude at point X, ratio of amplitude at point C to amplitude at point X, ratio of amplitude at point C to amplitude at point B, ratio of amplitude at point X to amplitude at point B, ratio of amplitude at point X to amplitude at point C, ratio of 75% horizontal time difference to cardiac cycle time, ratio of 50% horizontal time difference to cardiac cycle time, ratio of 25% horizontal time difference to cardiac cycle time, time difference from B to C The ratio of the time difference to the cardiac cycle duration, the ratio of the time difference from C to X to the cardiac cycle duration, the ratio of the time difference from B to X to the cardiac cycle duration, the ratio of the systolic to diastolic duration, the slope from C to X, the mean cardiac interval, the number of times the interval difference between adjacent cardiac cycles exceeds 50 milliseconds, the percentage of NN50 in the total number of RR intervals, the median cardiac cycle, the minimum cardiac cycle, the maximum cardiac cycle, and the difference between the maximum and minimum cardiac cycle.

[0122] Step 150: Dynamically correct the baseline peripheral vascular resistance prediction parameters and baseline arterial compliance prediction parameters based on specific indicators to generate target peripheral vascular resistance parameters and target arterial compliance parameters.

[0123] After outputting the basic peripheral vascular resistance prediction parameters and basic arterial compliance prediction parameters through the machine learning model, this embodiment will also dynamically correct them through special indicators to generate target peripheral vascular resistance parameters and target arterial compliance parameters. This will enable the target peripheral vascular resistance parameters and target arterial compliance parameters to accurately reflect the influence of neural regulation on peripheral vascular resistance and the influence of blood flow dynamics on arterial compliance, thereby making the subsequent blood pressure estimation results more accurate.

[0124] Specifically, the special indicators in this embodiment may include: the standard deviation of the cardiac cycle, the root mean square of the difference between adjacent cardiac cycles, the amplitude at point C, the slope from point B to point C, the area enclosed by points B, C, and X, and the time difference from point B to point C.

[0125] Accordingly, the baseline peripheral vascular resistance prediction parameters and baseline arterial compliance prediction parameters are dynamically corrected based on specific indicators to generate target peripheral vascular resistance parameters and target arterial compliance parameters, using the following formulas:

[0126] ;

[0127] ;

[0128] In the above formula, norm() is the normalization function. For the target peripheral vascular resistance parameter, Based on the parameters for predicting peripheral vascular resistance, It is the root mean square of the difference between adjacent cardiac cycles. The standard deviation of the cardiac cycle. The amplitude at point C, For target arterial compliance parameters, As a predictive parameter for basic arterial compliance, Let S be the slope from point B to point C. Let X be the area enclosed by points B, C, and X. The time difference between point B and point C. , , , , , and All are weighting coefficients, among which Greater than , Greater than ,and and All less than .

[0129] Since peripheral vascular resistance is affected by vascular tone, neural regulation, and blood flow intensity, this embodiment will dynamically correct the baseline peripheral vascular resistance prediction parameters by using the root mean square of the difference between adjacent cardiac cycles, the standard deviation of the cardiac cycle, and the amplitude at point C, in order to obtain the target peripheral vascular resistance parameters.

[0130] Specifically, the root mean square of the difference between adjacent cardiac cycles can reflect vagal nerve activity. When the root mean square of the difference between adjacent cardiac cycles is high, it indicates that the vagus nerve is active and can promote vasodilation. At this time, peripheral vascular resistance should decrease.

[0131] The standard deviation of the cardiac cycle reflects the overall heart rate variability. When the standard deviation of the cardiac cycle is high, it indicates that there is unstable neural regulation (such as sympathetic nerve activity), and peripheral vascular resistance should increase.

[0132] The amplitude at point C reflects the intensity of blood flow during ventricular contraction. When the amplitude at point C is high, it indicates that the current blood flow intensity is high, and peripheral vascular resistance should usually decrease.

[0133] In generating the target peripheral vascular resistance parameters, this embodiment also introduces... , and As a weighting coefficient, it reflects the contribution of each specific indicator to the target peripheral vascular resistance parameter. Furthermore, because vagal nerve activity has a greater impact on peripheral vascular resistance than overall heart rate variability, therefore... Should be greater than .

[0134] Since arterial compliance is affected by blood flow dynamics (i.e., ejection velocity and cumulative effect) and neural regulation, this embodiment will dynamically correct the basic arterial compliance prediction parameters by using the slope from point B to point C, the area enclosed by points B, C, and X, the time difference from point B to point C, and the root mean square of the difference between adjacent cardiac cycles, in order to obtain the target arterial compliance parameters.

[0135] Specifically, the slope from point B to point C can reflect the blood flow acceleration rate. When the slope from point B to point C is high, it indicates rapid blood flow changes. At this time, the artery elasticity is better, and therefore the artery compliance strain is greater.

[0136] The area enclosed by points B, C, and X can reflect the cumulative blood flow effect of the blood process. When the area enclosed by points B, C, and X is large, it proves that it has a strong buffering capacity, and therefore the arterial compliance strain is large.

[0137] The time difference between point B and point C can reflect the blood ejection time. When the time difference between point B and point C is small, it indicates that the blood is ejected rapidly and the artery elasticity is good, so the artery compliance strain is large.

[0138] The root mean square of the difference between adjacent cardiac cycles can reflect vagal nerve activity. When the root mean square of the difference between adjacent cardiac cycles is high, it indicates that the vagus nerve is active, which helps to improve vascular elasticity. At this time, arterial compliance should also be greater.

[0139] This embodiment also introduces [a method / mechanism] when generating target arterial compliance parameters. , , and As weighting coefficients, these reflect the contribution of each specific indicator to the target arterial compliance parameter. Ejection time and vagal nerve activity have relatively small effects on arterial compliance, while ejection acceleration has a greater impact on arterial compliance than cumulative blood flow effect. Should be greater than ,and and All should be less than .

[0140] It should be noted that the process of generating the above-mentioned special indicators through the signal characteristics of ECG signals and ICG signals is also existing technology. The specific methods are well known to those skilled in the art and are not part of the core inventive point of this application, so they will not be described in detail in this application.

[0141] Step 160: Input the target peripheral vascular resistance parameter and the target arterial compliance parameter into the two-element Windkessel model, and calculate and generate the blood pressure estimation result based on the conventional indicators, the target peripheral vascular resistance parameter, and the target arterial compliance parameter.

[0142] After obtaining the target peripheral vascular resistance parameters and the target arterial compliance parameters, this embodiment will calculate and generate blood pressure estimation results based on the two-element Windkessel model, conventional indicators, target peripheral vascular resistance parameters, and target arterial compliance parameters, so as to achieve the purpose of non-invasive blood pressure estimation for the subject.

[0143] Specifically, the conventional indicators in this embodiment may also include: cardiac output, systolic duration, heart rate, initial end-diastolic pressure, and cardiac cycle duration.

[0144] Accordingly, based on conventional indicators, target peripheral vascular resistance parameters, and target arterial compliance parameters, blood pressure estimation results can be calculated and generated using the following formula:

[0145] If, during systole, the ventricular ejection flow is assumed to be constant, then:

[0146] ;

[0147] ;

[0148] During diastole, if the ventricular ejection flow is set to 0, then:

[0149] ;

[0150] In the above formula, This refers to ventricular ejection flow during systole. For cardiac output, The duration of the systolic phase. Heart rate, For systolic pressure, For the target peripheral vascular resistance parameter, For target arterial compliance parameters, This is the initial pressure at the end of diastole. For diastolic blood pressure, The duration of the cardiac cycle, It is a time variable that characterizes a moment within the cardiac cycle.

[0151] It should be noted that the process of generating the above-mentioned conventional indicators through the signal characteristics of ECG signals and ICG signals is also existing technology. The specific methods are well known to those skilled in the art and are not part of the core inventive point of this application, so they will not be described in detail here.

[0152] The blood pressure estimate (i.e., diastolic and systolic pressure) of the subject can be obtained by calculating using the above formula within one cardiac cycle. Staff can repeat the above steps in each cardiac cycle to complete continuous blood pressure estimation.

[0153] Optionally, after calculating and generating the blood pressure estimation results, the non-invasive blood pressure estimation method in this embodiment further includes the following steps:

[0154] Obtain historical physiological characteristic indicators and historical blood pressure estimation results for multiple consecutive historical cardiac cycles, and calculate the changes in historical physiological characteristic indicators and historical blood pressure estimation results between adjacent historical cardiac cycles;

[0155] Obtain the physiological characteristic indicators of the previous cardiac cycle and calculate the real-time changes in physiological characteristic indicators between the current cardiac cycle and the previous cardiac cycle;

[0156] The real-time physiological characteristic index changes are input into a pre-trained regression model, which is configured to: based on the input real-time physiological characteristic index changes, and according to the mapping relationship between historical physiological characteristic index changes and historical blood pressure estimation result changes, output the predicted blood pressure estimation result changes.

[0157] Obtain the blood pressure estimation result of the previous cardiac cycle, and correct the blood pressure estimation result of the current cardiac cycle based on the blood pressure estimation result of the previous cardiac cycle and the change in the predicted blood pressure estimation result.

[0158] The above process can reduce the cumulative error and systematic bias of this method in long-term data. By predicting the change in blood pressure estimation results, the blood pressure estimation results of the current cardiac cycle are corrected. Therefore, this method can effectively improve the estimation stability and anti-drift ability of this method in real continuous data. It is especially suitable for error control and result consistency optimization in long-term monitoring tasks.

[0159] This embodiment provides a non-invasive blood pressure estimation method. It achieves noise reduction by preprocessing ECG and ICG signals and generates conventional and special physiological characteristic indicators based on signal features. Subsequently, a machine learning model generates baseline peripheral vascular resistance prediction parameters and baseline arterial compliance prediction parameters based on the conventional indicators, and then corrects these parameters using special indicators to obtain target peripheral vascular resistance parameters and target arterial compliance parameters. This ensures that the target peripheral vascular resistance parameters and target arterial compliance parameters fully reflect the influence of neural regulation and blood flow dynamics. Finally, a blood pressure estimation result is generated using the conventional indicators, target peripheral vascular resistance parameters, and target arterial compliance parameters. Compared with existing technologies, this method improves the accuracy of blood pressure estimation results and maintains good dynamics and stability even under complex conditions.

[0160] Furthermore, the inventors conducted practical application tests based on the non-invasive blood pressure estimation method provided in this embodiment. The tests showed that the cumulative error for systolic blood pressure estimation using this method was less than 5 mmHg in 84.00% of cases, less than 10 mmHg in 96.00% of cases, and less than 15 mmHg in 100.00% of cases. For diastolic blood pressure estimation, the cumulative error was less than 5 mmHg in 98.00% of cases, less than 10 mmHg in 100.00% of cases, and less than 15 mmHg in 100.00% of cases. According to the British Hypertension Society (BHS) standards, both the systolic and diastolic blood pressure estimations using this method were rated Grade A. The experimental results are shown in the table below:

[0161]

[0162] in addition, Figure 4 This is a scatter plot of the systolic pressure results generated during the actual testing process using this method. Figure 5 This is a scatter plot of diastolic blood pressure results generated during actual testing using this method. Figure 6 This is a scatter plot showing the consistency of systolic blood pressure error generated during actual testing using this method. Figure 7This is a scatter plot showing the consistency of diastolic blood pressure error generated during actual testing using this method. Figure 8 This is a histogram of systolic blood pressure error generated during actual testing using this method. Figure 9 This is a histogram of diastolic blood pressure error generated during actual testing using this method. (Through...) Figures 4-9 As shown, in actual blood pressure estimation tests based on this method, the predicted values ​​of systolic and diastolic blood pressure are highly consistent with the actual values. The 95% agreement intervals for systolic and diastolic blood pressure are -0.88±7.15 mmHg and -0.14±3.44 mmHg, respectively, both less than the clinically accepted ±10 mmHg threshold. The error histogram shows that approximately 85% and 98% of the errors for systolic and diastolic blood pressure, respectively, fall within ±5 mmHg, with the diastolic error being particularly concentrated. Therefore, this demonstrates that the method has good robustness on individualized datasets and provides reliable support for the clinical application of non-invasive blood pressure monitoring.

[0163] Example 2

[0164] Figure 2 This is a schematic diagram of the non-invasive blood pressure estimation device provided in Embodiment 2 of the present invention, as shown below. Figure 2 As shown, the non-invasive blood pressure estimation device includes:

[0165] The signal acquisition module 210 is used to acquire ECG signals and ICG signals, and to preprocess the ECG signals and ICG signals.

[0166] The indicator generation module 220 is used to generate physiological characteristic indicators based on the preprocessed ECG signal and the preprocessed ICG signal, and the physiological characteristic indicators include conventional indicators and special indicators.

[0167] Module 230 is used to determine the correlation between conventional indicators and baseline peripheral vascular resistance parameters and baseline arterial compliance parameters based on a two-element Windkessel model.

[0168] The prediction module 240 is used to input conventional indicators into a pre-trained machine learning model, which is configured to output predicted parameters of basic peripheral vascular resistance and basic arterial compliance based on the input conventional indicators and the correlation between the conventional indicators and the baseline peripheral vascular resistance parameters and the baseline arterial compliance parameters.

[0169] The correction module 250 is used to dynamically correct the baseline peripheral vascular resistance prediction parameters and the baseline arterial compliance prediction parameters according to specific indicators, and generate target peripheral vascular resistance parameters and target arterial compliance parameters.

[0170] The output module 260 is used to input the target peripheral vascular resistance parameters and the target arterial compliance parameters into the two-element Windkessel model, and calculate and generate blood pressure estimation results based on conventional indicators, target peripheral vascular resistance parameters, and target arterial compliance parameters.

[0171] This embodiment provides a non-invasive blood pressure estimation device that, through the cooperation of a signal acquisition module and an index generation module, generates physiological characteristic indicators based on ECG and ICG signals. Simultaneously, through the cooperation of a determination module and a prediction module, it predicts and generates baseline peripheral vascular resistance and arterial compliance parameters. Furthermore, a correction module dynamically corrects these parameters, resulting in target peripheral vascular resistance and arterial compliance parameters that reflect neural regulation and blood flow dynamics. Finally, the output module generates a blood pressure estimation result, achieving accurate, stable, and dynamically effective non-invasive blood pressure estimation.

[0172] Based on the above embodiments, the signal acquisition module includes:

[0173] The ICG signal preprocessing unit is used to perform adaptive denoising preprocessing on ICG signals based on wavelet transform.

[0174] The ECG signal preprocessing unit is used to perform noise reduction preprocessing on the ECG signal based on a 4th-order bandpass FIR filter.

[0175] Based on the above embodiments, the indicator generation module includes:

[0176] The ECG signal feature extraction unit is used to perform R-wave detection on the preprocessed ECG signal and extract the signal features of the R-wave in the ECG signal using the improved Pan-Tompkins algorithm based on Shannon energy envelope estimation.

[0177] The ICG signal feature extraction unit is used to perform time-frequency analysis on the preprocessed ICG signal based on S-transform and extract the signal features of points B, C and X in the ICG signal.

[0178] The generation unit is used to identify the signal characteristics of the R wave in the ECG signal and the signal characteristics of the B, C and X points in the ICG signal, and generate physiological characteristic indicators.

[0179] Based on the above embodiments, the non-invasive blood pressure estimation device further includes:

[0180] The correction module is used to acquire historical physiological characteristic indicators and historical blood pressure estimation results from multiple consecutive historical cardiac cycles, calculate the changes in historical physiological characteristic indicators and historical blood pressure estimation results between adjacent historical cardiac cycles; acquire the physiological characteristic indicators of the previous cardiac cycle, calculate the real-time changes in physiological characteristic indicators between the current cardiac cycle and the previous cardiac cycle; input the real-time changes in physiological characteristic indicators into a pre-trained regression model, which is configured to: based on the input real-time changes in physiological characteristic indicators, and according to the mapping relationship between the changes in historical physiological characteristic indicators and the changes in historical blood pressure estimation results, output the predicted changes in blood pressure estimation results; acquire the blood pressure estimation result of the previous cardiac cycle, and correct the blood pressure estimation result of the current cardiac cycle based on the blood pressure estimation result of the previous cardiac cycle and the predicted changes in blood pressure estimation results.

[0181] The non-invasive blood pressure estimation device provided in the embodiments of the present invention can execute the non-invasive blood pressure estimation method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0182] Example 3

[0183] Figure 3 This is a schematic diagram of the device provided in Embodiment 3 of the present invention. Figure 3 A block diagram of an exemplary device 12 suitable for implementing embodiments of the present invention is shown. Figure 3 The device 12 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0184] like Figure 3 As shown, device 12 is represented as a general-purpose computing device. Components of device 12 may include, but are not limited to: one or more processors or processing units 16, system memory 28, and a bus 18 connecting different system components (including system memory 28 and processing unit 16).

[0185] Bus 18 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. For example, these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0186] Device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by device 12, including volatile and non-volatile media, removable and non-removable media.

[0187] System memory 28 may include computer system readable media in the form of volatile memory, such as RAM 30 and / or cache 32. Device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 3 Not shown; usually referred to as a "hard drive"). Although Figure 3 As not shown, disk drives for reading and writing to removable non-volatile disks (e.g., "floppy disks") and optical disc drives for reading and writing to removable non-volatile optical discs (e.g., CD-ROMs, DVD-ROMs, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. System memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0188] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 42 typically perform the functions and / or methods described in the embodiments of the present invention.

[0189] Device 12 can also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), and with one or more devices that enable a user to interact with device 12, and / or with any device that enables device 12 to communicate with one or more other computing devices (e.g., network interface card, modem, etc.). This communication can be performed via I / O interface 22. Furthermore, device 12 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 20. As shown, network adapter 20 communicates with other modules of device 12 via bus 18. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0190] The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, such as implementing the non-invasive blood pressure estimation method provided in the embodiments of the present invention.

[0191] Example 4

[0192] Embodiment 4 of the present invention also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform any of the non-invasive blood pressure estimation methods provided in the above embodiments.

[0193] The computer storage medium of this invention can be any combination of one or more computer-readable media. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0194] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0195] The program code contained on a computer-readable medium may be transmitted using any suitable medium, including—but not limited to—wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0196] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0197] Example 5

[0198] This invention provides a computer program product, which includes a computer program and at least one line of program code. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can perform the non-invasive blood pressure estimation method provided in the above embodiments.

[0199] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. A non-invasive blood pressure estimation device, characterized in that, include: The signal acquisition module is used to acquire ECG and ICG signals and to preprocess the ECG and ICG signals. The indicator generation module is used to generate physiological characteristic indicators based on the preprocessed ECG signal and the preprocessed ICG signal, and the physiological characteristic indicators include conventional indicators and special indicators. The conventional indicators include: ventricular instantaneous ejection flow, cardiac output, systolic duration, heart rate, initial end-diastolic pressure, and cardiac cycle duration; the special indicators include: standard deviation of cardiac cycle, root mean square of the difference between adjacent cardiac cycles, amplitude at point C, slope from point B to point C, area enclosed by points B, C, and X, and time difference from point B to point C. The determination module is used to determine the correlation between conventional indicators and baseline peripheral vascular resistance parameters and baseline arterial compliance parameters based on a two-element Windkessel model; the mathematical expression for the correlation between the conventional indicators and baseline peripheral vascular resistance parameters and baseline arterial compliance parameters is as follows: ； In the above formula, Based on basic peripheral vascular resistance parameters, As a baseline arterial compliance parameter, This refers to the instantaneous ejection blood flow of the ventricle. This is the preset value for instantaneous arterial pressure. A time variable representing a moment within the cardiac cycle; The prediction module is used to input conventional indicators into a pre-trained machine learning model, which is configured to output predicted parameters of basic peripheral vascular resistance and basic arterial compliance based on the input conventional indicators and the correlation between the conventional indicators and the baseline peripheral vascular resistance parameters and the baseline arterial compliance parameters. The correction module is used to dynamically correct the baseline peripheral vascular resistance prediction parameters and baseline arterial compliance prediction parameters based on specific indicators, generating target peripheral vascular resistance parameters and target arterial compliance parameters. The dynamic correction of the baseline peripheral vascular resistance prediction parameters and baseline arterial compliance prediction parameters based on specific indicators to generate target peripheral vascular resistance parameters and target arterial compliance parameters is achieved using the following formula: ； ； In the above formula, For normalization function, For the target peripheral vascular resistance parameter, Based on the parameters for predicting peripheral vascular resistance, It is the root mean square of the difference between adjacent cardiac cycles. The standard deviation of the cardiac cycle. The amplitude at point C, For target arterial compliance parameters, As a predictive parameter for basic arterial compliance, Let S be the slope from point B to point C. Let X be the area enclosed by points B, C, and X. The time difference between point B and point C. , , , , , and All are weighting coefficients, among which Greater than , Greater than ,and and All less than ; The output module is used to input the target peripheral vascular resistance parameters and target arterial compliance parameters into the two-element Windkessel model, and calculate and generate blood pressure estimation results based on conventional indicators, target peripheral vascular resistance parameters, and target arterial compliance parameters. The calculation and generation of blood pressure estimation results based on conventional indicators, target peripheral vascular resistance parameters, and target arterial compliance parameters is achieved using the following formula: If, during systole, the ventricular ejection flow is assumed to be constant, then: ； ； During diastole, if the ventricular ejection flow is set to 0, then: ； In the above formula, This refers to ventricular ejection flow during systole. For cardiac output, The duration of the systolic phase. Heart rate, For systolic pressure, For the target peripheral vascular resistance parameter, For target arterial compliance parameters, This is the initial pressure at the end of diastole. For diastolic blood pressure, The duration of the cardiac cycle, It is a time variable that characterizes a moment within the cardiac cycle.

2. The non-invasive blood pressure estimation device according to claim 1, characterized in that: The preprocessing of ECG and ICG signals includes: Adaptive denoising preprocessing based on wavelet transform is performed on ICG signals; The ECG signal is subjected to denoising preprocessing based on a fourth-order bandpass FIR filter.

3. The non-invasive blood pressure estimation device according to claim 1, characterized in that: The process of generating physiological characteristic indicators based on preprocessed ECG and ICG signals includes: An improved Pan-Tompkins algorithm based on Shannon energy envelope estimation is used to detect R-waves in preprocessed ECG signals and extract the signal features of R-waves in ECG signals. The preprocessed ICG signal is subjected to time-frequency analysis based on S-transform, and the signal features of points B, C and X in the ICG signal are extracted. Identify the signal characteristics of the R wave in ECG signals and the signal characteristics of points B, C, and X in ICG signals to generate physiological characteristic indicators.

4. The non-invasive blood pressure estimation device according to claim 1, characterized in that: The non-invasive blood pressure estimation device also includes: The correction module is used to acquire historical physiological characteristic indicators and historical blood pressure estimation results from multiple consecutive historical cardiac cycles, calculate the changes in historical physiological characteristic indicators and historical blood pressure estimation results between adjacent historical cardiac cycles; acquire the physiological characteristic indicators of the previous cardiac cycle, calculate the real-time changes in physiological characteristic indicators between the current cardiac cycle and the previous cardiac cycle; input the real-time changes in physiological characteristic indicators into a pre-trained regression model, which is configured to: based on the input real-time changes in physiological characteristic indicators, and according to the mapping relationship between the changes in historical physiological characteristic indicators and the changes in historical blood pressure estimation results, output the predicted changes in blood pressure estimation results; acquire the blood pressure estimation result of the previous cardiac cycle, and correct the blood pressure estimation result of the current cardiac cycle based on the blood pressure estimation result of the previous cardiac cycle and the predicted changes in blood pressure estimation results.

Images