ECG synchronous feature recognition algorithm and device for four-electrode ICG impedance measurement and medium

Abstract

The invention relates to an ECG synchronous feature recognition algorithm and device for four-electrode ICG impedance measurement and a medium, and the algorithm comprises the steps: employing a standard quadrupole method for configuration, and synchronously collecting an ICG impedance change signal and an ECG impedance electrocardiosignal of a human body; a multi-stage multi-order filtering processing method is adopted, the collected ICG impedance change signals and the collected ECG impedance electrocardiosignals are preprocessed, and then ICG de-noised signals and ECG de-noised signals are obtained; drawing an impedance change diagram and an impedance electrocardiogram on the basis of the obtained ICG de-noised signal and the ECG de-noised signal, and identifying main feature points of ICG and ECG in the same period and extracting feature parameters on the basis of the drawn impedance change diagram and impedance electrocardiogram; through the synergistic effect of signal synchronous acquisition, multi-stage filtering and same-period feature recognition, the adaptive capacity of the algorithm to the physiological signal time-varying characteristics is enhanced, the precision and stability of cardiac function evaluation are greatly improved, and the technical defects of an existing ICG measurement algorithm are effectively overcome.

Description

Technical Field

[0001] This invention relates to the field of cardiac function monitoring technology, specifically to an ECG synchronization feature recognition algorithm, device, and medium for four-electrode ICG impedance measurement. Background Technology

[0002] Traditional invasive and minimally invasive cardiac function assessment techniques often rely on invasive procedures or complex equipment setups, causing inconvenience to patients and increasing medical costs. Intrathoracic electro-bioimpedance gait (ICG), as a non-invasive method for monitoring cardiac function, calculates impedance changes by applying a small alternating current to the chest and measuring the resulting voltage drop, thereby indirectly reflecting the heart's activity status.

[0003] However, existing ICG measurement algorithms suffer from low accuracy and poor stability when faced with individual differences, motion artifacts, and environmental interference, and lack the ability to adapt to the time-varying characteristics of physiological signals. Moreover, the eight-electrode acquisition method cannot quickly carry out measurements and physiological parameter calculations in inconvenient or more complex environments for patients, which affects its reliability and effectiveness in clinical applications.

[0004] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to overcome the above-mentioned defects and provide an ECG synchronization feature recognition algorithm, device and medium for four-electrode ICG impedance measurement.

[0006] To solve the above-mentioned technical problems, the technical solution provided by the present invention is as follows:

[0007] On one hand, this invention provides an ECG synchronous feature recognition algorithm for four-electrode ICG impedance measurement, comprising the following steps: S100, using a standard four-electrode configuration, synchronously acquiring ICG impedance change signals and ECG impedance electrocardiogram signals of the human body; S200, using a multi-level, multi-stage filtering processing method to preprocess the acquired ICG impedance change signals and ECG impedance electrocardiogram signals, thereby obtaining ICG denoised signals and ECG denoised signals; S300, based on the acquired ICG denoised signals and ECG denoised signals, drawing an impedance change diagram and an impedance electrocardiogram, and based on the drawn impedance change diagram and impedance electrocardiogram, identifying the main feature points of ICG and ECG in the same cycle and extracting feature parameters; S400, calculating physiological parameters for the extracted main feature points of ICG and ECG, outputting key cardiac function quantitative indicators to provide data support for accurate assessment of cardiac function.

[0008] Optionally, the standard quadrupole configuration for synchronously acquiring ICG impedance change signals and ECG impedance electrocardiogram signals of the human body includes: S110, selecting electrodes with good conductivity and low skin irritation, and arranging them on the surface of the chest and neck of the human body using a quadrupole configuration to ensure that the current is uniformly distributed in the region of interest and reduce the influence of edge effects on the measurement results; S120, implementing a low-frequency sinusoidal AC excitation signal in the FPGA through a program algorithm, setting the frequency range to 50kHz-75kHz; S130, using a high-precision analog-to-digital converter to implement the delay of the lead correction algorithm under algorithm control to eliminate time deviation during signal acquisition; S140, synchronously acquiring the response voltage signals corresponding to the ICG impedance change signals and ECG impedance electrocardiogram signals, setting the sampling rate to not less than 100Hz to ensure the capture of subtle changes within the cardiac cycle; S150, temporarily storing the acquired ICG impedance change signals and ECG impedance electrocardiogram signals in a high-speed cache for subsequent preprocessing unit calls.

[0009] Optionally, the multi-level, multi-order filtering method is used to preprocess the acquired ICG impedance change signal and ECG impedance electrocardiogram signal to obtain the denoised ICG signal and ECG signal, including: S210, using a fourth-order Butterworth low-pass filtering algorithm to filter the ICG impedance change signal and ECG impedance electrocardiogram signal in the 5-50kHz frequency band, retaining the frequency band containing cardiac activity information, suppressing DC offset and high-frequency noise, and outputting them as ICG low-pass signal and ECG low-pass signal respectively; S220, designing a 0.5Hz LMS adaptive high-pass filter to filter the output ICG low-pass signal and ECG low-pass signal. Filtering is performed to remove respiratory signal interference, and stable ICG high-pass and ECG high-pass signals are output respectively; S230, a polynomial baseline removal algorithm is introduced, and the algorithm parameters are dynamically updated according to the real-time acquired data to perform baseline correction on the ICG high-pass and ECG high-pass signals, compensate for long-term baseline trend changes, keep the signals near a stable zero mean, and output ICG baseline-removed signals and ECG baseline-removed signals respectively; S240, wavelet denoising is performed on the output ICG baseline-removed signals and ECG baseline-removed signals respectively to further remove residual noise, and ICG denoised signals and ECG denoised signals that can be used for feature point recognition are output respectively.

[0010] Optionally, based on the acquired denoised ICG and ECG signals, an impedance change graph and an impedance electrocardiogram are plotted. Based on the plotted impedance change graph and impedance electrocardiogram, the main feature points of the ICG and ECG in the same cycle are identified and feature parameters are extracted. This includes: S310, according to the sampling rate setting and clinical recommendations, dividing the ECG denoised signal into data blocks of 500 data points per buffer to obtain several... Denoising signal data blocks and plotting impedance electrocardiograms; S320, using an ECG peak recognition algorithm, for each The denoised signal data block is processed to find the R-peak feature points and extract the time-domain signal. and S330, for multiple Align the first R peak of the denoised signal data block according to Move to the left The minimum interval points are used as the starting point of the heartbeat cycle. If the start of the data block does not meet the requirements The minimum interval points are then calculated from the second... Move to the left Minimum interval points as At the same time, the next Decrease by 1 as the end point of the cycle Repeat this operation and divide the ECG period data; S340, mark the corresponding values ​​in the ECG period data. Peaks, using trough identification algorithms in Find and mark the left and right sides of the peak respectively. Fenghe Peak; S350, extract all from the ECG denoised signal , , The idx and val of the feature points are stored in a global variable structure and marked on the impedance electrocardiogram for use in the synchronous calculation of the ICG signal.

[0011] Optionally, the step of plotting an impedance change graph and an impedance electrocardiogram based on the acquired ICG denoised signal and ECG denoised signal, and identifying the main feature points of ICG and ECG in the same period and extracting feature parameters based on the plotted impedance change graph and impedance electrocardiogram, further includes: S360, inputting the ICG denoised signal, dividing the data into blocks according to the sampling rate synchronized with ECG, and obtaining... Denoising data blocks and establishing a global coordinate indexing system to ensure that the coordinates of ICG data blocks are consistent with... Maintain specific timing alignment and plot impedance changes; S370, using three-point and five-point center differential methods for... Differential calculations are performed on the denoised data blocks to obtain... Data block, and plot impedance differential graph; S380, traversal The data block uses the ICG peak identification algorithm to identify the maximum point in the impedance differential plot and determines it as... Points, and extract and S390. In the global coordinate system, and according to the size of the ECG periodic data block, the ICG data is periodically divided synchronously, and all data are marked on the impedance differential diagram. Point; S3100, based on the extracted Search to the left and confirm. Point location and extract S3110, according to Find the minimum point on the impedance differential graph to the right, and determine it as... Point, Extract Complete ICG periodic data , , Simultaneous identification of key feature points.

[0012] Optionally, physiological parameters are calculated for the extracted main feature points of the ICG and ECG, including: S410, retrieving the identified parameters from the global variable structure. , , and , , The idx and val data of feature points serve as the basis for calculating physiological parameters; S420, based on the data identified in ECG cycle data. Peaks, extracting adjacent peaks within the same period The peak idx difference is used to calculate the heart rate; S430, based on the ICG cycle data identified... , , Point, Extract and The time-domain difference is used to calculate the stroke volume; S440: The heart rate and stroke volume are calculated to obtain the cardiac output; S450: Based on the calculated heart rate, stroke volume, cardiac output and the characteristic parameters in the global variable structure, the parameters of cardiac output index, stroke index and vascular resistance are further calculated; S460: Based on the parameters calculated in S450, key quantitative indicators of cardiac function are output to provide data support for the accurate assessment of cardiac function.

[0013] Optionally, in step S420, the formula for calculating the heart rate is as follows:

[0014]

[0015] HR stands for heart rate, which is the number of heartbeats per minute. It refers to the index position of the i-th R-peak in the data block, and the sampling rate refers to the number of data points of the signal collected per unit time.

[0016] Optionally, in step S430, the formula for calculating the stroke volume is as follows:

[0017]

[0018] Where ρ is the blood conductivity, L is the distance between the second and third electrodes, Z0 is the fundamental impedance, and MAX is the impedance differential. Figure 3 The corresponding medium cycle Amplitude value. LVET is the left ventricular ejection interval.

[0019] On the other hand, the present invention also provides an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to execute an ECG synchronization feature recognition algorithm for four-electrode ICG impedance measurement.

[0020] On the other hand, the present invention also provides a computer-readable storage medium storing computer instructions for causing a computer to execute an ECG synchronization feature recognition algorithm for four-electrode ICG impedance measurement.

[0021] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0022] This invention constructs a complete four-electrode ICG impedance measurement and ECG synchronous feature recognition system. This scheme uses a standard four-electrode configuration instead of the traditional eight-electrode configuration, significantly simplifying the setup process of the measurement device. It eliminates the need for complex equipment settings, facilitating rapid measurement in situations where patients are physically unavailable or in complex clinical environments, effectively addressing the inconvenience of the eight-electrode acquisition method. Simultaneously, by synchronously acquiring human ICG impedance change signals and ECG impedance electrocardiogram signals, and combining this with multi-level, multi-stage filtering processing methods for targeted preprocessing of the raw signals, external interference can be effectively suppressed while accurately preserving key physiological information related to cardiac activity, significantly improving signal purity. Furthermore, by plotting impedance change graphs and impedance electrocardiograms, accurate identification of key feature points of ICG and ECG during the same cycle and extraction of feature parameters are achieved. Finally, key cardiac function quantitative indicators are calculated and output based on physiological parameters. The entire technical solution not only realizes non-invasive operation of cardiac function assessment, completely avoiding the physical discomfort and high medical costs caused to patients by traditional invasive / minimally invasive techniques, but also enhances the algorithm's adaptability to the time-varying characteristics of physiological signals through the synergistic effect of synchronous signal acquisition, multi-level filtering and same-cycle feature recognition, which greatly improves the accuracy and stability of cardiac function assessment and effectively makes up for the technical shortcomings of existing ICG measurement algorithms. Attached Figure Description

[0023] Figure 1 This is a flowchart of the ECG synchronization feature recognition algorithm for four-electrode ICG impedance measurement provided in this embodiment of the invention;

[0024] Figure 2 This is a sub-flowchart of the ECG synchronization feature recognition algorithm S100 for four-electrode ICG impedance measurement provided in this embodiment of the invention.

[0025] Figure 3 This is a sub-flowchart of the ECG synchronization feature recognition algorithm S200 for four-electrode ICG impedance measurement provided in this embodiment of the invention;

[0026] Figure 4 This is a sub-flowchart of the ECG synchronization feature recognition algorithm S300 for four-electrode ICG impedance measurement provided in this embodiment of the invention;

[0027] Figure 5 This is a sub-flowchart of the ECG synchronization feature recognition algorithm S400 for four-electrode ICG impedance measurement provided in this embodiment of the invention.

[0028] Figure 6 This is a schematic diagram of the flowchart of the ECG synchronization feature recognition algorithm for four-electrode ICG impedance measurement provided in an embodiment of the present invention.

[0029] Figure 7 This is the impedance electrocardiogram provided in the embodiments of the present invention;

[0030] Figure 8 This is an impedance variation diagram provided in an embodiment of the present invention;

[0031] Figure 9 This is the impedance differential diagram provided in the embodiments of the present invention;

[0032] Figure 10 This is a feature point calculation and feature parameter extraction annotation diagram provided in the embodiments of the present invention.

[0033] Figure 11 This is a graph showing the calculation results of the example parameters provided in the embodiments of the present invention. Detailed Implementation

[0034] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses consistent with some aspects of this disclosure as detailed in the appended claims.

[0035] In existing technologies, traditional cardiac function assessment techniques rely on invasive or minimally invasive procedures, which can cause patient discomfort and are complex to perform. While intrathoracic electro-bioimpedance technology offers the advantage of being non-invasive, current algorithms often employ an eight-electrode configuration, which is difficult to deploy in complex environments and is susceptible to individual differences, motion artifacts, and environmental noise, resulting in insufficient signal stability and limited feature extraction accuracy. For example, in emergency settings, the need for rapid device deployment conflicts with the complexity of multi-electrode operations. Furthermore, respiratory interference and baseline drift can mask crucial information about cardiac activity, affecting the reliability of cardiac output calculations.

[0036] To address these issues, researchers noted that while a four-electrode configuration simplifies equipment deployment, it requires solving the challenges of synchronized signal acquisition and noise suppression. Analysis revealed that the temporal synchronization of impedance and ECG signals is crucial for feature point identification, and traditional single-stage filtering cannot effectively separate respiratory interference from high-frequency noise. Further research showed that multi-stage filtering combined with dynamic baseline correction significantly improves signal quality. Based on this, a method combining impedance differential analysis with ECG cycle segmentation was proposed to achieve precise alignment of cross-modal feature points.

[0037] Therefore, this application proposes a technical solution including the following steps: synchronously acquiring impedance change signals and electrocardiogram signals of the human body; preprocessing through multi-level multi-order filtering; drawing impedance spectra based on the denoised signals and identifying synchronous feature points; and finally calculating and outputting cardiac function indicators through feature parameters. This application concludes in the following manner.

[0038] Example 1:

[0039] Please refer to the instruction manual appendix. Figures 1 to 11 As shown in the figure, this embodiment provides an ECG synchronization feature recognition algorithm for four-electrode ICG impedance measurement. The method includes the following steps:

[0040] The S100 uses a standard four-pole configuration to simultaneously acquire ICG (thoracic electrophysiological impedance) impedance change signals and ECG (electrocardiogram) impedance electrocardiogram signals from the human body.

[0041] S200. A multi-level, multi-stage filtering method is used to preprocess the acquired ICG impedance change signal and ECG impedance electrocardiogram signal to obtain the ICG denoised signal and ECG denoised signal.

[0042] S300. Based on the acquired ICG denoised signal and ECG denoised signal, draw an impedance change diagram and an impedance electrocardiogram, and based on the drawn impedance change diagram and impedance electrocardiogram, identify the main feature points of ICG and ECG under the same cycle and extract feature parameters.

[0043] S400: Calculate physiological parameters for the main feature points of the extracted ICG and ECG, and output key quantitative indicators of cardiac function to provide data support for accurate assessment of cardiac function.

[0044] In this embodiment, the raw bioimpedance and electrocardiogram (ECG) signals are first acquired synchronously using a four-electrode system, with the current excitation frequency set within a specific high-frequency range to reduce the influence of skin impedance. The preprocessing stage employs a cascaded filtering strategy: first, a steeply rolled-down low-pass filter preserves the cardiac activity frequency band; then, an adaptive high-pass filter eliminates respiratory interference; and finally, dynamic baseline correction compensates for long-term drift. In the feature recognition stage, the ECG R-peak is used as the period segmentation benchmark. The extreme points of the impedance differential signal are analyzed within the corresponding time window to determine the key time nodes of cardiac systole and diastole. During parameter calculation, the time interval and waveform amplitude between feature points are measured, and combined with known parameters such as blood conductivity, to derive core indicators such as stroke volume.

[0045] Compared to existing technologies, traditional eight-electrode systems require additional electrodes to detect respiratory interference. This solution replaces hardware redundancy with multi-level filtering at the algorithm level, simplifying the device structure while ensuring signal quality. Existing technologies typically process ECG and impedance signals separately; this solution innovatively establishes a cross-signal time correlation model, utilizing the explicit periodicity of the ECG signal to guide impedance feature extraction and improve the temporal accuracy of parameter calculation. Compared to fixed-parameter filtering methods, the dynamic baseline correction algorithm can adapt to the individualized physiological characteristics of different patients, especially performing better when dealing with slow baseline drift caused by postural changes.

[0046] Example 2:

[0047] Based on the above embodiments, in order to provide a clearer and more complete explanation of the technical solutions therein, this application also provides Embodiment Two. For example... Figures 2 to 11 As shown in this second embodiment, this application further proposes using a standard four-pole configuration to simultaneously acquire the ICG impedance change signal and ECG impedance electrocardiogram signal of the human body, i.e. Figure 2 As shown, S100 in this application may further include:

[0048] S110. Select electrodes with good conductivity and low skin irritation, and arrange them on the surface of the human chest and neck using the four-electrode method to ensure that the current is evenly distributed in the region of interest and reduce the influence of edge effects on the measurement results.

[0049] For example, medical conductive gel patch electrodes can be used, whose conductivity ensures uniform current distribution and reduces signal distortion caused by contact resistance.

[0050] S120: Implements low-frequency sinusoidal AC excitation signals within the FPGA (Field Programmable Gate Array) using a program algorithm, with a set frequency range of 50kHz-75kHz.

[0051] For example, direct digital frequency synthesis technology can be used to generate stable waveforms in the range of 50kHz-75kHz, thus avoiding the impact of high-frequency interference on biological signals.

[0052] S130. A high-precision analog-to-digital converter is used to realize the delay of the lead correction algorithm under algorithm control, so as to eliminate the time deviation in the signal acquisition process.

[0053] A high-precision analog-to-digital converter (ADC) refers to a conversion device with a resolution of 16 bits or more. For example, a Δ-Σ type ADC can be used to digitize a signal and a lead compensation algorithm can be used to compensate for the sampling clock delay, thereby eliminating phase errors in the signal transmission path.

[0054] S140: Simultaneously acquire the ICG impedance change signal and the corresponding response voltage signal of the ECG impedance electrocardiogram signal, and set the sampling rate to be no less than 100Hz to ensure the capture of subtle changes within the cardiac cycle.

[0055] Synchronous acquisition refers to controlling the simultaneous start of ICG and ECG signal acquisition modules through a hardware triggering mechanism. For example, using the same clock source to drive two ADC (analog-to-digital converter) circuits to ensure timestamp alignment.

[0056] S150. The acquired ICG impedance change signal and ECG impedance electrocardiogram signal are temporarily stored in a high-speed cache for subsequent preprocessing unit to call.

[0057] A cache is a storage unit with a dual-port structure. For example, SRAM or FIFO memory can be used to temporarily store raw signal data for subsequent processing modules to read as needed.

[0058] like Figure 3 As shown in this second embodiment, this application further proposes to use a multi-level, multi-stage filtering method to preprocess the acquired ICG impedance change signal and ECG impedance electrocardiogram signal, thereby obtaining the ICG denoised signal and ECG denoised signal. That is, S200 of this application may also include:

[0059] S210. A fourth-order Butterworth low-pass filtering algorithm is used to filter the ICG impedance change signal and ECG impedance electrocardiogram signal in the 5-50kHz frequency band, retaining the frequency band containing cardiac activity information, suppressing DC offset and high-frequency noise, and outputting them as ICG low-pass signals and ECG low-pass signals respectively.

[0060] The fourth-order Butterworth low-pass filter algorithm refers to a fourth-order low-pass filter with a flat passband response. For example, a filter design with a cutoff frequency of 50kHz can be used to preserve the characteristic frequency band of cardiac activity while suppressing high-frequency noise.

[0061] S220. Design a 0.5Hz LMS (Least Mean Square) adaptive high-pass filter to filter the output ICG low-pass signal and ECG low-pass signal, filter out respiratory signal interference, and output stable ICG high-pass signal and ECG high-pass signal respectively.

[0062] LMS adaptive high-pass filter refers to an adaptive filter based on the minimum mean square error criterion. For example, by setting a cutoff frequency of 0.5Hz, the filter coefficients are adjusted in real time to eliminate low-frequency interference in the respiratory signal.

[0063] S230. Introduce a polynomial baseline correction algorithm, dynamically update the algorithm parameters based on real-time acquired data, perform baseline correction on the ICG high-pass signal and ECG high-pass signal, compensate for long-term baseline trend changes, keep the signal near a stable zero mean, and output the ICG baseline correction signal and ECG baseline correction signal respectively.

[0064] For example, a cubic polynomial fitting method can be used to eliminate signal baseline drift by dynamically updating the fitting parameters.

[0065] S240. Perform wavelet denoising on the output ICG de-drift signal and ECG de-drift signal respectively to further remove residual noise, and output ICG denoised signal and ECG denoised signal that can be used for feature point recognition respectively.

[0066] For example, the sym8 wavelet basis function can be used for multi-scale decomposition, and residual noise in the signal can be removed by soft thresholding.

[0067] In summary, the preprocessing process (S200) described above adopts a phased, progressive processing architecture. Compared with existing technologies, traditional preprocessing methods often use single-order filters or fixed-parameter denoising, making it difficult to eliminate different interference sources simultaneously. Common respiratory interference suppression techniques in existing technologies often employ high-pass filtering with a fixed cutoff frequency, which cannot adapt to individual differences in respiratory rate. This solution combines frequency domain filtering with time domain correction through a multi-level processing architecture, which can more effectively maintain the integrity of the signal morphology compared to traditional methods. In particular, the application of adaptive filters can automatically adjust parameters according to the actual respiratory rate, avoiding the loss of effective signal caused by a fixed cutoff frequency.

[0068] Through the above technical solution, this application achieves hierarchical elimination of multi-source interference in bioimpedance signals, effectively solving the problems of residual respiratory interference and baseline drift existing in traditional methods. Fourth-order low-pass filtering preserves the complete characteristic frequency band of cardiac activity, adaptive high-pass filtering accurately removes respiratory interference without affecting the effective signal, dynamic baseline correction compensates for signal shifts in long-term measurements, and wavelet denoising further improves the signal-to-noise ratio. This progressive processing architecture improves the accuracy of subsequent feature point identification, such as reducing the error in R-wave peak detection, providing a reliable signal basis for the accurate calculation of cardiac function parameters.

[0069] like Figure 4 As shown in this second embodiment, this application further proposes to plot impedance change diagrams and impedance electrocardiograms based on the acquired denoised ICG and ECG signals, and to identify the main feature points of ICG and ECG in the same period and extract feature parameters based on the plotted impedance change diagrams and impedance electrocardiograms. That is, S300 of this application may also include the following steps:

[0070] S310. Based on the sampling rate setting and clinical recommendations, the ECG denoised signal is divided into data blocks of 500 data points each, resulting in several... Denoise the signal data block and plot the impedance electrocardiogram (e.g.) Figure 7 (As shown).

[0071] For example, a circular buffer can be used to ensure that each data block covers the entire heartbeat cycle by setting the data block length to match the sampling rate.

[0072] S320, employs the ECG peak recognition algorithm for each The denoised signal data block is processed to find the R-peak feature points and extract the time-domain signal. (ECG R-wave index, i.e., R-wave characteristic points at) The position number in the denoised signal data block is used to locate the time coordinates of the R wave, which is the reference for dividing the heartbeat cycle. (The amplitude of the R wave in an electrocardiogram, i.e., the voltage value of the ECG signal corresponding to the characteristic point of the R wave, reflects the intensity of the R wave and can help determine the quality of the ECG signal.)

[0073] S330, for multiple Align the first R peak of the denoised signal data block according to Move to the left (RR interval on an electrocardiogram, i.e., two consecutive intervals) The time interval between waves is a core parameter for calculating heart rate. The minimum interval point is used as the starting point of the heartbeat cycle. (ECG cycle start point); if the data block start point does not meet the requirements The minimum interval points are then calculated from the second... Move to the left Minimum interval points as At the same time, the next Decrease by 1 as the end point of the cycle (End of ECG cycle), repeat this operation and divide the ECG cycle data.

[0074] For example, when the sampling rate is 100Hz, the minimum interval points corresponding to a heart rate of 180 beats / minute are 33.

[0075] S340. Mark the corresponding data in the ECG cycle data. Peaks, using trough identification algorithms in Find and mark the left and right sides of the peak respectively. Fenghe peak.

[0076] The trough identification algorithm is a method based on local minimum search. Specifically, it can be implemented by combining derivative zero-crossing detection with amplitude screening to locate the starting point of Q waves and S waves.

[0077] S350, Extract all from the ECG denoised signal (Q wave on electrocardiogram) (Electrocardiogram R wave) The idx and val of the (ECG S wave) feature points are stored in a global variable structure and marked on the impedance ECG for use in synchronous calculation of the ICG signal.

[0078] In summary, during the ECG signal processing stage (S300), the denoised ECG signal is first divided into fixed-length data blocks. An ECG peak identification algorithm is used to locate the R-peak feature point within each data block. Subsequently, the starting point of the heartbeat cycle is dynamically adjusted based on the R-peak position. When the starting point of a data block cannot meet the minimum interval requirement, a backward recursive approach is used to redefine the cycle boundary, ensuring that each cycle contains a complete ECG waveform. Within the divided cycle data, the troughs of the Q and S waves are searched for on both sides based on the R-peak position. Finally, the timing information and amplitude parameters of each feature point are uniformly stored in a global structure, providing a time alignment reference for subsequent ICG signal synchronization analysis.

[0079] Compared to existing technologies, traditional methods typically divide ECG cycles using fixed time windows, failing to consider individual heart rate differences and incomplete signal initiation, leading to errors in feature point localization. This proposed solution dynamically adjusts the cycle start and boundary, combined with a data block recursion mechanism, to adapt to signal processing needs under different heart rate conditions. It also effectively solves the problem of cycle truncation at the beginning of data blocks, improving the synchronization and completeness of feature parameter extraction.

[0080] Through the above technical solution, this application achieves accurate identification and period division of ECG signal feature points, solving the waveform truncation problem caused by fixed window division in traditional methods, and ensuring complete capture of each feature point of the QRS complex wave. By storing feature parameters through a global variable structure, a unified time reference is provided for synchronous analysis of ICG signals, effectively improving the accuracy of subsequent physiological parameter calculations.

[0081] like Figure 4 As shown, in this second embodiment, S300 may further include:

[0082] S360. Input the ICG denoising signal, divide the data into blocks according to the sampling rate synchronized with ECG, and obtain... Denoising data blocks and establishing a global coordinate indexing system to ensure that the coordinates of ICG data blocks are consistent with... Maintain specific timing alignment and plot impedance changes (e.g.) Figure 8 (As shown).

[0083] Among them, the global coordinate indexing system refers to establishing a unified time-series coordinate system to... Denoising data blocks and all of the ECG denoising signals , , Feature points can be aligned along the timeline, for example, by mapping data block indices to timestamps.

[0084] S370, using the three-point and five-point central difference methods to... Differential calculations are performed on the denoised data blocks to obtain... Data blocks, and plot impedance differential diagrams (e.g.) Figure 9 (As shown).

[0085] The three-point and five-point center difference methods refer to calculating the differential value of the signal based on the numerical changes of adjacent sampling points. Specifically, they can be implemented using sliding window differential operations to accurately capture the extreme points of the slope of the impedance change curve.

[0086] S380, Traversal The data block (ICG impedance differential data block) uses the ICG peak identification algorithm to identify the maximum value point in the impedance differential graph and determine it as... Point (the point where the differential of the ICG impedance reaches its maximum value), and extract ( (point index) and ( Point amplitude);

[0087] The ICG peak identification algorithm determines the local maximum value by comparing the amplitude changes of adjacent data points. Specifically, it can be implemented by combining threshold comparison with extreme value verification, and is used to reliably identify the location of characteristic peaks in the impedance differential diagram.

[0088] S390. In the global coordinate system, and according to the size of the ECG periodic data block, synchronously divide the ICG data into periods, and mark all of them on the impedance differential diagram. point;

[0089] S3100, based on the extracted Search to the left and confirm. Point (starting point of ICG impedance differentiation) location and extraction ( Point index);

[0090] S3110, according to Find the minimum point on the impedance differential graph to the right, and determine it as... Point (minimum point of ICG impedance differential), extract ( Point index); complete ICG periodic data. , , Synchronous identification of key feature points (e.g.) Figure 10 (As shown).

[0091] Compared to existing technologies, traditional methods process ICG and ECG signals independently, which leads to inaccurate feature point correspondences due to time synchronization errors. This solution achieves precise alignment of the time axes of the two signals through a global coordinate indexing system and enhances feature point recognition sensitivity through differential operations, thus resolving parameter calculation errors caused by phase deviations between signals.

[0092] Through the above technical solutions, this application achieves precise time synchronization of ICG and ECG signal features, effectively improving the accuracy of cardiac function parameter calculation. By enhancing feature point recognition capabilities through differential operations, it overcomes the problem of missed feature point detection under low signal-to-noise ratio conditions inherent in traditional methods. The establishment of a global coordinate system ensures the temporal consistency of different physiological signals, providing a reliable data foundation for subsequent parameter calculations.

[0093] like Figure 5 As shown, in this second embodiment, S400 may further include:

[0094] S410. Retrieve the identified variable from the global variable structure. , , and , , The idx and val data of feature points serve as the basis for calculating physiological parameters.

[0095] The global variable structure refers to a dynamic data container used to store multi-cycle signal feature point information. Specifically, it can be implemented using a hash table or linked list data structure, and real-time updates ensure complete retrieval of feature data.

[0096] S420, identified from ECG cycle data Peaks, extracting adjacent peaks within the same period The peak idx difference is used to calculate the heart rate, and the calculation formula is as follows:

[0097] (1)

[0098] HR stands for heart rate, which is the number of heartbeats per minute. This refers to the index position of the i-th R-peak in the data block. The sampling rate refers to the number of data points of the signal collected per unit time. Specifically, a sampling frequency of not less than 100Hz can be used to ensure that the time position of the R-peak in the ECG signal can be accurately captured.

[0099] For example, when the sampling rate is 100Hz, if the index difference between adjacent R-peaks is 500 points, the corresponding time interval is 5 seconds, and the calculated heart rate is 60 / (5 / 60) = 72 beats / minute. This formula eliminates human measurement errors and achieves rapid quantification of heart rate parameters by synchronously dividing data blocks and automatically identifying R-peaks.

[0100] Through the above technical solution, this application can accurately quantify the heart rate, providing reliable basic data for the subsequent calculation of parameters such as cardiac output and vascular resistance, effectively improving the accuracy of cardiac function assessment. By deeply integrating formulaic calculations with signal processing, it solves the technical shortcomings of traditional methods in dynamic monitoring scenarios, such as poor real-time performance and weak anti-interference capabilities.

[0101] S430, based on the identification in the ICG periodic data , , Point, Extract and The stroke volume is calculated from the time-domain difference, using the following formula:

[0102] (2)

[0103] Where ρ is the blood conductivity (the electrical conductivity per unit volume of blood), L is the distance between the second and third electrodes (the physical distance between the flow injection electrode and the voltage detection electrode), Z0 is the fundamental impedance, and MAX is the impedance differential. Figure 3 The corresponding medium cycle Amplitude value. LVET is the left ventricular ejection interval.

[0104] S440. By calculating the heart rate and stroke volume, the cardiac output is calculated.

[0105] Cardiac output is the product of heart rate and stroke volume in the same cycle, that is:

[0106] (3)

[0107] S450, based on the calculated heart rate, stroke volume, cardiac output, and characteristic parameters in the global variable structure, further completes the parameter calculation of cardiac output index, stroke index, and vascular resistance.

[0108] S460, based on the parameters calculated in S450, outputs key quantitative indicators of cardiac function to provide data support for the accurate assessment of cardiac function (such as...). Figure 11 (As shown).

[0109] Example 3:

[0110] Based on the same general inventive concept, this invention proposes an electronic device including at least one processor and a memory communicatively connected to the at least one processor. The memory stores instructions executable by the at least one processor, which, when executed, enable the at least one processor to perform an ECG synchronization feature recognition algorithm for four-electrode ICG impedance measurement.

[0111] In some specific implementations, the processor may employ a multi-core architecture to process filtering and feature extraction tasks in parallel, and the memory may be divided into an instruction storage area and a data cache area to improve access efficiency. The communication connection may adopt a dual-bus structure, transmitting control instructions and sampled data separately.

[0112] Example 4:

[0113] Based on the same general inventive concept, this invention proposes a computer-readable storage medium that stores computer instructions for enabling a computer to execute an ECG synchronous feature recognition algorithm for four-electrode ICG impedance measurement. This algorithm includes synchronously acquiring ICG impedance change signals and ECG impedance electrocardiogram signals from the human body, obtaining denoised signals through multi-level multi-stage filtering preprocessing, identifying feature points in the same cycle based on impedance change maps and impedance electrocardiograms, extracting parameters, and finally calculating physiological parameters to output quantitative indicators of cardiac function.

[0114] Therefore, in summary, the advantages of this application and its embodiments compared to the prior art are as follows: This invention constructs a complete four-electrode ICG impedance measurement and ECG synchronous feature recognition system. This scheme uses a standard four-electrode configuration to replace the traditional eight-electrode configuration, significantly simplifying the arrangement of the measurement device. It eliminates the need for complex equipment setup, facilitating rapid measurement in situations where the patient is physically unwell or in complex clinical environments, effectively solving the inconvenience problem of the eight-electrode acquisition method. Simultaneously, by synchronously acquiring human ICG impedance change signals and ECG impedance electrocardiogram signals, and combining multi-level, multi-stage filtering processing methods to perform targeted preprocessing of the original signals, it can effectively suppress external interference and accurately retain key physiological information related to cardiac activity, significantly improving signal purity. Furthermore, by plotting impedance change diagrams and impedance electrocardiograms, it achieves accurate identification of key feature points of ICG and ECG in the same cycle and extraction of feature parameters, ultimately combining physiological parameters to calculate and output key quantitative indicators of cardiac function. The entire technical solution not only realizes non-invasive operation of cardiac function assessment, completely avoiding the physical discomfort and high medical costs caused to patients by traditional invasive / minimally invasive techniques, but also enhances the algorithm's adaptability to the time-varying characteristics of physiological signals through the synergistic effect of synchronous signal acquisition, multi-level filtering and same-cycle feature recognition, which greatly improves the accuracy and stability of cardiac function assessment and effectively makes up for the technical shortcomings of existing ICG measurement algorithms.

[0115] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0116] Furthermore, the terms "system" and "network" are often used interchangeably in this paper. The term "and / or" in this paper merely describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " in this paper generally indicates that the preceding and following related objects have an "or" relationship.

[0117] It should be understood that, in the embodiments of the present invention, "B corresponding to A" means that B is associated with A, and B can be determined based on A. However, it should also be understood that determining B based on A does not mean that B is determined solely based on A; B can also be determined based on A and / or other information.

[0118] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0119] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0120] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections through some interfaces, apparatuses, or units, or they may be electrical, mechanical, or other forms of connection.

[0121] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention, depending on actual needs.

[0122] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0123] From the above description of the embodiments, those skilled in the art will clearly understand that the present invention can be implemented in hardware, firmware, or a combination thereof. When implemented in software, the above-described functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of a computer program from one place to another. Storage media can be any available medium accessible to a computer. For example, but not limited to, computer-readable media can include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code having the form of instructions or data structures and accessible to a computer. Furthermore, any connection can suitably be a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of the medium. As used in this invention, disk and disc include compressed optical discs (CDs), laser discs, optical discs, digital versatile discs (DVDs), floppy disks, and Blu-ray discs, wherein disks typically magnetically copy data, while discs optically copy data using lasers. The combinations described above should also be included within the scope of protection for computer-readable media.

[0124] In summary, the above description is merely a preferred embodiment of the technical solution of the present invention and is not intended to limit the scope of protection of the present invention. 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. An ECG-synchronized feature recognition algorithm for four-electrode ICG impedance measurements, characterized in that, The method comprises the following steps: S100, using a standard four-pole method configuration, synchronously collecting ICG impedance change signals and ECG impedance electrocardio signals of a human body; S200, using a multi-stage multi-order filtering processing method, preprocessing the collected ICG impedance change signals and ECG impedance electrocardio signals, and then obtaining ICG denoising signals and ECG denoising signals; S300, based on the obtained ICG denoising signals and ECG denoising signals, drawing impedance change graphs and impedance electrocardio graphs, and based on the drawn impedance change graphs and impedance electrocardio graphs, identifying main feature points of ICG and ECG in the same cycle and extracting feature parameters; S400, calculating physiological parameters of the extracted main feature points of ICG and ECG, and outputting key cardiac function quantitative indexes to provide data support for accurate evaluation of cardiac function.

2. The ECG synchronized feature recognition algorithm for four-electrode ICG impedance measurement of claim 1, wherein, The method of synchronously collecting ICG impedance change signals and ECG impedance electrocardio signals of a human body using a standard four-pole method configuration comprises the following steps: S110, selecting electrodes with good conductivity and small skin irritation, and arranging the electrodes on the surface of the chest and neck of the human body in a four-pole method configuration, to ensure that the current is uniformly distributed in the region of interest and to reduce the influence of edge effects on the measurement results; S120, implementing a low-frequency sinusoidal alternating excitation signal in the FPGA through a program algorithm, and setting the frequency range to 50 kHz-75 kHz; S130, using a high-precision analog-to-digital converter to realize the delay of the lead correction algorithm under the control of the algorithm to eliminate the time deviation in the signal acquisition process; S140, synchronously collecting response voltage signals corresponding to the ICG impedance change signals and the ECG impedance electrocardio signals, and setting the sampling rate to be not less than 100 Hz to ensure that the subtle changes in the cardiac cycle are captured; S150, temporarily storing the collected ICG impedance change signals and ECG impedance electrocardio signals in a cache for subsequent calling by a preprocessing unit.

3. The ECG synchronized feature identification algorithm for four-electrode ICG impedance measurement of claim 1, wherein, The method of preprocessing the collected ICG impedance change signals and ECG impedance electrocardio signals using a multi-stage multi-order filtering processing method, and then obtaining ICG denoising signals and ECG denoising signals, comprises the following steps: S210, using a fourth-order Butterworth low-pass filtering algorithm to filter the ICG impedance change signals and the ECG impedance electrocardio signals in the 5-50 kHz frequency band, to retain the frequency band containing heart activity information, suppress direct current offset and high-frequency noise, and output the ICG low-frequency signals and the ECG low-frequency signals, respectively; S220, designing a 0.5 Hz LMS adaptive high-pass filter to filter the output ICG low-frequency signals and ECG low-frequency signals, to filter out respiratory signal interference, and output smooth ICG high-frequency signals and ECG high-frequency signals, respectively; S230, introducing a polynomial baseline removal algorithm, dynamically updating algorithm parameters according to real-time acquisition data, performing baseline correction on the ICG high-frequency signals and the ECG high-frequency signals, compensating for baseline long-term trend changes, keeping the signals near a stable zero mean value, and outputting ICG baseline drift signals and ECG baseline drift signals, respectively; S240, respectively, the output of the ICG de-biasing signal and ECG de-biasing signal signal wavelet noise reduction is performed, further remove residual noise, respectively, the ICG de-noising signal and ECG de-noising signal that can be used for feature point recognition are output.

4. The ECG synchronized feature identification algorithm for four-electrode ICG impedance measurement of claim 1, wherein, The impedance change graph and impedance electrocardiogram are drawn based on the obtained ICG de-noising signal and ECG de-noising signal, and the main feature points of ICG and ECG in the same cycle are identified and the feature parameters are extracted based on the drawn impedance change graph and impedance electrocardiogram, including: S310, according to the sampling rate setting and clinical suggestion, the ECG denoising signal is divided into data blocks with 500 data as a buffer, and a plurality of denoising signal data blocks are obtained, and an impedance electrocardiogram is drawn; S320, using ECG peak recognition algorithm, to each de-noise signal data block is processed, looking for R peak feature point, extracting the time domain signal in and ; S330, for multiple Align the first R peak of the denoised signal data block according to Move to the left The minimum interval points are used as the starting point of the heartbeat cycle. If the start of the data block does not meet the requirements The minimum interval points are then calculated from the second... Move to the left Minimum interval points as At the same time, the next Decrease by 1 as the end point of the cycle Repeat this operation and divide the ECG period data; S340, mark the corresponding peaks in the ECG cycle data, find and mark the peaks on the left and right sides of the peaks respectively using a trough identification algorithm; peaks. S350, Extract all from the ECG denoised signal , , The idx and val of the feature points are stored in a global variable structure and marked on the impedance electrocardiogram for use in the synchronous calculation of the ICG signal.

5. The ECG synchronized feature recognition algorithm for four-electrode ICG impedance measurements of claim 4, wherein, The impedance change graph and impedance electrocardiogram are drawn based on the obtained ICG de-noising signal and ECG de-noising signal, and the main feature points of ICG and ECG in the same cycle are identified and the feature parameters are extracted based on the drawn impedance change graph and impedance electrocardiogram, including: S360, input ICG de-noise signal, divide data block according to sampling rate synchronized with ECG, get de-noise data block, establish global coordinate index system, ensure ICG data block coordinate and maintain specific timing alignment, draw impedance change graph; S370、adopt three points and five points center difference method to the denoising data block differential calculation, get Data block, and draw impedance differential diagram; Data block, and draw impedance differential diagram; S380, Traversal The data block uses the ICG peak identification algorithm to identify the maximum point in the impedance differential plot and determines it as... Points, and extract and ; S390, in the global coordinate system, and according to the size of the ECG cycle data block, the ICG data is periodically divided synchronously, and all points are marked in the impedance differential diagram; S3100, according to the extracted Looking left, determine Point position and extract ; S3110、According to The minimum point in the impedance differential graph is found to the right, and is determined as The point is extracted ; The main feature points in the ICG cycle data are synchronized and identified. , , Synchronization identification of main feature points.

6. The ECG synchronized feature recognition algorithm for four-electrode ICG impedance measurements of claim 5, wherein, The physiological parameter calculation of the extracted main feature points of ICG and ECG includes: S410, calling the identified , , and , , idx and val data of the feature points as the basic data for physiological parameter calculation; S420, identified from ECG cycle data Peaks, extracting adjacent peaks within the same period The peak idx difference is used to calculate the heart rate; S430, according to the ICG_period data identified 、 、 point, extract and time domain difference value, calculate the stroke volume; S440, the heart rate and stroke volume are calculated by calculation, and the cardiac output is calculated; S450, based on the calculated heart rate, stroke volume, cardiac output and feature parameters in the global variable structure, the parameter calculation of cardiac output index, stroke index and vascular resistance is further completed; S460, based on the parameters calculated in S450, the key cardiac function quantitative index is output, so as to provide data support for accurate evaluation of cardiac function.

7. The ECG-synchronized feature recognition algorithm for four-electrode ICG impedance measurements of claim 6, characterized in that, In step S420, the calculation formula of the heart rate is as follows: wherein HR is the number of heartbeats per minute, i.e. the heart rate, is the index position of the ith R-peak in the data block, and the sampling rate is the number of data points collected per unit of time.

8. The ECG synchronization feature identification algorithm for four electrode ICG impedance measurement according to claim 6, characterized in that, In step S430, the calculation formula of the stroke volume is as follows where p is the blood conductivity, L is the distance between the two three electrodes, Z0 is the base impedance, MAX is the amplitude value of the impedance in the corresponding period of the impedance differential graph 3. LVET is the left ventricular ejection interval. wave amplitude. LVET is the left ventricular ejection interval.

9. An electronic device, comprising: The electronic device includes: At least one processor; and The memory is in communication connection with the at least one processor; wherein The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the ECG synchronous feature recognition algorithm of the four-electrode ICG impedance measurement of any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions for causing a computer to execute the ECG synchronous feature recognition algorithm of the four-electrode ICG impedance measurement of any one of claims 1-8.

Images