Pulse wave signal processing methods, devices and systems

Abstract

This invention provides a pulse wave signal processing method and apparatus. The method includes: acquiring and storing raw pulse wave signals; truncating the raw pulse wave signals using a sliding time window function and constructing a fitting function; adjusting the parameters of the sliding time window function and / or the fitting function to make the fitted signal corresponding to each truncated signal segment approximate the signal segment; generating a pulse wave signal set based on the signal segments or the corresponding fitted signals; splicing the signal segments in the pulse wave signal set into signal segments with certain signal characteristics to obtain a target periodic signal set; detecting the signal segments in the target periodic signal set and extracting heart rhythm signal features and signal features between heart rhythm beats as secondary signal features; and determining whether to output a normal pulse wave signal or identify motion artifacts or noise based on the secondary signal features. Using this invention, high-quality pulse wave signals can be extracted under complex interference and large motion conditions.

Description

Technical Field

[0001] This invention relates to the field of biomedical signal processing technology, specifically to a pulse wave signal processing method, device, and system. Background Technology

[0002] A pulse wave is a signal generated by the beating of the human heart, propagating along arteries and blood flow to the periphery of the body. Pulse waves can be acquired using signal sensors such as pressure sensors, photoelectric sensors, and ultrasound sensors. Pressure sensors can be placed directly inside arteries or on the skin surface near arteries to measure pressure changes in the pulse wave and convert them into electrical signals. Photoelectric sensors detect differences in the intensity of transmitted or reflected light absorbed by human blood and tissues, recording changes in blood vessel volume during the cardiac cycle; these changes can be converted into electrical signals synchronized with the pulse wave. Ultrasound sensors are placed on the skin surface near arteries, sending ultrasound signals through the skin to the arteries and surrounding tissues, generating reflected echoes. The sensor detects these echo changes and converts them into electrical signals synchronized with the pulse wave.

[0003] Regardless of the type of sensor used, the acquired pulse wave signal is easily affected by various factors such as the subject's own activities, the movement or positional changes of the acquisition device, and changes in the surrounding environment, resulting in interference signals with various modes and characteristics. These interference signals are superimposed on the desired biological signal, causing baseline drift, waveform distortion, signal distortion, etc., resulting in a decrease in the quality of the biological signal and negatively impacting subsequent signal processing analysis and extraction of physiological information components.

[0004] In addition, because the contribution of the motion component in the pulse wave signal often exceeds that of the anatomy component by a certain amount during human movement, motion artifacts are generated in the pulse wave signal. Motion artifacts can lead to incorrect interpretations and degrade the accuracy and reliability of cardiovascular parameter estimations.

[0005] Therefore, how to eliminate interference and motion artifacts in pulse wave signals and improve the quality of pulse wave signals is an important problem that the industry needs to solve. Summary of the Invention

[0006] This invention provides a pulse wave signal processing method, apparatus, and system to effectively remove interference noise and motion artifacts from pulse wave signals, and to extract high-quality pulse wave signals under complex interference and large motion conditions.

[0007] Therefore, the present invention provides the following technical solution:

[0008] This invention provides a pulse wave signal processing method, the method comprising:

[0009] Acquire and store the raw pulse wave signal;

[0010] The original pulse wave signal is truncated using a sliding time window function, and a fitting function is constructed. The parameters of the sliding time window function and / or the fitting function are adjusted so that the fitting signal corresponding to each truncated signal segment approximates the original signal segment, thereby obtaining a fitting signal segment corresponding to the original signal segment. A pulse wave signal set is generated based on the original signal segment or the corresponding fitting signal segment.

[0011] The signal segments in the pulse wave signal set are spliced ​​together to form signal segments with certain signal characteristics to obtain the target periodic signal set.

[0012] The signal segments in the target periodic signal set are detected, and the heart rhythm signal features and the signal features between heart rhythm beats are extracted as secondary signal features.

[0013] Based on the secondary signal characteristics of the signal segments in the target periodic signal set, it is determined whether these signal segments meet the decision conditions; if the decision conditions are met, these signal segments are spliced ​​together and output as a normal pulse wave signal; if the decision conditions are not met, these signal segments are regarded as motion artifacts or noise.

[0014] Optionally, generating a pulse wave signal set based on the signal segment includes: recording the signal segment into the pulse wave signal set.

[0015] Optionally, generating a pulse wave signal set based on the signal segment includes:

[0016] Generate fitted signals corresponding to each signal segment;

[0017] The fitted signal is recorded in the pulse wave signal set.

[0018] Optionally, the fitting function is a monotonic function or a periodic function.

[0019] Optionally, the step of constructing the fitting function and adjusting the parameters of the sliding time window function and / or the fitting function to make the fitted signal corresponding to each intercepted signal segment approximate the signal segment includes:

[0020] Construct a fitting function, and use the fitting function to generate a fitting signal corresponding to the signal segment;

[0021] Calculate the signal characteristics of the fitted signal and the corresponding signal segment;

[0022] If the fitted signal and the signal characteristics of the corresponding signal segment meet the set conditions, then the fitted signal is taken as the fitted signal segment corresponding to the signal segment;

[0023] Otherwise, adjust the time window function parameters and / or fitting function parameters until the fitted signal and the signal characteristics of the corresponding signal segment meet the set conditions.

[0024] Optionally, splicing signal segments from the pulse wave signal set into signal segments with certain signal characteristics includes: sequentially splicing multiple signal segments from the pulse wave signal set that are temporally adjacent and have signal characteristics that meet set conditions into signal segments with certain signal characteristics.

[0025] Optionally, the step of detecting whether a signal segment satisfies the decision conditions based on the secondary signal features of the signal segments in the target periodic signal set includes: judging whether the self-signal features of two adjacent signal segments in the target periodic signal set, the similarity and correlation between the two signal segments, the variability between the signal segments, and the continuity of the signal segments satisfy the decision conditions based on the secondary signal features.

[0026] Alternatively, one of the following methods can be used to determine whether these signal segments meet the decision conditions: voting method, stacking method, ensemble learning method, or Bayesian fusion method.

[0027] Optionally, the signal characteristics include any one or more of the following: obvious monotonicity or periodicity, strong similarity or correlation during the period, relatively slow variability during the period, and a certain degree of continuity.

[0028] Alternatively, the monotonicity or periodicity of the signal can be calculated using any of the following methods: time-domain analysis, frequency-domain analysis, or statistical methods.

[0029] Alternatively, the similarity or correlation between signals can be calculated using any of the following methods: distance metric method, statistical feature comparison method, correlation analysis method, time-frequency analysis method, machine learning or deep learning method.

[0030] Optionally, the variability between signals can be calculated using any of the following methods: time-domain analysis, autocorrelation or cross-correlation analysis, frequency-domain analysis, or statistical methods.

[0031] Optionally, the method further includes: performing noise reduction and filtering on the original pulse wave signal before truncating it.

[0032] Optionally, the filtering process of the original pulse wave signal includes: using any of the following filtering methods to perform noise reduction and filtering on the original pulse wave signal: low-pass filtering, high-pass filtering, band-pass filtering, median filtering, sliding window filtering, wavelet transform, Kalman filtering, empirical mode decomposition (EMD), variational mode decomposition (VMD), and local mean decomposition (LMD).

[0033] The present invention also provides a pulse wave signal processing device, the device comprising:

[0034] The signal acquisition module is used to acquire and store the raw pulse wave signal;

[0035] The segmentation module is used to truncate the original pulse wave signal using a sliding time window function, construct a fitting function, adjust the parameters of the sliding time window function and / or the fitting function parameters, so that the fitting signal corresponding to each truncated signal segment approximates the signal segment; and generate a pulse wave signal set based on the signal segment or the corresponding fitting signal.

[0036] The splicing module is used to splice signal segments in the pulse wave signal set into signal segments with certain signal characteristics to obtain a target periodic signal set.

[0037] The signal detection module is used to detect signal segments in the target periodic signal set, extract secondary signal features between signal segments and between signal segments; detect whether the signal segments meet the decision conditions based on the secondary signal features of the signal segments in the target periodic signal set; if the decision conditions are met, these signal segments are spliced ​​together and output as a normal pulse wave signal; if the decision conditions are not met, these signal segments are regarded as motion artifacts or noise.

[0038] Optionally, the device further includes a preprocessing module for performing noise reduction and filtering on the original pulse wave signal before intercepting it.

[0039] The present invention also provides a pulse wave signal processing system, the system comprising: a signal acquisition device and the pulse wave signal processing device;

[0040] The acquisition device is used to acquire human pulse wave signals and transmit the acquired pulse wave signals for a certain period of time to the pulse wave signal processing device.

[0041] The pulse wave signal processing device is used to process the human pulse wave signal and output a normal pulse wave signal.

[0042] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, performs the steps of the pulse wave signal processing method.

[0043] The present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the steps of the pulse wave signal processing method when running the computer program.

[0044] The pulse wave signal processing method, apparatus, and system provided by this invention perform noise reduction and filtering on the original pulse wave signal, then segment it using a sliding time window function. For each segmented signal segment, a manually generated fitting signal is used for fitting. The segmentation parameters and fitting signal parameters are adjusted to ensure that the error between the fitted signal and the signal segment is less than a preset value. The signal segments or their corresponding fitted signal segments are then used to generate a pulse wave signal set. The signal segments within the pulse wave signal set are then spliced ​​together to form signal segments with certain signal characteristics. Secondary features of the signal segments are then extracted. Based on the pulse wave signal characteristics, the signal characteristics between adjacent signals, and the continuity of the signal segments, a decision rule fusion method is used to splice these signal segments that conform to the decision rules into a complete signal. This adaptively filters out signal interference and motion artifacts in the pulse wave signal, achieving high-quality pulse wave signal extraction under complex interference and large motion conditions. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly described below. Obviously, the drawings described below are merely some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0046] Figure 1 This is a flowchart of a pulse wave signal processing method provided by the present invention;

[0047] Figure 2 This is a flowchart illustrating the generation of a fitted signal corresponding to a signal segment in an embodiment of the present invention;

[0048] Figure 3 This is a schematic diagram of a pulse wave signal processing device provided by the present invention;

[0049] Figure 4 This is a schematic diagram of a pulse wave signal processing structure provided by the present invention. Detailed Implementation

[0050] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0051] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0052] Analyzing and calculating blood pressure, blood oxygen saturation, and other physiological characteristics using pulse waves typically requires high-quality pulse wave signals as basic data. High-fidelity signals can more accurately represent pulse wave details and reflect dynamic changes in the cardiovascular system; continuous signals facilitate data analysis and result output during long-term monitoring, ensuring data continuity and comparability; stable signals help improve the signal-to-noise ratio and reduce measurement errors caused by interference or motion artifacts. Therefore, removing signal interference and motion artifacts from the pulse wave signals acquired by sensors can directly improve the accuracy and reliability of blood pressure, blood oxygen saturation, and other physiological characteristic analyses.

[0053] To address this, the present invention provides a pulse wave signal processing method and apparatus. The method involves denoising and filtering the original pulse wave signal, then segmenting it using a sliding time window function. The segmented signal segments are first spliced ​​together to form signal segments with certain signal characteristics. Then, based on the pulse wave signal characteristics, the correlation between adjacent signals, and the continuity of the signal segments, these signal segments are spliced ​​together into a complete signal.

[0054] like Figure 1 The diagram shown is a flowchart of a pulse wave signal processing method provided by the present invention, which includes the following steps:

[0055] In step 101, the raw pulse wave signal is acquired and stored.

[0056] The raw pulse wave signal can be the raw pulse wave signal collected by the sensor.

[0057] In a non-limiting embodiment, the original pulse wave signal can also be denoised and filtered to remove baseline drift and obtain a preprocessed signal.

[0058] The purpose of filtering the original pulse wave signal is to remove baseline drift. Specifically, any of the following filtering methods can be used to denoise and filter the original pulse wave signal: low-pass filtering, high-pass filtering, band-pass filtering, median filtering, sliding window filtering, wavelet transform, Kalman filtering, Empirical Mode Decomposition (EMD) and its extensions, Variational Mode Decomposition (VMD) and its extensions, and Local Mean Decomposition (LMD) and its extensions.

[0059] For example, the raw analog signal of the pulse wave is acquired from the sensor at a sampling frequency of 200Hz. A 32-bit analog-to-digital converter converts the acquired analog signal into a raw decimal digital signal. This signal may contain interference from various sources, such as DC components, baseline drift, motion artifacts, or electromagnetic interference.

[0060] A Butterworth digital bandpass filter was used, with the passband frequency set to 0.5Hz–20Hz, passband ripple better than 3dB, stopband frequency less than 0.25Hz and greater than 30Hz, and stopband attenuation better than 30dB. The original digital signal was filtered to obtain a preprocessed signal. Compared to the original signal, the preprocessed signal had unwanted DC components, baseline drift, low-frequency and high-frequency components removed, while retaining the main pulse wave information.

[0061] In step 102, the original pulse wave signal is truncated using a sliding time window function, and a fitting function is constructed. The parameters of the sliding time window function and / or the fitting function are adjusted so that the fitting signal corresponding to each truncated signal segment approximates the signal segment. A pulse wave signal set is generated based on the signal segment or the corresponding fitting signal.

[0062] When truncating the preprocessed signal, a time window function of length w can be preset. Based on this time window function, a signal of length w is truncated from the preprocessed signal to obtain multiple signal segments.

[0063] In some embodiments, the final extracted signal segments can be recorded into the pulse wave signal set, and motion artifacts can be adaptively eliminated by splicing these signal segments to obtain a high-quality pulse wave signal.

[0064] In other embodiments, the fitted signals corresponding to each of the final extracted signal segments can be recorded in the pulse wave signal set, and the motion artifacts can be adaptively eliminated by splicing these fitted signal segments to obtain high-quality pulse wave signals.

[0065] The constructed fitting function can be a monotonic function or a periodic function, and this embodiment of the invention does not limit the specific function.

[0066] For each signal segment, a fitted signal of length w is generated using the fitting function. By adjusting the time window function parameters and / or the fitting function parameters, the fitted signal can be made to approximate the corresponding signal segment.

[0067] like Figure 2 The diagram shown is a flowchart of generating a fitted signal corresponding to a signal segment in an embodiment of the present invention, including the following steps:

[0068] Step 201: Construct the fitting function.

[0069] Step 202: Use the fitting function to generate a fitted signal corresponding to the truncated signal segment.

[0070] Step 203: Calculate the signal characteristics of the fitted signal and the corresponding signal segment.

[0071] The signal characteristics may include, but are not limited to, any one or more of the following: obvious monotonicity or periodicity (such as the time repetition of typical feature points or segments, periodic time-frequency domain characteristics, etc.), strong similarity or correlation between periods (such as small Euclidean distance between periods, large correlation coefficient, etc.), and slow signal characteristic changes during periods (such as the amplitude change rate, rise slope change rate, peak-to-peak change rate, etc., which are consistent with the possible amplitude change rate of a normal pulse wave).

[0072] The periodicity characteristics of the signal can be calculated using, but is not limited to, any of the following methods:

[0073] (1) Time-domain analysis method

[0074] For example, calculate whether signal segments with time-domain waveforms that have the same peak, valley, and slope are repeated; calculate the autocorrelation function of the signal and observe whether it has a significant peak under certain specific delays.

[0075] (2) Frequency Domain Analysis Method

[0076] For example, when a signal is subjected to a Fourier transform, by performing a Discrete Fourier Transform (DFT) calculation on the signal, a periodic signal will usually exhibit discrete frequency components in the spectrum.

[0077] (3) Statistical methods

[0078] For example, statistical indicators of signal periodicity, such as periodicity intensity (PPI) or periodicity factor (PF); analyzing the energy distribution of a signal at different frequencies. Periodic signals typically have a high energy concentration at specific frequencies.

[0079] The similarity or correlation of the signals can be calculated using, but is not limited to, any of the following methods:

[0080] (1) Distance metric methods

[0081] For example, the Euclidean distance, Manhattan distance, etc., between the extracted signal segment and the corresponding fitted signal can be calculated to assess the similarity or correlation between the two signals.

[0082] (2) Statistical characteristic comparison method

[0083] For example, the statistical characteristics of the extracted signal segments and the fitted signal, such as mean, variance, period length, time length, slope, amplitude, kurtosis, skewness, etc., can be compared to assess the similarity or correlation between the two signals.

[0084] (3) Similarity or correlation analysis methods

[0085] For example, the Pearson correlation coefficient and cross-correlation function of the extracted signal segment and the fitted signal can be calculated to assess the similarity or correlation between the two signals.

[0086] (4) Time-frequency analysis method

[0087] For example, by using Fourier transform and wavelet transform, the truncated signal segment and the fitted signal can be decomposed into wavelet components of different frequencies, and then the energy or features of these wavelet components can be compared to evaluate the similarity or correlation between the two signals; or, the frequency domain features and time domain features of the target signal and the fitted signal can be calculated and compared to evaluate the similarity or correlation between the two signals.

[0088] (5) Machine learning methods.

[0089] For example, machine learning techniques such as Support Vector Machine (SVM) and neural networks can be used for pattern recognition and classification to evaluate the similarity or correlation between the extracted signal segments and the fitted signal.

[0090] Step 204: Determine whether the signal characteristics of the fitted signal and the corresponding signal segment meet the set conditions; if yes, proceed to step 205; otherwise, proceed to step 206.

[0091] Step 205: Record the fitted signal into the pulse wave signal set.

[0092] Step 206: Adjust the time window function parameters and / or fitting function parameters, then return to step 202.

[0093] It should be noted that if the time window function is adjusted, the signal segment corresponding to the fitted signal will also change, that is, the start and end points of the signal segment will be redefined.

[0094] For each captured signal segment, follow the above... Figure 2 The process shown generates the corresponding fitted signal and the corresponding target signal.

[0095] Each target signal is recorded as a final extracted signal segment in the pulse wave signal set, or each fitted signal is recorded in the pulse wave signal set.

[0096] The following example, using a distance metric method, further illustrates the process of generating a set of pulse wave signals in step 102.

[0097] A time window function of length w is preset. The initial value of w can be set to, for example, 0.25s. The value of w can be optimized and determined based on factors such as the sampling frequency of the pulse wave signal, the signal-to-noise ratio of the pulse wave signal, and the heart rhythm cycle characteristics contained in the pulse wave signal. The time window function is optimized based on experience.

[0098] Parameters can improve computational efficiency and the resolution of fragment decomposition.

[0099] A maximum allowable fitting error ε is preset. The initial value of ε can be set to 8, for example. Optimizing ε based on experience can improve computational efficiency and the resolution of fragment decomposition.

[0100] Starting from point A0 of the preprocessed signal, a signal of length w is extracted and the endpoint is B0. The extracted signal segment S0 is taken as the target signal.

[0101] Based on the characteristics of the target signal S0, a quadratic polynomial is constructed as the fitting function. The formula for the quadratic polynomial function is as follows:

[0102] f(x) = ax 2 +bx+c (1)

[0103] Based on the fitting function, a fitting signal C0 of length w is generated. By adjusting the parameters a, b, and c of the fitting function, the fitting signal C0 is made to approximate the target signal S0 as closely as possible.

[0104] The formula for calculating the distance d between two points is as follows:

[0105]

[0106] Select a point A on the target signal S0 i Calculate point Ai The minimum distance d from each point on the fitted signal C0 is denoted as A. i The distance d from the point to the fitted signal C0.

[0107] Adjust the time window function parameters and the fitting signal parameters a, b, and c. When the distance d from all points on the target signal S0 to the fitting signal C0 is not greater than the maximum allowable fitting error ε, record the starting point A0 and ending point B0 of the current target signal S0 and add the target signal S0 to the pulse wave signal set; or record the starting point A0′ of the current fitting signal C0 and add the fitting signal C0 to the pulse wave signal set.

[0108] Repeat the above steps until all the preprocessed signals have been processed. The preprocessed signals are divided into several target signal segments, and fitting signals corresponding to each target signal segment are generated.

[0109] Continue to refer to Figure 1 In step 103, the signal segments in the pulse wave signal set are spliced ​​together to form signal segments with certain signal characteristics, thereby obtaining the target periodic signal set.

[0110] Specifically, multiple signal segments that are temporally adjacent and have similar or related pulse wave signal characteristics in the pulse wave signal set are sequentially spliced ​​together to form a signal segment with certain signal characteristics.

[0111] For example, n1 temporally adjacent signal segments from a pulse wave signal set are spliced ​​together to form a new signal segment. This new signal segment has similar or related pulse wave signal characteristics and represents m1 pulse wave cycles. These spliced ​​signal segments representing m1 pulse wave cycles are then added to the target periodic signal set. This process is repeated until all signal segments in the pulse wave signal set have been processed, and these processed signal segments are then added to the target periodic signal set.

[0112] The signal characteristics mainly refer to the signal features of the original or fitted pulse wave signal within a certain window function. These signal characteristics may include, but are not limited to: 1. obvious monotonicity or periodicity; 2. similarity or correlation between adjacent signal segments; 3. slow variation between adjacent signal segments; 4. a certain degree of repeatability.

[0113] Signal segments in a pulse wave signal set can be spliced ​​using, but are not limited to, any of the following methods.

[0114] (1) Slope feature method

[0115] Calculate the slope value of each signal segment in the pulse wave signal set; splice together n1 adjacent signal segments with the same slope to merge them into a new signal segment, which represents the rising or falling segment of m1 complete pulse wave cycles; record this new signal segment into the target cycle signal set.

[0116] For example, calculate the slope values ​​of two adjacent signal segments in a pulse wave signal set, multiply these two slope values, and if the product of the slope values ​​of the current signal segment and the next adjacent signal segment is greater than or equal to 0, merge these two signal segments into the current signal segment.

[0117] Repeat the above process, calculating the slope values ​​of the current signal segment and the next signal segment, multiplying these two slope values ​​until the product of the slopes of two adjacent segments is less than 0. At this point, the merged current signal segment is added to the target periodic signal set.

[0118] Repeat the above process until all signal segments in the pulse wave signal set have been processed and recorded in the target periodic signal set. All signal segments in the target periodic signal set have similar signal characteristics.

[0119] (2) Peak Periodicity Characteristic Method

[0120] Calculate the peak-to-peak value (the amplitude difference between the maximum positive peak and the maximum negative peak value of the signal waveform within one pulse wave cycle, i.e., the total amplitude difference between the peak and trough values ​​of the waveform) and the distance between peak-to-peak values ​​(the distance difference between the maximum positive peak value and the maximum negative peak value of the signal waveform within one pulse wave cycle) of each signal segment in the pulse wave signal set. Then, concatenate n1 adjacent signal segments with similar peak-to-peak values ​​to form a new signal segment, which represents several complete pulse wave cycles; or concatenate n1 adjacent signal segments with similar peak-to-peak distances to form a new signal segment, which represents m1 complete pulse wave cycles. Finally, record these new signal segments in the target period signal set.

[0121] Continue to refer to Figure 1 In step 104, signal segments in the target periodic signal set are detected, and cardiac rhythm signal features and signal features between cardiac rhythm beats are extracted as secondary signal features.

[0122] The secondary signal features mainly refer to the signal features of signal segments or combinations of signal segments within the target periodic signal set. The heart rhythm refers to the rhythm and regularity of the heartbeat, which can be a single heartbeat cycle or multiple heartbeat cycles. When extracting the secondary signal features, it is necessary to first identify each heartbeat beat of the first spliced ​​signal segment before extracting the secondary signal features.

[0123] It should be noted that the secondary signal characteristics are similar to the signal characteristics in step 103 above, and may include, but are not limited to: 1. obvious monotonicity or periodicity; 2. similarity or correlation between adjacent signal segments; 3. slow variability between adjacent signal segments; 4. a certain degree of repeatability. However, both have the following characteristics:

[0124] The signal features mentioned in step 103 above are used to perform the first splicing of the original signal decomposed by the time window function to obtain signal segments with monotonic or certain periodic characteristics. These signal features include: obvious periodicity (e.g., the time repetition of typical characteristic points of a beat cycle consistent with human physiological characteristics, periodic time-frequency domain characteristics, etc.), strong similarity or correlation between cycles (e.g., small Euclidean distance between cycles, large correlation coefficient, etc.), and relatively slow changes in signal characteristics during the cycle (e.g., the possible amplitude change rate, rising slope change rate, peak-to-peak change rate of a normal pulse wave, etc., consistent with human physiological characteristics).

[0125] The signal features described in step 103 have the following characteristics: 1. Since a fitting function is constructed, the characteristics of the fitting function itself limit many characteristics of the pulse wave signal that can be fitted with it. Therefore, the signal features in step 103 focus more on the similarity and consistency between the fitted signal and the pulse wave signal, whether the signal itself has periodicity, and whether different signal segments have similarity and slow changes. 2. The judgment rules do not emphasize features related to the physiological characteristics of heart rhythm, but only emphasize the features of the signal itself.

[0126] The secondary signal features mentioned in step 104 above are as follows: 1. The rhythm of the pulse wave is first identified, using one or more heartbeat cycles as the segment for feature extraction; 2. The judgment rule emphasizes whether it is similar to a reasonable physiological signal.

[0127] A normal pulse wave, with each beat corresponding to one heart rhythm beat, should have the following characteristics:

[0128] 1. The beat has obvious periodicity (the periodicity emphasized here includes the periodicity between each beat and the periodicity between n beats);

[0129] 2. Strong similarity and correlation during the beat cycle (single beats are similar to each other, and n beats are also similar to each other).

[0130] 3. Relatively slow variability during the beat cycle (between each single beat, or between n beats, the amplitude, length, time domain characteristic value, frequency domain characteristic value, etc., vary within a certain range. If the range is exceeded, it is considered that at least one of the two pulse wave segments is an abnormal pulse wave).

[0131] 4. The beat has a certain continuity (a high-quality, stable pulse wave signal should have a certain period length).

[0132] The periodicity calculation of signal segments can be performed using, but is not limited to, any of the following methods:

[0133] (1) Time-domain analysis method

[0134] For example, calculate whether signal segments with time-domain waveforms that have the same peak, valley, and slope are repeated; calculate the autocorrelation function of the signal and observe whether it has a significant peak under certain specific delays.

[0135] (2) Frequency Domain Analysis Method

[0136] For example, by performing a Fourier transform on a signal, specifically a Discrete Fourier Transform (DFT), periodic signals typically exhibit discrete frequency components in the spectrum.

[0137] (3) Statistical methods

[0138] For example, statistical indicators of signal periodicity, such as periodicity intensity (PPI) or periodicity factor (PF); analyzing the energy distribution of a signal at different frequencies. Periodic signals typically have a high energy concentration at specific frequencies.

[0139] The similarity or correlation between two adjacent signal segments can be calculated using, but is not limited to, any of the following methods:

[0140] (1) Distance metric methods

[0141] For example, calculating the Euclidean distance or Manhattan distance between two adjacent signal segments can be used to assess the similarity or correlation between the two adjacent signal segments.

[0142] (2) Statistical characteristic comparison method

[0143] For example, comparing the statistical characteristics of two adjacent signal segments, such as mean, variance, period length, time length, slope, amplitude, kurtosis, skewness, etc., can help assess the similarity or correlation between the two adjacent signal segments.

[0144] (3) Correlation analysis method

[0145] For example, the Pearson correlation coefficient and cross-correlation function of two adjacent signal segments can be calculated to assess the similarity or correlation between the two adjacent signal segments.

[0146] (4) Time-frequency analysis method

[0147] For example, by using Fourier transform and wavelet transform, two adjacent signal segments can be decomposed into wavelet components of different frequencies. Then, the energy or characteristics of these wavelet components can be compared to assess the similarity or correlation between the two adjacent signal segments. Alternatively, the frequency domain characteristics and time domain characteristics of two adjacent signal segments can be calculated and compared to assess their similarity or correlation.

[0148] (5) Machine Learning Methods

[0149] Machine learning techniques, such as support vector machines and neural networks, are used for pattern recognition and classification to assess the similarity or correlation between two adjacent signal segments.

[0150] The calculation of relatively slow variability during the period can be performed using, but is not limited to, any of the following methods:

[0151] 1. Time Domain Analysis

[0152] For example, calculating the amplitude of a signal fluctuation, and then calculating the difference between the maximum and minimum values ​​based on the fluctuation amplitude, can reflect the degree of change in the signal. Alternatively, calculating the standard deviation of the signal can measure the variability of the signal amplitude.

[0153] 2. Autocorrelation analysis

[0154] For example, by calculating the autocorrelation function, we can observe the similarity of signals under different time delays. If the autocorrelation function decreases rapidly, it indicates that the signal is highly variable; conversely, it indicates that the signal is less variable.

[0155] 3. Frequency Domain Analysis

[0156] For example, performing a Fourier transform on a signal allows you to observe its energy distribution in the frequency domain. A wider spectrum generally indicates greater signal variability because it contains more frequency components. Alternatively, calculating the power spectral density (PSD) can provide a more precise assessment of the signal's frequency domain variability.

[0157] 4. Statistical characteristics

[0158] For example, calculating the entropy of a signal can measure its uncertainty. The higher the entropy value, the greater the variability and complexity of the signal. Alternatively, calculating the coefficient of variation of a signal (the ratio of the standard deviation to the mean) can help assess the relative variability of the signal, and is suitable for comparing signals of different dimensions.

[0159] The following example uses a statistical feature comparison method to illustrate how the signal characteristics such as period length, time length, amplitude, and slope of two adjacent signal segments are compared in this embodiment of the invention. The inherent characteristics of the pulse wave and correlation conditions are set as follows: a confidence interval for period length [p0, p1], a confidence interval for pulse wave time length [t0, t1], a confidence interval for pulse wave amplitude change rate [k0, k1], a confidence interval for pulse wave period length change rate [q0, q1], and a confidence interval for pulse wave slope change rate [s0, s1].

[0160] Set the continuity condition for the number of pulse wave signal segments: confidence value r for the number of continuous wave segments.

[0161] Furthermore, adjacent signal segments P0 and P1 are selected from the target periodic signal set, and the period length L of these two signal segments is calculated. p0 and L p1 (or time length T) p0 and T p1 Calculate the rise amplitude A of the pulse wave period for these two signal segments. u0 and A u1 and the magnitude of the decline A d0 and A d1 Calculate the slope S of the pulse wave signal for these two signal segments. p0 and S p1 If L p0 and L p1 The pulse wave period length confidence interval [p0, p1] is satisfied, and the pulse wave period rise amplitude A u0 and A u1 and the magnitude of the decline A d0 and A d1 The ratio satisfies the confidence interval [k0, k1] of the rate of change of pulse wave amplitude, and the slope of the pulse wave signal is S. p0 and S p1 .

[0162] Continue to refer to Figure 1 In step 105, the signal segments in the target periodic signal set are judged to determine whether they meet the decision conditions based on the secondary signal characteristics of the signal segments. If they meet the decision conditions, the signal segments are spliced ​​together and output as a normal pulse wave signal. If they do not meet the decision conditions, the signal segments are regarded as motion artifacts or noise.

[0163] Specifically, based on the secondary signal characteristics, it is determined whether the self-signal characteristics of two adjacent signal segments in the target periodic signal set, the similarity and correlation between the two signal segments, the variability between the signal segments, and the continuity of the signal segments meet the decision conditions. If the decision conditions are met, these adjacent signal segments are spliced ​​together to form a complete signal output. If the decision conditions are not met, these signals are regarded as motion artifacts or interference signals.

[0164] Repeat the above process until all signal segments in the target periodic signal set have been processed. The resulting spliced ​​signal is the final pulse wave signal with signal interference and motion artifacts removed.

[0165] For example, n1 temporally adjacent signal segments from a pulse wave signal set are spliced ​​together to form a new signal segment. This new signal segment has similar or related pulse wave signal characteristics and represents m1 pulse wave cycles. These spliced ​​signal segments representing m1 pulse wave cycles are then added to the target periodic signal set. This process is repeated until all signal segments in the pulse wave signal set have been processed, and these processed signal segments are then added to the target periodic signal set.

[0166] To improve the system's accuracy in pulse wave recognition, increase the probability of correctly identifying signal segments, and reduce the probability of missing correct signal segments, the system's robustness is enhanced. This makes the recognition of signal segments less susceptible to errors or uncertainties in individual signal features, improving the system's adaptability to abnormal situations. After calculating the periodicity, period-interval similarity and correlation, period-interval variability, and periodic continuity characteristics of the signal segments, a decision fusion method is used to fuse these characteristic results for analysis, ultimately outputting a reasonable and reliable result.

[0167] When determining whether a signal segment in the target periodic signal set satisfies the decision conditions, a decision fusion approach can be used, such as, but not limited to, any of the following methods:

[0168] 1. Voting method

[0169] For example, a simple voting method can be used to determine whether all signal features meet preset conditions. If all or most of the signal segments meet the preset conditions, then the decision rule is satisfied. Alternatively, a weighted voting method can be used, assigning different weights to different signal features. If the signal segments satisfy the preset conditions after weighting, then these signal segments are judged to satisfy the decision rule.

[0170] 2. Stacking method

[0171] For example, by using the feature values ​​of multiple signal segments, a higher-level judgment condition can be generated. If a signal segment meets the judgment condition, then the signal segment is judged to meet the decision rule.

[0172] 3. Integrated learning

[0173] For example, using the Bagging method, multiple models are trained by resampling the signal segment set using random forests, and their predictions are combined to generate a higher-level decision rule. If a signal segment meets a preset condition, it is determined that the signal segment satisfies the decision rule. Alternatively, using Boosting, multiple models are trained by progressively adjusting the sample weights to generate a higher-level decision rule. If a signal segment meets a preset condition, it is determined that the signal segment satisfies the decision rule.

[0174] 4. Bayesian fusion

[0175] By combining the judgment results of different signal segment feature values ​​using Bayes' theorem, a higher-level judgment rule is generated. If a signal segment meets the preset conditions, then these signal segments are judged to satisfy the decision rule.

[0176] These methods can be selected and combined based on specific pulse wave signal attributes and characteristics to achieve more effective decision fusion.

[0177] The following example uses a simple voting method to illustrate how the invention performs decision fusion based on the secondary signal characteristics of the target signal segment set, ultimately outputting a pulse wave signal that removes signal interference and motion artifacts.

[0178] Set the following preset conditions to determine whether a signal segment and its adjacent segments in a target signal set satisfy the following relationship:

[0179]

[0180] or

[0181]

[0182] If the above preset conditions are met, then the two signal segments P0 and P1 satisfy the inherent signal characteristics of a pulse wave. These two signal segments are then spliced ​​together to form a pulse wave periodic signal S0. Otherwise, the two signal segments P0 and P1 are considered noise.

[0183] Repeat the above steps to obtain the combined pulse wave period signal S0……S n .

[0184] Furthermore, the time length L of the aforementioned pulse wave period signals S0 and S1 is calculated. s0 and Ls1 and pulse wave amplitude A s0 and A s1 If the time length L s0 and L s1 The ratio satisfies the confidence interval [q0, q1] of the pulse wave period length variation law, and the pulse wave amplitude A s0 and A s1 The ratio satisfies the confidence interval [s0, s1] of the pulse wave period length variation law, and its calculation relationship is as follows:

[0185]

[0186] If the above conditions are met, then the two combined pulse waves satisfy the correlation condition, and the periodic signals S0 and S1 are denoted as normal pulse wave signals. Otherwise, the periodic signals S0 and S1 are denoted as noise.

[0187] Repeat the above steps. If the number of consecutive normal pulse wave cycles in condition (3) or (4) and (5) is greater than or equal to the confidence value r of the number of consecutive pulse segments (e.g., r = 10, that is, a pulse wave with more than 10 heart rhythm beats is considered a stable and reliable segment), then the entire pulse wave cycle is counted as a normal pulse wave signal; otherwise, the pulse wave cycle is discarded.

[0188] It should be noted that the simple voting decision rules (3), (4), or (5) given in this embodiment are several sets of prior setting conditions based on the characteristics of the pulse wave signal. The above setting conditions are only one example listed in this invention. This invention cannot enumerate other possible setting conditions, and other conditions set by those skilled in the art are also within the protection scope of this invention.

[0189] In this embodiment of the invention, the original pulse wave is decomposed into different signal segments according to a sliding time window function (the sliding time windows may partially overlap). Since the signal segments may contain interference, motion artifacts, noise, and other signals that are inconvenient to process, the segmented signal segments are spliced ​​together to form signal segments with certain signal characteristics. Then, the pulse wave signal characteristics, the signal characteristics between adjacent signals, the rate of change between adjacent signals, and the continuity of the signal segments are calculated. Decision conditions are set, and these signal segments that meet the decision rules are spliced ​​together into a complete signal. This allows for adaptive filtering of signal interference and motion artifacts in the pulse wave signal, achieving the extraction of high-quality pulse wave signals under complex interference and large motion conditions.

[0190] Alternatively, fitting techniques can be used to reduce the resolution of signal segments, blurring them into a fitted signal. The fitted signal highlights macroscopic features while removing microscopic features, making it easier to analyze and process, and leading to more accurate final decisions.

[0191] Accordingly, the present invention also provides a pulse wave signal processing device, such as... Figure 3 The diagram shown is a structural schematic of a pulse wave signal processing device provided by the present invention.

[0192] The pulse wave signal processing device 300 includes the following modules:

[0193] Signal acquisition module 301 is used to acquire raw pulse wave signals;

[0194] The segmentation module 302 is used to truncate the original pulse wave signal using a sliding time window function, construct a fitting function, adjust the parameters of the sliding time window function and / or the parameters of the fitting function, so that the fitting signal corresponding to each truncated signal segment approximates the signal segment; and generate a pulse wave signal set based on the signal segment or the corresponding fitting signal.

[0195] The splicing module 303 is used to splice signal segments in the pulse wave signal set into signal segments with certain signal characteristics to obtain a target periodic signal set.

[0196] The signal detection module 304 is used to detect signal segments in the target periodic signal set, extract heart rhythm signal features and signal features between heart rhythm beats as secondary signal features; determine whether these signal segments meet the decision conditions based on the secondary signal features of the signal segments in the target periodic signal set; if the decision conditions are met, these signal segments are spliced ​​together and output as a normal pulse wave signal; if the decision conditions are not met, these signal segments are regarded as motion artifacts or noise.

[0197] In another non-limiting embodiment, the pulse wave signal processing device 300 may further include a preprocessing module (not shown) for performing noise reduction and filtering processing on the original pulse wave signal.

[0198] The specific implementation methods of the above modules can be found in the descriptions in the previous embodiments of the present invention, and will not be repeated here.

[0199] Accordingly, the present invention also provides a pulse wave signal processing system, such as... Figure 4 The diagram shown is a structural schematic of the system.

[0200] The pulse wave signal processing system includes a signal acquisition device 400 and the aforementioned pulse wave signal processing device 300. In this embodiment, the pulse wave signal processing device 300 further includes a storage module 30.

[0201] The signal acquisition device 400 is used to acquire human pulse wave data and transmits the acquired pulse wave signal for a certain period of time to the pulse wave signal processing device 300.

[0202] The signal acquisition device 400 may include various sensors, such as, but not limited to, photoelectric sensors, pressure sensors, and ultrasonic sensors, which can be used to acquire pulse wave signals. It is used to acquire the raw pulse wave signal of the subject. Acquisition sites include, but are not limited to, locations where pulse wave signals can be acquired, such as the wrist, fingertips, base of the fingers, arm, ankle, neck, and ear. The signal acquisition device 400 can acquire pulse wave signals for a certain duration and then send them to the pulse wave signal processing device 300. The signal acquisition module 301 in the pulse wave signal processing device 300 receives the signal and stores it as a raw pulse wave signal in the storage module 30.

[0203] In some embodiments, the signal acquisition device 400 may further perform noise reduction and filtering on the acquired pulse wave signal before sending it to the pulse wave signal processing device 300. For example, the signal acquisition device 400 may be equipped with a corresponding filtering module, which may be a Butterworth bandpass filter with a passband frequency of 0.5Hz to 20Hz and a filter order of 4.

[0204] In other embodiments, the preprocessing module in the pulse wave signal processing device 300 may first perform noise reduction and filtering on the received raw pulse wave signal, and then perform subsequent processing on the filtered pulse wave signal. This embodiment of the present invention does not limit this.

[0205] After the pulse wave signals in the storage module 30 reach a certain quantity, the segmentation module 302, splicing module 303 and signal detection module 304 perform corresponding calculations and processing to remove motion artifacts and noise, and output high-quality pulse wave signals.

[0206] The following example illustrates the processing procedure for the received pulse wave signal.

[0207] A time window function of length w is preset. The segmentation module 302 extracts a pulse wave signal from the signal acquisition module 301, and extracts a signal of length w from it as the target signal y. i :

[0208]

[0209] Construct a fitting function using the following quadratic polynomials as basis functions:

[0210] f(x j )=a2x j 2 +a1x j +a0,j∈[0,w] (7)

[0211] The time window function is a rectangular window function, with the initial value of the length w set to 0.25s.

[0212] Calculate the fitting function that satisfies the conditions. For example, using the least squares method, fit f(x) to the objective function. By performing a fitting operation, a set of basis function parameters a0, a1, and a2 are obtained. The fitting function f(x) and the objective function are then calculated. The sum of squares of the errors between time windows. A maximum permissible error ε is preset (e.g., ε is 8). The time window length w is iteratively updated, and the iterative update ends when the calculated sum of squares of errors is less than or equal to ε. As shown in the following formula:

[0213]

[0214] At this point, record the pulse wave signal corresponding to the time window length w. To pulse wave signal set Alternatively, the calculated fitted signal f1(x) can be recorded as part of the pulse wave signal set.

[0215] Repeat the above steps until all pulse wave data has been processed. The resulting pulse wave signal set is as follows:

[0216]

[0217] The segmentation module 302 stores the above calculation results into the storage module 30.

[0218] In subsequent processing, the pulse wave signal set... Capturing signal segments or The processing of the fitted signal segments is the same. Therefore, the following uses a pulse wave signal set as an example. The following explanation uses the extraction of signal segments as an example.

[0219] Then, the splicing module 303 extracts the pulse wave signal set stored in the storage module 30. The characteristics of the signal segments are calculated and compared with the set reasonable characteristic range conditions. If the characteristics of the periodic signal segments meet the set conditions, these periodic signals are spliced ​​together into a single periodic signal and recorded in the target periodic signal set.

[0220] Repeat the above steps until the pulse wave signals are collected. All signal segments were processed.

[0221] Specifically, the target signal can be... To calculate the monotonicity of a signal, find its derivative. Then, combine adjacent signal segments with the same monotonicity to form a new signal segment. Specifically, combine adjacent signal segments that are both increasing in monotonicity, and combine adjacent signal segments that are both decreasing in monotonicity.

[0222] After completing the above splicing, if the monotonic properties of two adjacent signal segments are opposite, then continue to splice these two signal segments with opposite monotonic properties together to form a periodic signal with complete fluctuations.

[0223] Furthermore, confidence intervals are set for the rate of change of length of adjacent signal segments [l0, l1], the rate of change of amplitude of adjacent signal segments [k0, k1], and the rate of change of slope of adjacent signal segments [s0, s1]. In the pulse wave signal set... In the process, the current signal segment and the next adjacent signal segment are selected. If the length change rate of the two signal segments meets the range [l0, l1], the amplitude change rate of the two adjacent signal segments meets the range [k0, k1], and the slope change rate of the two adjacent signal segments meets the range [s0, s1], then these periodic signal segments are recorded as a reasonable period; otherwise, the current signal segment is recorded as noise.

[0224] Repeat the above steps to add the spliced ​​periodic signal segments to the target periodic signal set.

[0225] For example, the values ​​of the above parameters are: [l0,l1] is [0.5,1.5], [k0,k1] is [0.5,1.5], and [s0,s1] is [-1.4,-0.6].

[0226] The splicing module 303 stores the splicing result obtained above into the storage module 30.

[0227] Then, the signal detection module 304 extracts the target periodic signal set stored in the storage module 30. The intrinsic characteristics, correlation characteristics, and continuity characteristics of these periodic signal segments are calculated, and these characteristics are compared with the set reasonable characteristic range conditions. If the characteristics of the periodic signal segments meet the set conditions, these periodic signal segments are counted as reasonable pulse waves, and they are spliced ​​together into a continuous pulse wave. Otherwise, this one periodic signal or several periodic signals are counted as noise.

[0228] Specifically, a confidence interval [p0, p1] is set for the length of the periodic signal. If the length of one or more periodic signals meets the range [p0, p1], then the period of this one periodic signal is counted as a reasonable pulse wave period, or several periodic signals are spliced ​​together and counted as a reasonable pulse wave period. Otherwise, this one or several periodic signals are recorded as noise.

[0229] Furthermore, a confidence interval [q0, q1] is set for the rate of change of the length of adjacent pulse wave cycles, and a confidence interval [k0, k1] is set for the rate of change of the amplitude. The current pulse wave cycle signal and the signal of the next adjacent pulse wave cycle are selected. If the rate of change of the length of these two signals satisfies the range [q0, q1], and if the rate of change of the amplitude of these two signals satisfies the range [k0, k1], then these two pulse wave cycle signals are counted as reasonable pulse wave cycles. Otherwise, the current pulse wave cycle signal is recorded as noise.

[0230] Furthermore, setting the minimum correlation coefficient between adjacent periodic signals to ∈, the correlation coefficient ρ between the current periodic signal and the next adjacent periodic signal is calculated using the following formula:

[0231]

[0232] If the correlation coefficient ρ ≥ ∈ between the two periodic signals mentioned above, then these two pulse wave periodic signals are recorded as reasonable pulse wave periods. Otherwise, the current pulse wave periodic signal is recorded as noise.

[0233] Specifically, in this embodiment, the normal heart rate range for a person is generally 40 beats / minute to 220 beats / minute. Therefore, the confidence interval for the periodic signal length [p0, p1] is set to [fs·60 / 220, fs·60 / 40], the confidence interval for the rate of change of adjacent pulse wave period length [q0, q1] is set to [0.5, 1.5], the confidence interval for the rate of change of adjacent pulse wave amplitude [k0, k1] is set to [0.4, 1.6], and the minimum correlation coefficient between the two periodic signals is set to 0.7.

[0234] Repeat the above steps to record the spliced ​​reasonable pulse wave target signals into the set.

[0235] For sets The system splices signal segments and calculates the number of reasonable pulse wave cycles that occur consecutively. A minimum continuous cycle value *r* is set (e.g., *r* is set to 5). If the number of reasonable consecutive pulse wave cycles is greater than or equal to *r*, these consecutive pulse wave cycles are spliced ​​together into a single continuous pulse wave signal. This process is repeated to generate several continuous pulse wave signal segments, which serve as the final output after removing motion artifacts and interference.

[0236] Through the above process, the pulse wave signal processing device 300 can adaptively remove pulse wave signal interference and motion artifacts.

[0237] It should be noted that, in specific applications, the segmentation module 302, splicing module 303 and signal detection module 304 can be independent modules or integrated into one module. This embodiment of the invention does not limit this.

[0238] Compared with existing technologies, the pulse wave signal processing method, apparatus, and system provided by this invention have the following advantages:

[0239] (1) The motion artifacts and noise in the pulse wave signal can be removed without the aid of an accelerometer or other sensors. This overcomes the dependence of the motion artifact removal process on the accelerometer, reduces the complexity of the data in the denoising process, makes the implementation process more real-time, and improves the efficiency of removing motion artifacts from the pulse wave.

[0240] (2) Motion artifacts and noise in pulse wave signals can be removed without the aid of multi-channel pulse wave acquisition. This overcomes the dependence of the motion artifact removal process on multi-channel pulse wave sampling, greatly reduces the amount of data acquisition, reduces the implementation cost and computing cost of hardware equipment, makes the implementation process more real-time, and improves the efficiency of removing motion artifacts from pulse waves.

[0241] (3) This method removes motion artifacts and noise from pulse wave signals without relying on pre-acquired signals for model training. It overcomes the dependence on pre-acquired data and the limitations of pre-trained models due to variations in parameters and signal details caused by different devices and sensors. It sets simple parameters based solely on the fundamental characteristics of the pulse wave signal itself, identifies effective pulse wave periods using the data features of the acquired pulse wave signal, and utilizes the correlation between periods to identify and remove interference signals and motion artifacts while preserving normal pulse wave signals. This significantly improves the adaptability of the method and system, and reduces implementation and computational costs.

[0242] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0243] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0244] In the several embodiments provided by the present invention, it should be understood that the disclosed apparatus can be implemented in other ways.

[0245] The present invention also provides a storage medium, which is a computer-readable storage medium storing a computer program thereon, the computer program being executable when it runs. Figure 1 or Figure 2 The method shown may include some or all of the steps. The storage medium may include read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, etc. The storage medium may also include non-volatile memory or non-transitory memory, etc.

[0246] Accordingly, the present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program executable on the processor, and the processor executes the computer program. Figure 1 or Figure 2 All or part of the steps in the pulse wave signal processing method.

[0247] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article indicates that the preceding and following related objects have an "or" relationship.

[0248] In the embodiments of this application, "multiple" refers to two or more.

[0249] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means.

[0250] The embodiments of the present invention have been described in detail above. Specific implementation methods have been used to illustrate the present invention. The descriptions of the embodiments above are only for the purpose of helping to understand the methods and systems of the present invention, and are merely some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention, and the content of this specification should not be construed as a limitation of the present invention. Therefore, any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A pulse wave signal processing method, characterized in that, The method includes: Acquire and store the raw pulse wave signal; The original pulse wave signal is truncated using a sliding time window function, and a fitting function is constructed. The parameters of the sliding time window function and / or the fitting function are adjusted so that the fitting signal corresponding to each truncated signal segment approximates the original signal segment, thereby obtaining a fitting signal segment corresponding to the original signal segment. A pulse wave signal set is generated based on the original signal segment or the corresponding fitting signal segment. The signal segments in the pulse wave signal set are spliced ​​together to form signal segments with certain signal characteristics to obtain the target periodic signal set. The signal segments in the target periodic signal set are detected, and the heart rhythm signal features and the signal features between heart rhythm beats are extracted as secondary signal features. Based on the secondary signal characteristics of the signal segments in the target periodic signal set, it is determined whether these signal segments meet the decision conditions; if they meet the decision conditions, these signal segments are spliced ​​together and output as a normal pulse wave signal; if they do not meet the decision conditions, these signal segments are regarded as motion artifacts or noise. The step of generating a pulse wave signal set based on the signal segment includes: Generate fitted signals corresponding to each signal segment; The fitted signal is recorded in the pulse wave signal set; The step of constructing a fitting function and adjusting the parameters of the sliding time window function and / or the fitting function to make the fitted signal corresponding to each intercepted signal segment approximate the signal segment includes: Construct a fitting function, and use the fitting function to generate a fitting signal corresponding to the signal segment; Calculate the signal characteristics of the fitted signal and the corresponding signal segment; If the fitted signal and the signal characteristics of the corresponding signal segment meet the set conditions, then the fitted signal is taken as the fitted signal segment corresponding to the signal segment; Otherwise, adjust the time window function parameters and / or fitting function parameters until the fitted signal and the signal characteristics of the corresponding signal segment meet the set conditions; When extracting heart rhythm signal features and signal features between heart rhythm beats as secondary signal features, it is necessary to first identify each heartbeat beat of the spliced ​​signal segment, and then extract the secondary signal features. The step of detecting whether a signal segment satisfies the decision condition based on the secondary signal characteristics of the signal segment in the target periodic signal set includes: Based on the secondary signal characteristics, the self-signal characteristics of two adjacent signal segments in the target periodic signal set, the similarity and correlation between the two signal segments, the variability between the signal segments, and the continuity of the signal segments are judged using a decision fusion method to determine whether the decision conditions are met.

2. The pulse wave signal processing method according to claim 1, characterized in that, The step of generating a pulse wave signal set based on the signal segment includes: The signal segment is recorded in the pulse wave signal set.

3. The pulse wave signal processing method according to claim 1, characterized in that, The fitting function is a monotonic function or a periodic function.

4. The pulse wave signal processing method according to claim 1, characterized in that, The step of splicing signal segments from the pulse wave signal set into signal segments with certain signal characteristics includes: Multiple signal segments that are temporally adjacent and have signal characteristics that meet set conditions are sequentially spliced ​​together to form a signal segment with certain signal characteristics.

5. The pulse wave signal processing method according to claim 4, characterized in that, Use any of the following methods to determine whether these signal segments meet the decision conditions: voting method, stacking method, ensemble learning method, Bayesian fusion method.

6. The pulse wave signal processing method according to any one of claims 1-5, characterized in that, The signal characteristics include any one or more of the following: obvious monotonicity or periodicity, strong similarity or correlation during the period, relatively slow variability during the period, and a certain degree of continuity.

7. The pulse wave signal processing method according to claim 6, characterized in that, Use any of the following methods to calculate the monotonicity or periodicity of a signal: time-domain analysis, frequency-domain analysis, or statistical methods.

8. The pulse wave signal processing method according to claim 6, characterized in that, Use any of the following methods to calculate the similarity or correlation between signals: distance metric method, statistical feature comparison method, correlation analysis method, time-frequency analysis method, machine learning or deep learning method.

9. The pulse wave signal processing method according to claim 6, characterized in that, Use any of the following methods to calculate the variability between signals: time-domain analysis, autocorrelation or cross-correlation analysis, frequency-domain analysis, or statistical methods.

10. The pulse wave signal processing method according to claim 1, characterized in that, The method further includes: Before truncating the original pulse wave signal, the original pulse wave signal is subjected to noise reduction and filtering processing.

11. The pulse wave signal processing method according to claim 10, characterized in that, The filtering process for the original pulse wave signal includes: The original pulse wave signal is denoised and filtered using any of the following filtering methods: low-pass filtering, high-pass filtering, band-pass filtering, median filtering, sliding window filtering, wavelet transform, Kalman filtering, empirical mode decomposition (EMD), variational mode decomposition (VMD), and local mean decomposition (LMD).

12. A pulse wave signal processing device, characterized in that, The device includes: The signal acquisition module is used to acquire and store the raw pulse wave signal; The segmentation module is used to truncate the original pulse wave signal using a sliding time window function, construct a fitting function, adjust the parameters of the sliding time window function and / or the fitting function parameters, so that the fitting signal corresponding to each truncated signal segment approximates the signal segment; and generate a pulse wave signal set based on the signal segment or the corresponding fitting signal. The splicing module is used to splice signal segments in the pulse wave signal set into signal segments with certain signal characteristics to obtain a target periodic signal set. The signal detection module is used to detect signal segments in the target periodic signal set, extract heart rhythm signal features and signal features between heart rhythm beats as secondary signal features; determine whether these signal segments meet the decision conditions based on the secondary signal features of the signal segments in the target periodic signal set; if the decision conditions are met, these signal segments are spliced ​​together and output as a normal pulse wave signal; if the decision conditions are not met, these signal segments are regarded as motion artifacts or noise. in, The step of generating a pulse wave signal set based on the signal segment includes: Generate fitted signals corresponding to each signal segment; The fitted signal is recorded in the pulse wave signal set; The step of constructing a fitting function and adjusting the parameters of the sliding time window function and / or the fitting function to make the fitted signal corresponding to each intercepted signal segment approximate the signal segment includes: Construct a fitting function, and use the fitting function to generate a fitting signal corresponding to the signal segment; Calculate the signal characteristics of the fitted signal and the corresponding signal segment; If the fitted signal and the signal characteristics of the corresponding signal segment meet the set conditions, then the fitted signal is taken as the fitted signal segment corresponding to the signal segment; Otherwise, adjust the time window function parameters and / or fitting function parameters until the fitted signal and the signal characteristics of the corresponding signal segment meet the set conditions; When extracting heart rhythm signal features and signal features between heart rhythm beats as secondary signal features, it is necessary to first identify each heartbeat beat of the spliced ​​signal segment, and then extract the secondary signal features. The step of detecting whether a signal segment satisfies the decision condition based on the secondary signal characteristics of the signal segment in the target periodic signal set includes: Based on the secondary signal characteristics, the self-signal characteristics of two adjacent signal segments in the target periodic signal set, the similarity and correlation between the two signal segments, the variability between the signal segments, and the continuity of the signal segments are judged using a decision fusion method to determine whether the decision conditions are met.

13. The pulse wave signal processing device according to claim 12, characterized in that, The device further includes: The preprocessing module is used to perform noise reduction and filtering on the original pulse wave signal before it is truncated.

14. A pulse wave signal processing system, characterized in that, The system includes: a signal acquisition device, and a pulse wave signal processing device as described in claim 12 or 13; The acquisition device is used to acquire human pulse wave signals and transmit the acquired pulse wave signals for a certain period of time to the pulse wave signal processing device. The pulse wave signal processing device is used to process the human pulse wave signal and output a normal pulse wave signal.

15. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is run by the processor, it performs the steps of the pulse wave signal processing method according to any one of claims 1 to 11.

16. An electronic device comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor runs the computer program, it performs the steps of the pulse wave signal processing method according to any one of claims 1 to 11.

Images