Heart sound analysis methods and electronic equipment

Abstract

This application discloses a method and electronic device for analyzing heart sounds, relating to the field of biomedical signal processing technology. The method includes: acquiring a target electrocardiogram (ECG) signal, a target heart sound signal, and a target pulse signal; determining a first heart sound and a second heart sound from the target heart sound signal based on a first time point corresponding to the R-wave peak point in the target ECG signal and a second time point corresponding to the trough point in the target pulse signal, and determining the first heart sound envelope peak value and the second heart sound envelope peak value; identifying three valid heart sound peak points in the first heart sound based on the first heart sound envelope peak value, and identifying two valid heart sound peak points in the second heart sound based on the second heart sound envelope peak value; and performing heart sound analysis based on the five valid heart sound peak points to obtain the heart sound analysis result. This application can more accurately resolve heart sound signals in low signal-to-noise ratio environments, thereby improving the practicality and reliability of heart sound signals in health monitoring and clinical decision support.

Description

[0001] Declaration regarding priority reference: This application claims priority to the patent filed on September 15, 2025, with application number 202511316987.2 and application title "Method and Electronic Device for Heart Sound Analysis", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of biomedical signal processing technology, and in particular to methods and electronic devices for heart sound analysis. Background Technology

[0003] In recent years, with the rapid popularization of wearable medical devices and remote health monitoring technologies, phonocardiogram (PCG) signals have gradually become an important data source in fields such as cuffless blood pressure estimation, cardiac function assessment, and arrhythmia analysis due to their advantages such as high physiological relevance, low energy consumption for acquisition, and strong non-invasiveness. In the composition of heart sounds, the first heart sound (S1) and the second heart sound (S2), as landmark events, correspond to the mechanical vibrations caused by valve closure during the initial stages of cardiac systole and diastole, respectively. They have clear physiological significance and relatively stable temporal positions, and are therefore often used as benchmarks for heart sound segmentation, feature extraction, and time-series analysis.

[0004] Current cardiac sound analysis techniques largely rely on traditional methods such as envelope extraction, energy thresholding, or template matching, focusing only on the "main peak" or global extremum of S1 / S2, while neglecting the physiological information contained in its internal multi-peak structure. These methods fail to effectively distinguish the origins and temporal relationships of different vibrational components, leading to significant limitations in feature modeling, dynamic tracking, and pathological reasoning. Furthermore, in practical applications, cardiac sound signals acquired through wearable devices often suffer from low signal-to-noise ratios, motion artifacts, environmental noise interference, and sensor positional variations, further increasing the difficulty of accurately identifying and segmenting the S1 / S2 components. Existing peak detection algorithms perform poorly in such real-world scenarios, easily resulting in false positives, false negatives, or location drift, severely limiting the effectiveness of cardiac sound analysis in continuous monitoring and early diagnosis.

[0005] Therefore, how to accurately analyze heart sound signals in low signal-to-noise ratio environments to improve the practicality and reliability of heart sound signals in health monitoring and clinical decision support has become an urgent technical problem to be solved. Summary of the Invention

[0006] The main purpose of this application is to provide a method and electronic device for analyzing heart sounds, aiming to address the technical problem of how to more accurately resolve heart sound signals in low signal-to-noise ratio environments, so as to improve the practicality and reliability of heart sound signals in health monitoring and clinical decision support.

[0007] To achieve the above objectives, this application provides a method for analyzing heart sounds, comprising: Acquire the target physiological signals of the target heartbeat, wherein the target physiological signals include the target electrocardiogram signal, the target heart sound signal, and the target pulse signal; Identify the peak point of the R wave in the target electrocardiogram signal and take the time corresponding to the peak point as the first time point; identify the trough point of the target pulse signal and take the time corresponding to the trough point as the second time point; Based on the first time point and the second time point, the target heart sound signal is segmented to obtain the first heart sound and the second heart sound, and the first heart sound envelope peak value corresponding to the first heart sound and the second heart sound envelope peak value corresponding to the second heart sound are determined. Based on the first heart sound envelope peak value, a first threshold is determined, and based on the second heart sound envelope peak value, a second threshold is determined; based on the first threshold, three valid heart sound peak points in the first heart sound are identified, and based on the second threshold, two valid heart sound peak points in the second heart sound are identified; Heart sound analysis is performed based on the three effective heart sound peak points in the first heart sound and the two effective heart sound peak points in the second heart sound to obtain the heart sound analysis results.

[0008] In addition, to achieve the above objectives, this application also provides an electronic device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the heart sound analysis method as described above.

[0009] In addition, to achieve the above objectives, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the heart sound analysis method as described above.

[0010] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the heart sound analysis method as described above.

[0011] The applicant, through significant time and resources, conducted extensive experiments and research. The study revealed that S1 (the first heart sound, occurring at the onset of ventricular systole) is primarily generated by vibrations caused by mitral and tricuspid valve closure and ventricular myocardial contraction (corresponding to the three effective peak points of the first heart sound mentioned above). It often presents as multiple transient mechanical vibration peaks, reflecting multiple coupled structural responses during the rise of left ventricular pressure. S1 is highly consistent with pre-ejection phase, ventricular pressure changes, and active left ventricular contraction, serving as an important reference for deducing pulse transit time (PTT) and cardiac output. S2 (the second heart sound, occurring at the onset of ventricular diastole) is primarily generated by vibrations caused by aortic and pulmonary valve closure (corresponding to the two effective peak points of the second heart sound mentioned above). Its morphology often exhibits slight splitting due to the pressure difference between the left and right atria and is influenced by respiratory regulation (e.g., increased splitting of S2 during inspiration).

[0012] This application provides a method and electronic device for analyzing heart sounds. The technical solution involves acquiring a target physiological signal of a target heartbeat, including a target electrocardiogram (ECG) signal, a target heart sound signal, and a target pulse signal. The method identifies the R-wave peak point of the target ECG signal, using the time corresponding to the R-wave peak point as a first time point. It also identifies the trough point of the target pulse signal, using the time corresponding to the trough point as a second time point. Based on the first and second time points, the target heart sound signal is segmented to obtain a first heart sound and a second heart sound. The method determines the first heart sound envelope peak value corresponding to the first heart sound and the second heart sound envelope peak value corresponding to the second heart sound. Then, based on the first heart sound envelope peak value, a first threshold is determined, and based on the second heart sound envelope peak value, a second threshold is determined. Finally, the method identifies three valid heart sound peak points in the first heart sound according to the first threshold and identifies the third valid heart sound peak point according to the second threshold. The two effective heart sound peak points in the two heart sounds are then used to perform heart sound analysis based on the three effective heart sound peak points in the first heart sound and the two effective heart sound peak points in the second heart sound. This results in robust identification of multiple effective S1 and S2 peaks in each heart sound cycle. Compared to related technologies that only extract the "main peak" or maximum amplitude point of S1 and S2, this embodiment can analyze the complex internal signal structure of S1 and S2, and integrate the physiological correspondence and temporal distribution characteristics between different peaks in S1 and S2. This provides more accurate and valuable information in modeling, tracking, and clinical reasoning, and provides a reliable basis for subsequent PTT measurement, cardiac output estimation, and cuffless blood pressure prediction. This enables the embodiment to more accurately resolve heart sound signals in low signal-to-noise ratio environments, thereby improving the practicality and reliability of heart sound signals in health monitoring and clinical decision support. Attached Figure Description

[0013] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0014] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0015] Figure 1 This is a flowchart illustrating the first embodiment of the heart sound analysis method of this application; Figure 2 This is a flowchart illustrating the second embodiment of the heart sound analysis method of this application; Figure 3 This is a flowchart illustrating the third embodiment of the heart sound analysis method of this application; Figure 4 This is a schematic diagram of the process for identifying valid heart sound peaks in a specific embodiment of this application; Figure 5 This is a schematic diagram of the electronic device structure of the hardware operating environment involved in the heart sound analysis method in the embodiments of this application.

[0016] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0017] 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 application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0019] Currently, heart sound signals, as important physiological information reflecting cardiac mechanical activity, have significant application value in cardiovascular health monitoring, non-invasive blood pressure estimation, cardiac output assessment, and early heart disease screening. Among them, the temporal characteristics of the first heart sound (S1) and the second heart sound (S2), especially their internal multi-peak structure, contain deep physiological information such as valve closure dynamics, hemodynamic status, and myocardial contraction coordination.

[0020] However, existing related technologies generally have the following drawbacks in heart sound analysis: Coarse-grained feature extraction: Most methods only focus on the "main peak" or global maximum point of S1 / S2, ignoring the temporal distribution and amplitude variation of its internal multi-peak structure. This results in the inability to fully explore the physiological details of the heart sound signal, limiting its ability to distinguish pathological conditions (such as valvular regurgitation and conduction block). Weak anti-interference capability: Heart sound signals collected by wearable devices are often affected by factors such as low signal-to-noise ratio, motion artifacts, environmental noise and unstable sensor coupling. Traditional fixed threshold or template matching methods are prone to false detection, false negative detection or peak location drift, and lack robustness. Disconnect between segmentation and recognition: Existing technologies mostly adopt isolated signal segmentation and peak detection processes, lacking a time-guided mechanism based on multimodal physiological signals (such as ECG and pulse), which leads to inaccurate S1 / S2 boundary division, thereby affecting the accuracy of subsequent peak recognition.

[0021] Therefore, how to achieve robust, accurate, and interpretable identification and analysis of the multi-peak structure within S1 / S2 in a low signal-to-noise ratio environment has become a key challenge in improving the practicality of heart sound signals in continuous health monitoring and clinical decision support.

[0022] The main solution of this application embodiment is as follows: Acquire the target physiological signal of the target heartbeat, wherein the target physiological signal includes a target electrocardiogram (ECG) signal, a target heart sound signal, and a target pulse signal; identify the R-wave peak point of the target ECG signal and take the time corresponding to the R-wave peak point as a first time point; identify the trough point of the target pulse signal and take the time corresponding to the trough point as a second time point; based on the first time point and the second time point, segment the target heart sound signal to obtain a first heart sound and a second heart sound, and determine the first heart sound envelope peak value corresponding to the first heart sound and the second heart sound envelope peak value corresponding to the second heart sound; based on the first heart sound envelope peak value, determine a first threshold and based on the second heart sound envelope peak value, determine a second threshold; identify three valid heart sound peak points in the first heart sound according to the first threshold and two valid heart sound peak points in the second heart sound according to the second threshold; perform heart sound analysis based on the three valid heart sound peak points in the first heart sound and the two valid heart sound peak points in the second heart sound to obtain heart sound analysis results.

[0023] Compared to related technologies that only extract the "main peak" or maximum amplitude point of S1 and S2, the embodiments of this application can robustly identify multiple effective S1 and S2 peaks in each heart sound cycle, analyze the complex internal signal structure of S1 and S2, and fuse the physiological correspondence and temporal distribution characteristics between different peaks in S1 and S2. This provides more accurate and valuable information in modeling, tracking, and clinical reasoning, and provides a reliable foundation for subsequent PTT measurement, cardiac output estimation, and cuffless blood pressure prediction. In this way, the embodiments of this application can more accurately resolve heart sound signals in low signal-to-noise ratio environments, thereby improving the practicality and reliability of heart sound signals in health monitoring and clinical decision support.

[0024] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0025] This application proposes a heart sound analysis method according to a first embodiment.

[0026] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the heart sound analysis method of this application.

[0027] In this embodiment, the heart sound analysis method includes steps S100~S500: Step S100: Obtain the target physiological signal of the target heartbeat, wherein the target physiological signal includes the target electrocardiogram signal, the target heart sound signal and the target pulse signal; As those skilled in the art will know, a cardiac cycle refers to the complete mechanical activity cycle of the heart completing one contraction and relaxation. It is usually marked by the R wave or the trough of the pulse wave and includes the ventricular systolic and diastolic phases. It is the basic time unit for analyzing physiological signals such as heart sounds, electrocardiograms, and pulses.

[0028] Physiological signals are measurable physical quantities that reflect the physiological state or functional activities of an organism. Common types include electrical signals (such as electrocardiogram and electroencephalogram), acoustic signals (such as heart sounds and breath sounds), and mechanical signals (such as pulse and blood pressure), which are used for non-invasive monitoring of human health status.

[0029] Among them, the electrocardiogram (ECG) signal refers to the cardiac electrical activity signal collected by sensors such as ECG surface electrodes, which reflects the depolarization and repolarization process of the myocardium. Its typical waveforms include the P wave, QRS complex (including R wave), and T wave. The R wave is the positive wave with the highest amplitude in the QRS complex and is often used to calibrate the onset time of cardiac electrical activity.

[0030] Phonocardiogram (PCG) refers to the acoustic signal collected by sensors such as microphones and accelerometers, which is generated by mechanical vibrations such as the closure of heart valves and blood flow turbulence. It mainly includes the first heart sound (S1, corresponding to the closure of the mitral and tricuspid valves) and the second heart sound (S2, corresponding to the closure of the aortic and pulmonary valves). Its internal multi-peak structure contains information on valvular dynamics and cardiac function status.

[0031] Pulse waveform (PPG) refers to the peripheral arterial pulsation signal collected by photoplethysmography (PPG), pressure sensors, etc. It reflects the changes in blood volume caused by the heart pumping blood. Its periodic trough points usually correspond to the onset of ventricular diastole and can be used to calibrate the end of pulse wave conduction time (PTT).

[0032] It should be noted that, in this embodiment, the target heartbeat refers to a complete heartbeat cycle to be analyzed, typically starting from one R-wave peak and ending before the next R-wave peak. The target physiological signal refers to the multimodal physiological signals synchronously acquired within the target heartbeat, including the target electrocardiogram signal, the target heart sound signal, and the target pulse signal. These three signals are strictly aligned in time to ensure physiological consistency in subsequent joint analysis based on temporal relationships.

[0033] In one feasible implementation, the step of acquiring the target physiological signal of the target heartbeat in step S100 above may include steps S110 to S130: Step S110: Synchronously acquire raw physiological signals, including raw electrocardiogram signals, raw pulse signals, and raw heart sound signals; It should be noted that raw physiological signals refer to physiological signals in analog or digital form that are directly acquired by sensors without any preprocessing. They retain all time and frequency domain information of the physiological signals, but also contain non-physiological components such as environmental noise, motion artifacts, and power supply interference. They are the basic input for all subsequent signal processing and feature extraction.

[0034] In this embodiment, the raw physiological signals include raw electrocardiogram signals, raw pulse signals, and raw heart sound signals.

[0035] Among them, the raw electrocardiogram signal is the raw voltage waveform of the heart's electrical activity directly acquired by sensors such as ECG surface electrodes. It is usually a time-varying electrical signal in the range of microvolts (μV) to millivolts (mV). Its core feature is the periodic QRS complex, especially the R wave peak point, which has a high signal-to-noise ratio and strong detectability.

[0036] The raw pulse signal is a peripheral arterial pulsation signal caused by changes in peripheral arterial blood volume, which is directly collected by PPG sensors, pressure sensors, etc. Its typical waveform has a periodic rising edge (ejection phase) and falling edge (diastole phase), and the trough point corresponds to the moment when the vascular pressure is the lowest. It can be used to calibrate the end point of pulse wave propagation time.

[0037] The raw heart sound signal is an acoustic or vibration signal collected by sensors such as microphones, piezoelectric sensors, and accelerometers, which is caused by mechanical vibrations such as heart valve closure and blood flow impact. The frequency range is mainly concentrated in 20–500 Hz, and the energy is concentrated in the S1 and S2 intervals. However, it is easily affected by breathing, environmental noise and motion interference, and the signal-to-noise ratio is low.

[0038] In this embodiment, the three types of raw physiological signals can be synchronously acquired through a multimodal sensor array (including ECG surface electrodes, PPG sensors, accelerometers, etc.) to ensure that each physiological signal is strictly aligned on the time axis. For example, in wearable bracelets or smart clothing devices, ECG electrodes, PPG optical modules, and PCG vibration sensors are integrated to achieve hardware-level synchronous acquisition of the three types of raw physiological signals: ECG, PPG, and PCG. The sampling rates are typically set to ≥250Hz (ECG), ≥100Hz (PPG), and ≥1kHz (PCG) to meet the requirements for accurate capture of high-frequency heart sound components.

[0039] This implementation method constructs a multimodal physiological signal data foundation for single heartbeat and heart sound analysis by simultaneously acquiring three raw physiological signals. This ensures the physiological consistency of subsequent cross-modal temporal correlations (such as R-wave peak point → S1 → pulse signal trough point), which is a prerequisite for achieving high-precision joint analysis. Compared with time-division acquisition or asynchronous recording by multiple devices, this method significantly reduces PTT calculation errors and heart sound localization deviations caused by time offsets, and is particularly suitable for ensuring long-term stability in dynamic monitoring scenarios.

[0040] Step S120: Preprocess the original physiological signal to obtain the preprocessed physiological signal. The preprocessing includes at least one of filtering, normalization and noise reduction. The preprocessed physiological signal includes the preprocessed electrocardiogram signal, the preprocessed pulse signal and the preprocessed heart sound signal. Those skilled in the art will recognize that filtering refers to using digital filters to remove unwanted components in specific frequency bands from the original signal. Common types include low-pass filtering, high-pass filtering, band-pass filtering, and power frequency filtering. Normalization refers to unifying the amplitudes of signals from different channels or individual signals to the same dimension or range. Common methods include min-max normalization and Z-score normalization. Noise reduction refers to using signal enhancement algorithms to suppress non-physiological noise and improve the signal-to-noise ratio. Common techniques include wavelet thresholding, empirical mode decomposition, variational mode decomposition, and adaptive filtering.

[0041] It should be noted that, in this embodiment, the preprocessed physiological signal refers to a high-quality physiological signal optimized by one or more of the above preprocessing methods. Its signal-to-noise ratio is significantly improved, the baseline is more stable, and the characteristic waveform is clearer, making it suitable for subsequent peak detection, envelope extraction, and multi-peak identification.

[0042] This implementation significantly improves the quality and usability of physiological signals across various modalities by systematically preprocessing the raw physiological signals. This process not only enhances the detectability of key feature points (such as R-wave peaks, pulse signal troughs, and S1 / S2 main peaks) but also reduces the risk of false triggering and missed identification due to noise interference, serving as the first line of defense for ensuring the overall algorithm's robustness. The preprocessing stage is particularly crucial in complex usage scenarios common to wearable devices, such as movement, sweating, and loosening.

[0043] Step S130: Based on the preprocessed physiological signals, obtain the target physiological signal of the target heartbeat.

[0044] This embodiment extracts a signal segment from the continuously acquired preprocessed physiological signal stream, within the complete cardiac cycle of the target heartbeat, and uses it as the target heart sound signal for subsequent heart sound analysis.

[0045] This implementation method systematically completes the conversion from raw physiological signals to target heartbeat physiological signals through a three-stage process of "synchronous acquisition → multidimensional preprocessing → periodic slicing". This process not only ensures the temporal consistency and quality reliability of multimodal signals, but also constructs standardized data units for single cardiac cycle analysis, laying a solid data foundation for subsequent accurate heart sound analysis based on electro-acoustic-mechanical multi-anchor points. It is an indispensable preliminary step for achieving high-precision and highly robust heart sound analysis.

[0046] This embodiment acquires multimodal target physiological signals synchronously within the target heartbeat and constructs an ECG-PCG-PPG joint analysis framework. This provides a data foundation for subsequent precise heart sound segmentation and peak identification based on electro-acoustic-mechanical time-series anchors, enabling the extraction of analysis objects with clear physiological context (i.e., effective heart sound peak points) from the target physiological signals. This effectively avoids misjudgments caused by signal asynchrony or segment misalignment, and improves the reliability and repeatability of the overall heart sound analysis.

[0047] Step S200: Identify the peak point of the R wave in the target ECG signal and take the time corresponding to the peak point of the R wave as the first time point; identify the trough point of the target pulse signal and take the time corresponding to the trough point as the second time point; As those skilled in the art will know, a peak point is the point in time when the local amplitude of a signal waveform reaches its maximum value. It usually corresponds to the extreme value of the intensity of a certain physiological event, such as the R wave of an electrocardiogram, the main peak of a heart sound, and the peak of a pulse wave.

[0048] A trough point is a point in a signal waveform where the amplitude reaches its minimum. It typically corresponds to a low point in a physiological process, such as the diastolic minimum of a pulse wave or the oscillation interval in a heart sound signal. Trough points, along with peak points, constitute the periodic characteristics of a signal and can be used to calibrate sequential events and calculate physiological parameters.

[0049] In this embodiment, the R-wave peak point refers to the highest positive extreme value of the QRS complex amplitude in the target ECG signal, representing the onset of ventricular depolarization. It has a high signal-to-noise ratio and strong detectability, and is the standard starting marker of the cardiac cycle. The trough point of the target pulse signal refers to the point at which the waveform of the target pulse signal drops to its lowest amplitude within a cardiac cycle, usually occurring in the early diastolic phase of the ventricle, corresponding to the lowest arterial pressure. The time interval between the R-wave peak point and the trough point, that is, the time interval between the first time point (T1) and the second time point (T2), is the pulse wave conduction time (PTT), an important indicator for assessing vascular elasticity and blood pressure status.

[0050] This embodiment identifies the R-wave peak point of ECG and the trough point of PPG as dual-modal time-series anchor points, constructing an "electro-mechanical" dual-end reference system to provide a high-precision time boundary for subsequent accurate segmentation of heart sound signals. Compared to methods that rely solely on a single signal (such as ECG alone) for segmentation, this embodiment significantly improves the physiological accuracy of S1 / S2 interval division, especially maintaining stable positioning under motion or low signal-to-noise ratio scenarios, thus enhancing the robustness of the algorithm.

[0051] Step S300: Based on the first time point and the second time point, the target heart sound signal is segmented to obtain the first heart sound and the second heart sound, and the first heart sound envelope peak value corresponding to the first heart sound and the second heart sound envelope peak value corresponding to the second heart sound are determined. It should be noted that, in this embodiment, the first heart sound (S1) refers to the sound produced by the vibrations caused by the closure of the mitral and tricuspid valves and the contraction of the ventricular myocardium. It occurs in the early stage of ventricular systole, usually after the peak of the R wave and before the trough of the pulse signal, and its signal energy is concentrated in the 30–100 Hz frequency range. The second heart sound (S2) refers to the sound produced by the vibrations caused by the closure of the aortic and pulmonary valves. It occurs in the early stage of ventricular diastole, usually after the trough of the pulse signal, and its frequency is slightly higher than S1.

[0052] In one feasible implementation, step S300 above, which involves segmenting the target heart sound signal based on a first time point and a second time point to obtain a first heart sound and a second heart sound, may include steps S310 to S320: Step S310: The target heart sound signal between the first time point and the second time point is taken as the first heart sound; It should be noted that in this embodiment, the first time point is the time corresponding to the peak point of the R wave (T1), which represents the start of cardiac electrical activity, and the second time point is the time corresponding to the trough point of the pulse signal (T2), which represents the lowest point of peripheral arterial volume / pressure, and is usually synchronized with the hemodynamic response after the aortic valve closes.

[0053] In this embodiment, the time interval between T1 and T2 is defined as the S1 interval, and the heart sound signal segment of the S1 interval is defined as the first heart sound (S1). This conforms to the physiological timing logic that S1 occurs in the early stage of ventricular systole and can accurately cover the main energy range of S1, which includes the three stages of mitral valve closure, tricuspid valve closure and ventricular myocardial contraction.

[0054] This implementation method directly extracts the S1 signal segment using T1 and T2 as boundaries, achieving precise segmentation based on multimodal physiological time-series anchor points. This avoids the problems of S1 truncation or S2 component mixing caused by traditional fixed-delay methods (such as 100ms after the R wave) when heart rate variability is large. This method fully utilizes the high robustness of ECG and PPG features, improving the physiological consistency and stability of S1 interval division, and is particularly suitable for the continuous monitoring needs of wearable devices in dynamic scenarios.

[0055] And in step S320, the target heart sound signal between the second time point and the third time point is taken as the second heart sound; The third time point is the time point that continues for a preset duration after the second time point. The preset duration is determined based on the product of the duration of the cardiac cycle and a preset coefficient factor, with the preset coefficient factor ranging from 0.35 to 0.45.

[0056] In this embodiment, the time interval between the second time point T2 and the third time point T3 is defined as the S2 interval, and the heart sound signal segment of the S2 interval is defined as the second heart sound (S2). This is consistent with the physiological timing logic that S2 occurs in the early diastolic phase of the ventricle and is usually located near the trough of the pulse wave. It can accurately cover the main energy range of S2, which includes the two stages of aortic valve closure and pulmonary valve closure.

[0057] In this embodiment, the preset coefficient factor is a proportional parameter used to determine the duration of S2 (i.e., the length of the S2 interval, T3-T2). It is essentially a dimensionless empirical weighting coefficient used to proportionally map the total duration of the current cardiac cycle (i.e., the duration of the heartbeat) to the expected duration of the S2 component.

[0058] From a physiological perspective, the second heart sound (S2) occurs during the early diastolic phase of the ventricle, primarily generated by the closure of the aortic and pulmonary valves. Its duration is closely related to the timing of cardiac mechanical activity. Because changes in heart rate significantly affect the time allocation of different cardiac phases (e.g., a faster heart rate leads to a more pronounced shortening of the systolic phase and an overall compression of the diastolic phase), the duration of S2 is not fixed but dynamically adjusts with the cardiac cycle. Therefore, using a fixed duration window (e.g., a uniform 300ms) to segment S2 can easily lead to signal truncation or aliasing when the heart rate is too fast or too slow.

[0059] To address this, this implementation introduces a preset coefficient factor k, and calculates the duration ΔT of S2 using the formula ΔT = VVI × k, where VVI is the trough interval of the current heartbeat (i.e., the duration of the cardiac cycle, which is also equal to the duration of the heartbeat, specifically the time interval between the PPG trough point in the current heartbeat and the PPG trough point in the next heartbeat). This design ensures that the duration of S2 is linearly correlated with the cardiac cycle, achieving adaptive segmentation.

[0060] In this embodiment, the preset coefficient factor k ranges from 0.35 to 0.45, a range derived from extensive clinical data statistics and algorithm verification. In practical applications, users can flexibly set it according to their actual needs, or dynamically determine it based on the duration of the cardiac cycle and the duration of S1 (i.e., the length of the S1 interval, T2-T1). It is easy to understand that for cardiac cycles of the same duration, there is a certain proportional relationship between the duration of S1 and the duration of S2. This proportional relationship differs for cardiac cycles of different durations, and can be specifically determined through extensive experiments.

[0061] This implementation defines the S2 interval by introducing a dynamic time window proportional to the cardiac cycle, achieving adaptability in S2 segmentation. Compared to a fixed-duration window (e.g., a fixed 300ms), this strategy effectively addresses fluctuations in the duration of S2 caused by changes in the cardiac cycle, preventing the loss of S2 components due to an excessively short window or the introduction of interference from the next heartbeat due to an excessively long window. By using T2 as the physiological starting point, this implementation significantly improves segmentation accuracy and algorithm robustness while ensuring the integrity of S2, making it particularly suitable for monitoring scenarios involving significant cardiac cycle fluctuations, such as motion or pathological conditions.

[0062] In summary, this implementation method constructs a physiologically consistent and highly adaptive heart sound component segmentation mechanism through a two-stage segmentation strategy of "T1-T2 interval definition S1" and "T2-T3 dynamic interval definition S2". This mechanism fully utilizes the high-precision temporal features of ECG and PPG, combined with adaptive heart rate parameter adjustment, significantly improving the accuracy and generalization ability of S1 / S2 segmentation. This provides a high-quality input foundation for subsequent envelope extraction, peak identification, and heart sound analysis, and is a key supporting link for achieving highly robust heart sound analysis.

[0063] In this embodiment, the first heart sound envelope peak value refers to the absolute value of the amplitude of the highest point in the heart sound envelope signal in the S1 interval, and the second heart sound envelope peak value refers to the absolute value of the amplitude of the highest point in the heart sound envelope signal in the S2 interval.

[0064] The upper envelope of the phonocardiogram (Upper PCG Envelope) is an amplitude envelope curve extracted using signal processing techniques, reflecting the time-varying characteristics of the positive vibrational energy of the phonocardiogram signal. It is formed by connecting the local positive maxima of the target phonocardiogram signal and characterizes the maximum positive intensity of the phonocardiogram at each moment. It is mainly used to identify the energy distribution, start and end times, and peak positions of major phonocardiogram components such as S1 and S2. The lower envelope of the phonocardiogram (Lower PCG Envelope) is an amplitude envelope curve extracted using signal processing techniques, reflecting the time-varying characteristics of the negative vibrational energy of the phonocardiogram signal. It is formed by connecting the local negative minima of the target phonocardiogram signal and characterizes the maximum negative intensity of the phonocardiogram at each moment. It is mainly used to characterize the quiet period between S1 and S2, the vibration decay process, and the signal floor fluctuations.

[0065] In this embodiment, the upper and lower envelope signals of the heart sounds mentioned above can be extracted from the target heart sound signal using envelope detection methods such as Hilbert transform, wavelet envelope, and RMS (Root Mean Square) sliding window. Among them, Hilbert transform is a signal processing technique that can be used to extract the analytic signal of a real signal and then obtain its instantaneous amplitude envelope.

[0066] In one feasible implementation, the steps in step S300 above, which determine the first heart sound envelope peak value corresponding to the first heart sound and the second heart sound envelope peak value corresponding to the second heart sound, may include steps S330 to S350: Step S330: Perform Hilbert transform on the target heart sound signal to obtain the heart sound envelope signal; It should be noted that, in this embodiment, the heart sound envelope signal (PCG Envelope) refers to the amplitude variation curve of the heart sound signal extracted by Hilbert transform, which includes two parts: the upper heart sound envelope signal and the lower heart sound envelope signal.

[0067] This implementation obtains the heart sound envelope signal through Hilbert transform, which effectively improves the recognizability of the main energy peaks of S1 / S2 heart sounds. In particular, it can still stably capture the start and end and intensity changes of heart sound components under low signal-to-noise ratio conditions, providing a highly robust basic signal for subsequent envelope peak extraction and dynamic threshold generation.

[0068] Step S340: The peak value of the heart sound envelope signal between the first time point and the second time point is taken as the first heart sound envelope peak value; As those skilled in the art will know, the envelope peak refers to the local maximum value of the envelope signal within a certain time interval.

[0069] In this embodiment, the heart sound peak value refers to the peak value of the heart sound envelope signal within a certain time interval, representing the moment when the heart sound energy is strongest within that time interval.

[0070] It is easy to understand that, since the upper envelope signal of the heart sound in the heart sound envelope signal is formed by connecting the local positive maxima of the target heart sound signal, the envelope peak of the heart sound envelope signal in a certain time interval, that is, the heart sound peak, actually belongs to the upper envelope signal of the heart sound.

[0071] This implementation searches for the peak value of the heart sound envelope signal (or the heart sound envelope signal) within the S1 interval between the first time point T1 and the second time point T2, and uses it as the first heart sound envelope peak value As1env_max corresponding to the first heart sound. This avoids the peak selection problem caused by fixed time windows or missegmentation. As1env_max not only reflects the true energy level of S1, but also has the ability to resist noise interference (because the envelope has smoothed the fluctuations of the original signal), providing a stable and reliable reference benchmark for the subsequent adaptive threshold mechanism.

[0072] And in step S350, the peak value between the second time point and the third time point is taken as the second heart sound envelope peak value; The third time point is the time point that continues for a preset duration after the second time point. The preset duration is determined based on the product of the duration of the cardiac cycle and a preset coefficient factor, with the preset coefficient factor ranging from 0.35 to 0.45.

[0073] This implementation method searches for the heart sound peak value of the heart sound envelope signal (or the heart sound upper envelope signal) within the S2 interval between the second time point T2 and the third time point T3, and uses it as the second heart sound envelope peak value As2env_max corresponding to the second heart sound. This achieves accurate quantification of the maximum intensity of S2 energy, overcomes the problem of peak omission or misinclusion caused by heart rate changes in a fixed time window, and ensures that As2env_max always represents the true main peak energy of S2, providing a physiologically consistent and robust input basis for the subsequent generation of the second threshold th2.

[0074] In summary, this implementation method systematically achieves accurate identification of the first and second heart sound envelope peaks through a process of "Hilbert transform → interval search → dynamic window extraction". This method integrates signal energy enhancement, physiological timing constraints, and heart rate adaptive mechanisms, significantly improving the accuracy and environmental adaptability of envelope peak extraction. It is a key preliminary step in constructing a dynamic threshold system and achieving robust multi-peak identification, effectively supporting the stable operation of the overall heart sound analysis scheme in low signal-to-noise ratio wearable scenarios.

[0075] Step S400: Determine a first threshold based on the first heart sound envelope peak value, and determine a second threshold based on the second heart sound envelope peak value; identify three valid heart sound peak points in the first heart sound according to the first threshold value, and identify two valid heart sound peak points in the second heart sound according to the second threshold value; It should be noted that the effective heart sound peak point refers to the positive local maximum point in S1 and S2 that represents a vibrational event with clear physiological significance, such as mitral valve closure, tricuspid valve closure, and aortic valve opening.

[0076] In this embodiment, the first threshold (Threshold 1, th1) is the amplitude judgment threshold used to identify the effective heart sound peak points in S1. Its value is dynamically generated based on the first heart sound envelope peak As1env_max to ensure that the multi-peak structure inside S1 can be stably extracted under different signal intensities.

[0077] Correspondingly, the second threshold (Threshold 2, th2) is the amplitude judgment threshold used to identify the effective heart sound peak points in S2. Its value is dynamically generated based on the second heart sound envelope peak As2env_max to ensure that the multi-peak structure inside S2 can be stably extracted under different signal intensities.

[0078] This embodiment constructs a dual threshold system that adaptively matches the energy levels of S1 / S2, thereby achieving robust recognition of the multi-peak structure within heart sounds. This avoids the problems of missed or false detections caused by traditional fixed thresholds in scenarios with individual differences or changes in signal-to-noise ratio, and provides high-quality feature input for subsequent structured analysis.

[0079] In one feasible implementation, the steps of determining the first threshold based on the first heart sound envelope peak value and determining the second threshold based on the second heart sound envelope peak value in step S400 may include steps S410-S420: Step S410: Adjust the peak value of the first heart sound envelope based on the first adjustment coefficient to obtain a first threshold, wherein the first threshold is less than the peak value of the first heart sound envelope; It should be noted that the first adjustment coefficient α is a dimensionless scaling factor between 0 and 1, used to proportionally attenuate the first heart sound envelope peak As1env_max to generate the first threshold th1.

[0080] For example: th1 = α × As1env_max; The initial value of α ranges from 0.15 to 0.25, preferably 0.2. This coefficient reflects the tolerance for the energy intensity of the "effective peak" in the first heart sound S1: too large a coefficient may lead to noise being misjudged as an effective peak; too small a coefficient may miss weak but physiologically significant secondary peaks.

[0081] The core of this strategy is to set a reasonable identification threshold downwards based on the absolute value of the total energy of S1 (i.e. the peak value of the first heart sound envelope), so as to ensure that the identified peak value is significantly different from the background noise and can also accommodate the amplitude differences caused by the different transmission paths of multiple vibration sources inside S1.

[0082] This implementation linearly scales As1env_max using a first adjustment coefficient, constructing a simple, efficient, and computationally inefficient dynamic thresholding mechanism. This method achieves stable adaptation across individuals and devices without requiring additional model training or complex parameter tuning, making it particularly suitable for real-time processing scenarios in resource-constrained wearable terminals.

[0083] And in step S420, the peak value of the second heart sound envelope is adjusted based on the second adjustment coefficient to obtain a second threshold, wherein the second threshold is less than the peak value of the second heart sound envelope.

[0084] It should be noted that the second adjustment coefficient β is a dimensionless scaling factor between 0 and 1, used to proportionally attenuate the peak value of the second heart sound envelope As2env_max to generate the second threshold th2.

[0085] For example: th2 = β × As2env_max; The initial value of β ranges from 0.15 to 0.25, preferably 0.2, and can be the same as or independently configured with α. This coefficient reflects the tolerance for the energy intensity of the "effective peak" in the second heart sound S2: too large a coefficient may lead to noise being misjudged as an effective peak; too small a coefficient may miss weak but physiologically significant secondary peaks.

[0086] Since S2 often splits due to fluctuations in respiration and blood pressure, its main peak may be weak. Therefore, the value of β can be dynamically adjusted according to the actual signal quality. For example, when the signal-to-noise ratio is low, β can be appropriately reduced to improve the recall rate.

[0087] The core of this strategy is to set a reasonable identification threshold downwards based on the absolute value of the overall maximum energy of S2 (i.e. the peak value of the second heart sound envelope), so as to ensure that the identified peak value is significantly different from the background noise and can also accommodate the amplitude differences caused by the different transmission paths of multiple vibration sources inside S2.

[0088] This implementation uses a second adjustment coefficient to scale As2env_max, achieving adaptive identification of effective peak values ​​within the S2 interval. Combined with a T2-T3 dynamic window, this mechanism can effectively capture S2 split peaks or delayed-closing components, improving sensitivity to valvular dysfunction and enhancing the clinical value of heart sound analysis results.

[0089] This implementation constructs a simple, efficient, and computationally inefficient dynamic thresholding mechanism. This method achieves stable adaptation across individuals and devices without requiring additional model training or complex parameter tuning, making it particularly suitable for real-time processing scenarios in resource-constrained wearable terminals.

[0090] It is worth mentioning that, in addition to the method described above of determining the recognition threshold (first threshold and second threshold) by preset adjustment coefficients (first adjustment coefficient and second adjustment coefficient), one of the following methods can also be used to determine the recognition threshold: The sliding adaptive thresholding method based on historical statistical mean maintains a recent multi-beat heart sound envelope peak sequence within a sliding time window (such as As1env_max and As2env_max of the past 10 heart beats), calculates the mean and standard deviation of As1env_max and As2env_max within the sliding time window, and sets the recognition threshold to the mean minus k times the standard deviation (k∈[0.5,1.5]), realizing long-term tracking and dynamic response of individualized heart sound intensity trends. It is suitable for signal drift scenarios caused by changes in body position and activity status in long-term monitoring.

[0091] A nonlinear threshold generation method based on machine learning models utilizes a pre-trained lightweight regression model (such as linear support vector regression, decision trees, or lightweight neural networks). Input features include at least the peak value of the first or second heart sound envelope, and may also include the current heart rate, the duration of the current cardiac cycle, and an estimated signal-to-noise ratio. The output is the optimal suggested value for the threshold. This method can integrate multi-dimensional contextual information to achieve more intelligent threshold decision-making and is suitable for advanced health analysis devices.

[0092] Although the above alternatives are implemented in different ways, they all follow the core idea of ​​"taking the heart sound energy benchmark as the core and dynamically generating the recognition threshold".

[0093] In summary, this embodiment constructs a dynamic threshold system based on As1env_max and As2env_max, and combines various implementation paths such as scaling, historical statistics, or model prediction to achieve highly robust identification of the three valid heart sound peaks in S1 and the two valid heart sound peaks in S2. This mechanism effectively overcomes the poor adaptability of fixed thresholds in scenarios with individual differences, motion interference, and device differences, significantly improving the completeness and accuracy of multi-peak structure extraction. It is a key technical step in achieving refined heart sound analysis and lays a solid foundation for subsequent physiological parameter modeling and pathological pattern recognition (i.e., heart sound analysis) based on multi-peak temporal and amplitude characteristics.

[0094] Step S500: Perform heart sound analysis based on the three effective heart sound peak points in the first heart sound and the two effective heart sound peak points in the second heart sound to obtain the heart sound analysis results.

[0095] It should be noted that the heart sound analysis results refer to the various physiological parameters and status judgments generated by analyzing multiple effective heart sound peak points, which can be used for health monitoring, disease early warning, or auxiliary diagnosis.

[0096] In this embodiment, heart sound analysis may include, but is not limited to, the following: Multi-peak time series modeling: Calculate the interval between the three peaks in S1 and analyze their changing trends to evaluate valve closure coordination.

[0097] Amplitude ratio analysis: Compare the amplitude ratio of the main peak values ​​of S1 and S2 to help determine the cardiac function status.

[0098] PTT Refinement Correction: The first effective peak value of S1, P1, is used instead of the traditional main peak of S1 as the starting point of PTT to improve the accuracy of blood pressure estimation.

[0099] Long-term trend tracking: Construct daily S1 three-peak amplitude change curves for remote monitoring of chronic diseases.

[0100] This embodiment overcomes the limitations of traditional "single-peak" analysis by performing structured analysis on the S1 triple peak and S2 double peak, enabling in-depth mining of the complex vibrational modes within heart sounds. The generated analysis results can not only be used to estimate parameters such as cuffless blood pressure and cardiac output, but also support early heart disease screening and personalized health management, significantly enhancing the clinical practical value and user engagement of heart sound signals in wearable devices.

[0101] This embodiment acquires multimodal physiological signals (ECG, PCG, PPG) simultaneously from the target heartbeat. Based on dual anchor points of R-wave peak and pulse trough, it achieves precise S1 / S2 segmentation. Combining a dynamic threshold driven by the heart sound envelope peak and a progressive search mechanism, it robustly identifies three valid heart sound peaks in S1 and two valid heart sound peaks in S2. Structured analysis is then performed based on the multi-peak sequence, thus proposing a hierarchical multimodal heart sound analysis method for low signal-to-noise ratio environments. This method integrates four-dimensional information: electrical timing (ECG-R-wave peak), mechanical response (PPG-pulse trough), acoustic energy (PCG-heart sound envelope), and morphological features (peak-trough phase relationship). It solves the problems of misjudgment and missed judgment caused by traditional methods relying only on a single signal or main peak. Through a progressive process of "coarse segmentation → fine identification → structural analysis," it significantly improves the accuracy, robustness, and physiological interpretability of heart sound analysis, providing a highly reliable intelligent heart sound analysis engine for wearable health devices.

[0102] Compared to related technologies that only extract the "main peak" or maximum amplitude point of S1 and S2, this embodiment can robustly identify multiple effective S1 and S2 peaks in each cardiac cycle, analyze the complex internal signal structure of S1 and S2, and fuse the physiological correspondence and temporal distribution characteristics between different peaks in S1 and S2. This provides more accurate and valuable information in modeling, tracking, and clinical reasoning, and provides a reliable foundation for subsequent PTT measurement, cardiac output estimation, and cuffless blood pressure prediction. In this way, the embodiment of this application can more accurately resolve heart sound signals in low signal-to-noise ratio environments, thereby improving the practicality and reliability of heart sound signals in health monitoring and clinical decision support.

[0103] Based on the first embodiment described above, a heart sound analysis method according to a second embodiment of this application is proposed.

[0104] In the second embodiment of this application, the same or similar content as in the above embodiments can be referred to the above description, and will not be repeated hereafter.

[0105] Please refer to Figure 2 , Figure 2 This is a flowchart illustrating the second embodiment of the heart sound analysis method of this application.

[0106] In this embodiment, the step S400 above, which identifies the three valid heart sound peak points in the first heart sound based on the first threshold, may include steps S430 to S450: Step S430: Based on the peak points in the first heart sound that are greater than the first threshold within the first time interval, identify the first valid heart sound peak point in the first heart sound; It should be noted that, in this embodiment, the first time interval refers to a time window of a certain duration starting from the onset of the first heart sound (i.e., the first time point T1), used to preferentially search for the earliest effective vibration event in S1—mitral valve closure (M1). This first time interval is usually set from T1 to T1+45ms, covering the mitral valve closure event in the early stage of S1, which conforms to the physiological timing characteristics of the initial stage of S1.

[0107] It should also be noted that, in this embodiment, the first effective heart sound peak point in the first heart sound refers to the positive local maximum point in S1 within the first time interval, representing the vibration event of mitral valve closure M1.

[0108] This embodiment avoids the problem of mislocalization of M1 event caused by blindly scanning the entire S1 interval by limiting the search range of the "first time interval" to the first effective peak, and improves the accuracy of identifying the effective heart sound peak point corresponding to the M1 event in the early components of S1.

[0109] In one feasible implementation, step S430 above may include steps S431 to S434: Step S431: Detect peak points that are greater than the first threshold within the first time interval; As those skilled in the art will know, peak detection refers to the process of identifying local maxima in a signal waveform, which is usually achieved by the first derivative sign-changing method (from positive to negative) or the sliding window comparison method (the current point is greater than N sampling points before and after it).

[0110] This implementation method detects peak points of the target heart sound signal within the first time interval and selects candidate peak points with amplitudes greater than the first threshold th1, thereby achieving preliminary screening of high-energy events in the early stage of S1, distinguishing potential physiological vibrations from background noise, and providing a candidate set for subsequent accurate determination.

[0111] Step S432: If a peak point greater than the first threshold is detected within the first time interval, the first valid heart sound peak point in the first heart sound is determined based on the largest peak point within the first time interval. In this embodiment, when one or more candidate peak points with amplitudes exceeding th1 are detected within the first time interval, the one with the largest amplitude is selected as the first candidate valid peak point, and it is further determined whether it is a valid heart sound peak point. This strategy is based on the physiological fact that the valid heart sound peak point corresponding to the M1 event in S1 is often the peak point with the highest amplitude in the first time interval of S1.

[0112] For example, in one feasible implementation, the heart sound envelope signal includes an upper heart sound envelope signal and a lower heart sound envelope signal. The step S432 above, which determines the first valid heart sound peak point in the first heart sound based on the largest peak point within the first time interval, may include steps A10 to A50: Step A10: Take the largest peak point in the first time interval as the first heart sound peak point in the first heart sound. In this example, the largest peak point of the first heart sound within the first time interval is taken as the first heart sound peak point (i.e., the first candidate valid peak point) and enters the subsequent morphological verification process to verify whether it is a valid heart sound peak point and whether it can represent the M1 event in S1.

[0113] Step A20: The heart sound envelope signal between the first time point and the second time point is taken as the first heart sound envelope valley value, wherein the first heart sound envelope valley value belongs to the heart sound lower envelope signal; As those skilled in the art will know, the envelope trough is the local minimum value of the envelope signal within a certain time interval.

[0114] In this example, the heart sound valley value refers to the valley value of the heart sound envelope signal within a certain time interval, representing the weakest moment of heart sound energy within that time interval.

[0115] It is easy to understand that, since the lower envelope signal of the heart sound envelope signal is formed by connecting the local negative minimum points of the target heart sound signal, the envelope valley value of the heart sound envelope signal in a certain time interval, that is, the heart sound valley value, actually belongs to the lower envelope signal of the heart sound.

[0116] In this example, the first heart sound envelope valley value refers to the absolute value of the amplitude of the lowest point in the heart sound envelope signal in the S1 interval.

[0117] This example searches for the valley value of the heart sound envelope signal (or the lower envelope signal) within the S1 interval between the first time point T1 and the second time point T2, and uses it as the first heart sound envelope valley value As1env_min corresponding to the first heart sound. This avoids the problem of incorrect valley value selection caused by fixed time windows or missegmentation. This As1env_min not only reflects the true energy level of S1, but also has the ability to resist noise interference (because the envelope has smoothed the fluctuations of the original signal), providing a stable and reliable reference benchmark for the subsequent adaptive thresholding mechanism.

[0118] Step A30: Adjust the first heart sound envelope trough value based on the third adjustment coefficient to obtain the third threshold, wherein the third threshold is less than the first heart sound envelope trough value; In this example, the third threshold (Threshold 3, th3) ​​is the amplitude judgment threshold used to identify the first target heart sound valley value in S1. Its value is dynamically generated based on the first heart sound envelope valley value As1env_min, ensuring that the first target heart sound valley value in S1 can be accurately identified under different signal intensities.

[0119] It should be noted that the third adjustment coefficient γ is a dimensionless proportionality coefficient between 0 and 1, used to proportionally attenuate the first heart sound envelope valley value As1env_min to generate the third threshold th3.

[0120] For example: th3 = γ × As1env_min; The initial value of γ ranges from 0.15 to 0.25, preferably 0.2.

[0121] The core of this strategy is to set a reasonable recognition threshold downwards based on the absolute value of the minimum overall energy of S1 (i.e., the first heart sound envelope valley value), so as to ensure that the identified valley value is significantly different from the background noise and can also accommodate the amplitude differences caused by the different transmission paths of multiple vibration sources inside S1.

[0122] This example demonstrates a simple, efficient, and computationally inefficient dynamic thresholding mechanism by linearly scaling As1env_min using a third adjustment factor. This method achieves stable adaptation across individuals and devices without requiring additional model training or complex parameter tuning, making it particularly suitable for real-time processing scenarios in resource-constrained wearable devices.

[0123] It is easy to understand that, in addition to the method of determining the third threshold based on the third adjustment coefficient provided in this example, the third adjustment threshold can also be determined by other methods based on the first heart sound envelope valley value. For details, please refer to the implementation method of determining the first threshold and the second threshold described in the first embodiment above. This example will not elaborate on it here.

[0124] Step A40: Detect whether there is a first target heart sound valley point within a preset time after the peak point of the first heart sound, wherein the first target heart sound valley point is greater than the third threshold and the phase of the first target heart sound valley point is opposite to that of the peak point of the first heart sound; It should be noted that, in this example, the first target heart sound valley point refers to a local minimum value that appears within a preset time period (e.g., within 50ms) after the first candidate peak point (i.e., the first heart sound peak point) in S1 and meets the following conditions: the amplitude is greater than th3 (excluding abnormal noise), the phase is opposite to the first candidate peak point in S1 (the peak value is positive and the valley value is negative), and the time is immediately following it, which conforms to the decay response law of mechanical vibration.

[0125] The preset duration mentioned above can be obtained through extensive experimental determination.

[0126] This example implements dual verification of the "peak-valley" physiological waveform structure: it requires both a main peak with sufficient intensity and a symmetrical valley with a reasonable response, thereby effectively eliminating interference items such as isolated electrical impulses and knocking noise, and ensuring that the first valid heart sound peak point determined at the end can accurately represent the M1 event.

[0127] Step A50: If a first target heart sound valley point exists, then the first heart sound peak point of the first heart sound is determined as the first valid heart sound peak point in the first heart sound.

[0128] This example significantly improves the physiological rationality and anti-interference ability of identifying the first valid heart sound peak point in S1 through a dual criterion mechanism of "peak search + valley verification". Only when both conditions of "high amplitude peak value" and "reasonable symmetrical valley value" are met is the heart sound peak point confirmed as a valid heart sound peak point, avoiding the risk of misjudgment in complex noise environments by the single threshold method.

[0129] Furthermore, in a feasible implementation, after step A40 above, the heart sound analysis method may further include steps A60-A70: Step A60: If there is no first target heart sound valley value, the third adjustment coefficient is reduced by the third preset step value to obtain the updated third adjustment coefficient. Then, the step of adjusting the first heart sound envelope valley value based on the third adjustment coefficient is returned to be executed until the first target heart sound valley value is detected within a preset time after the first heart sound peak value, or until the third threshold is less than the preset value. It should be noted that in this embodiment, the third preset step value Δγ is a small proportional increment (e.g., 0.02) used to gradually reduce the γ value, thereby reducing th3, and ultimately gradually relaxing the screening conditions for the first target heart sound valley point, increasing the sensitivity to weak amplitude valley points in S1, until the first target heart sound valley point is screened out from the S1 interval, or the screening conditions are relaxed to the limit (i.e., the third threshold is less than the preset value), determining that there is no first target heart sound valley point. The preset value is a preset empirical lower limit used to prevent the identification thresholds (including the first threshold, second threshold, third threshold, etc.) from being excessively relaxed, leading to false noise reception.

[0130] The adaptive adjustment mechanism of the third threshold in this embodiment reflects a progressive verification strategy of "from strict to lenient": initially, strict standards are used to ensure high accuracy; if failure occurs, the conditions are gradually relaxed to improve recall, taking into account both accuracy and robustness.

[0131] Step A70: If, until the third threshold is less than the preset value, no first target heart sound valley point is detected within the preset time after the first heart sound peak point, then discard the current target heart beat, take other heart beats as new target heart beats, and return to the step of obtaining the target physiological signal of the target heart beat.

[0132] This implementation method avoids the algorithm getting stuck in an invalid loop on severely noisy or distorted heartbeats by setting a "failure discard + retry mechanism". Selecting the next heartbeat to continue heart sound analysis ensures the continuity and stability of the overall monitoring process, and is especially suitable for the automatic fault-tolerant operation of long-term wearable devices.

[0133] Step S433: If no peak point greater than the first threshold is detected in the first time interval, the first adjustment coefficient is reduced by a first preset step value to obtain the updated first adjustment coefficient. Then, the process returns to the step of adjusting the first heart sound envelope peak based on the first adjustment coefficient to obtain the first threshold, until a peak point greater than the first threshold is detected in the first time interval, or until the first threshold is less than the preset value. It should be noted that in this embodiment, the first preset step value Δα is a small proportional decrease (such as 0.01) used to gradually reduce the value of α, thereby reducing th1, and finally gradually relaxing the screening conditions for heart sound peak points, increasing the sensitivity to weak amplitude peak points in S1, until heart sound peak points are screened out from the S1 interval, or the screening conditions are relaxed to the limit (that is, the first threshold is less than the preset value), and it is determined that there are no heart sound peak points.

[0134] In this implementation, dynamic threshold reduction of the third threshold is achieved: when the initial threshold is too high and no peak can be detected, the system automatically lowers the threshold and attempts to find effective signals at a lower intensity, which reflects the adaptive capability of the algorithm.

[0135] Step S434: If no peak point greater than the first threshold is detected in the first time interval until the first threshold is less than the preset value, then discard the current target heartbeat, take other heartbeats as new target heartbeats, and return to the step of obtaining the target physiological signal of the target heartbeat.

[0136] This implementation method avoids the algorithm getting stuck in an invalid loop on severely noisy or distorted heartbeats by setting a "failure discard + retry mechanism". Selecting the next heartbeat to continue heart sound analysis ensures the continuity and stability of the overall monitoring process, and is especially suitable for the automatic fault-tolerant operation of long-term wearable devices.

[0137] This fault-tolerant mechanism ensures that the system does not waste resources on severely poor signals. By skipping abnormal heartbeats and switching to the next cardiac cycle to continue heart sound analysis, it maintains the continuity and reliability of overall monitoring and is a key design feature for achieving long-term stable operation.

[0138] Step S440: Based on the first effective heart sound peak point in the first heart sound, determine the second time interval; based on the peak points in the second time interval that are greater than the first threshold, identify the second effective heart sound peak point in the first heart sound; wherein, the second time interval is the first preset time interval after the first effective heart sound peak point in the first heart sound. It should be noted that, in this embodiment, the first preset time interval refers to a time window formed by extending a first preset duration (e.g., 30-60 ms) backward from the first valid heart sound peak point in S1, used to search for the secondary vibration event in S1—tricuspid valve closure T1. The first preset duration can be a preset value to ensure that the second time interval effectively covers the tricuspid valve closure event in S1.

[0139] It should also be noted that, in this embodiment, the second effective heart sound peak point in the first heart sound refers to the positive local maximum point in S1 within the second time interval, representing the vibration event of tricuspid valve closure T1.

[0140] This embodiment employs a "progressive window positioning" strategy: starting from the previous valid heart sound peak point, the search range for the next valid heart sound peak point is limited. This avoids blindly scanning across the entire S1 interval and misidentifying peaks, improving the accuracy of identifying valid heart sound peak points corresponding to T1 events in the S1 component, and ensuring the temporal logic correctness of the S1 multi-peak structure. This implementation simulates the propagation sequence of cardiac mechanical activity, improving the physiological consistency of multi-peak identification.

[0141] In one feasible implementation, the step of identifying the second valid heart sound peak point in the first heart sound based on the peak points greater than the first threshold within the second time interval in step S440 above may include steps S441 to S444: Step S441: Detect peak points that are greater than the first threshold within the second time interval; Step S442: If a peak point greater than the first threshold is detected in the second time interval, then the second valid heart sound peak point in the first heart sound is determined based on the largest peak point in the second time interval. For example, in one feasible implementation, the step S442 above, which determines the second valid heart sound peak point in the first heart sound based on the largest peak point within the second time interval, may include steps B10 to B30: Step B10: Take the largest peak point in the second time interval as the second heart sound peak point in the first heart sound; Step B20: Detect whether there is a second target heart sound valley point within a preset time after the peak point of the second heart sound of the first heart sound, wherein the second target heart sound valley point is greater than the third threshold and the phase of the second target heart sound valley point is opposite to that of the peak point of the second heart sound of the first heart sound; It should be noted that in this example, the second target heart sound valley point refers to a local minimum value that appears within a preset time (e.g., within 50ms) after the second heart sound peak point in S1 and meets the following conditions: the amplitude is greater than th3 (excluding abnormal noise), the phase is opposite to the second heart sound peak point in S1 (the peak value is positive and the valley value is negative), and it follows immediately in time, conforming to the decay response law of mechanical vibration.

[0142] Step B30: If a second target heart sound valley point exists, then the second heart sound peak point of the first heart sound is determined as the second valid heart sound peak point in the first heart sound.

[0143] The operation steps and technical effects of this example are similar to those of steps A10 to A50 in step S432, and will not be repeated here. Those skilled in the art can refer to the examples corresponding to steps A10 to A50 above.

[0144] Furthermore, in one feasible implementation, after step B20 described above, the heart sound analysis method may further include steps B40 to B50: Step B40: If there is no second target heart sound valley value, the third adjustment coefficient is reduced by the third preset step value to obtain the updated third adjustment coefficient. Then, the step of adjusting the first heart sound envelope valley value based on the third adjustment coefficient is returned to be executed until the second heart sound peak value of the first heart sound is detected and there is a second target heart sound valley value within a preset time, or until the third threshold is less than the preset value. Step B50: If, until the third threshold is less than the preset value, no second target heart sound valley point is detected within the preset time after the second heart sound peak point of the first heart sound, then discard the current target heart beat, take other heart beats as new target heart beats, and return to the step of obtaining the target physiological signal of the target heart beat.

[0145] The operation steps and technical effects of this embodiment are similar to those of the embodiments corresponding to steps A60 to A70 in step S432, and will not be described in detail here. Those skilled in the art can refer to the embodiments corresponding to steps A60 to A70 above.

[0146] Step S443: If no peak point greater than the first threshold is detected in the second time interval, the first adjustment coefficient is reduced by a first preset step value to obtain the updated first adjustment coefficient. Then, the process returns to the step of adjusting the first heart sound envelope peak value based on the first adjustment coefficient to obtain the first threshold, until a peak point greater than the first threshold is detected in the second time interval, or until the first threshold is less than the preset value. Step S444: If no peak point greater than the first threshold is detected in the second time interval until the first threshold is less than the preset value, then discard the current target heartbeat, take other heartbeats as new target heartbeats, and return to the step of obtaining the target physiological signal of the target heartbeat.

[0147] The operation steps and technical effects of this embodiment are similar to those of the embodiments corresponding to steps S431 to S434 in step S430, and will not be repeated here. Those skilled in the art can refer to the embodiments corresponding to steps S431 to S434 above.

[0148] Step S450: Based on the second valid heart sound peak point in the first heart sound, determine the third time interval; based on the peak points in the third time interval that are greater than the first threshold, identify the third valid heart sound peak point in the first heart sound, wherein the third time interval is the second preset time interval after the second valid heart sound peak point in the first heart sound.

[0149] It should be noted that, in this embodiment, the second preset time interval refers to a time window formed by extending a second preset duration (e.g., 30-60 ms) backward from the second effective heart sound peak point in S1, used to search for vibration events in S1—ventricular myocardial contraction. The second preset duration can be a preset value to ensure that the third time interval effectively covers the ventricular myocardial contraction events in S1.

[0150] It should also be noted that, in this embodiment, the third effective heart sound peak point in the first heart sound refers to the positive local maximum point representing the ventricular myocardial contraction vibration event in S1 within the third time interval.

[0151] In one feasible implementation, the step of identifying the third valid heart sound peak point in the first heart sound based on the peak points greater than the first threshold within the third time interval in step S450 above may include steps S451 to S454: Step S451: Detect peak points that are greater than the first threshold within the third time interval; Step S452: If a peak point greater than the first threshold is detected in the third time interval, the third valid heart sound peak point in the first heart sound is determined based on the largest peak point in the third time interval. For example, in one feasible implementation, the step S452 above, which determines the third valid heart sound peak point in the first heart sound based on the largest peak point within the third time interval, may include steps C10 to C30: Step C10: Take the largest peak point in the third time interval as the third heart sound peak point in the first heart sound; Step C20: Detect whether there is a third target heart sound valley point within a preset time after the third heart sound peak point of the first heart sound, wherein the third target heart sound valley point is greater than the third threshold, and the third target heart sound valley point is out of phase with the third heart sound peak point of the first heart sound; It should be noted that in this example, the third target heart sound valley point refers to a local minimum value that appears within a preset time (e.g., within 50ms) after the third heart sound peak point in S1 and meets the following conditions: the amplitude is greater than th3 (excluding abnormal noise), the phase is opposite to the third heart sound peak point in S1 (the peak value is positive and the valley value is negative), and it follows immediately in time, which conforms to the decay response law of mechanical vibration.

[0152] Step C30: If a third target heart sound valley point exists, then the third heart sound peak point of the first heart sound is determined as the second valid heart sound peak point in the first heart sound.

[0153] The operation steps and technical effects of this example are similar to those of steps A10 to A50 in step S432, and will not be repeated here. Those skilled in the art can refer to the examples corresponding to steps A10 to A50 above.

[0154] Furthermore, in a feasible implementation, after step C20 above, the heart sound analysis method may further include steps C40 to C50: Step C40: If there is no third target heart sound valley value, the third adjustment coefficient is reduced by the third preset step value to obtain the updated third adjustment coefficient. Then, the step of adjusting the first heart sound envelope valley value based on the third adjustment coefficient is returned to be executed until a third target heart sound valley value is detected within a preset time after the third heart sound peak value of the first heart sound, or until the third threshold is less than the preset value. Step C50: If, until the third threshold is less than the preset value, no third target heart sound peak point is detected within the preset time after the third target heart sound peak point, then discard the current target heart beat, take other heart beats as new target heart beats, and return to the step of obtaining the target physiological signal of the target heart beat.

[0155] The operation steps and technical effects of this embodiment are similar to those of the embodiments corresponding to steps A60 to A70 in step S432, and will not be described in detail here. Those skilled in the art can refer to the embodiments corresponding to steps A60 to A70 above.

[0156] Step S453: If no peak point greater than the first threshold is detected in the third time interval, the first adjustment coefficient is reduced by a first preset step value to obtain the updated first adjustment coefficient. Then, the process returns to the step of adjusting the first heart sound envelope peak value based on the first adjustment coefficient to obtain the first threshold, until a peak point greater than the first threshold is detected in the third time interval, or until the first threshold is less than the preset value. Step S454: If no peak point greater than the first threshold is detected in the third time interval until the first threshold is less than the preset value, then discard the current target heartbeat, take other heartbeats as new target heartbeats, and return to the step of obtaining the target physiological signal of the target heartbeat.

[0157] The operation steps and technical effects of this embodiment are similar to those of the embodiments corresponding to steps S431 to S434 in step S430, and will not be repeated here. Those skilled in the art can refer to the embodiments corresponding to steps S431 to S434 above.

[0158] This embodiment constructs a complete identification process for the three-peak structure within S1 through a "chain-progressive" search mechanism, achieving systematic analysis of the complex vibration modes of S1. This structured method of extracting heart sound features breaks through the limitations of traditional single-peak analysis, providing high-dimensional feature inputs for subsequent valve coordination assessment and cardiac function status modeling. It significantly improves the completeness and accuracy of identifying the multi-peak structure of S1 in complex scenarios such as low signal-to-noise ratio and motion interference. It is a key supporting technology for realizing refined heart sound analysis and advanced applications such as cuffless blood pressure and cardiac function assessment, fully demonstrating the innovative value and practical potential of this application in the field of wearable health monitoring.

[0159] Based on the above embodiments, a heart sound analysis method according to a third embodiment of this application is proposed.

[0160] In the third embodiment of this application, the same or similar content as the above embodiments can be referred to the above description, and will not be repeated hereafter.

[0161] Please refer to Figure 3 , Figure 3 This is a flowchart illustrating the third embodiment of the heart sound analysis method of this application.

[0162] In this embodiment, the step S400 above, which identifies two valid heart sound peak points in the second heart sound based on the second threshold, may further include steps S460 to S470: Step S460: Based on the peak points in the fourth time interval of the second heart sound that are greater than the second threshold, identify the first valid heart sound peak point in the second heart sound; It should be noted that, in this embodiment, the fourth time interval refers to a time window of a certain duration starting from the beginning of the second heart sound (i.e., the second time point T2), used to preferentially search for the earliest effective vibration event in S2—aortic valve closure (A2). This fourth time interval is usually set to the time corresponding to the peak of the second heart sound envelope from T2, covering the aortic valve closure event in S2, which conforms to the physiological temporal characteristics of the initial stage of S1.

[0163] It should also be noted that, in this embodiment, the first effective heart sound peak point in the second heart sound refers to the positive local maximum point in the second time interval S2 that represents the vibration event of aortic valve closure A2.

[0164] In one feasible implementation, step S460 above may include steps S461 to S464: Step S461: Detect peak points that are greater than the second threshold within the fourth time interval; Step S462: If a peak point greater than the second threshold is detected in the fourth time interval, the first valid heart sound peak point in the second heart sound is determined based on the largest peak point in the fourth time interval. For example, in one feasible implementation, the heart sound envelope signal includes an upper heart sound envelope signal and a lower heart sound envelope signal. The step S462 above, which determines the first valid heart sound peak point in the first heart sound based on the largest peak point within the fourth time interval, may include steps D10 to D50: Step D10: Take the largest peak point in the fourth time interval as the first heart sound peak point in the second heart sound. Step D20: The heart sound envelope signal between the first time point and the third time point is taken as the second heart sound envelope valley value, wherein the second heart sound envelope valley value belongs to the heart sound lower envelope signal; In this example, the second heart sound envelope valley value refers to the absolute value of the amplitude of the lowest point in the heart sound envelope signal in the S2 interval.

[0165] Step D30: Adjust the second heart sound envelope trough value based on the fourth adjustment coefficient to obtain the fourth threshold, wherein the fourth threshold is less than the second heart sound envelope trough value; In this example, the fourth threshold (Threshold 4, th4) is the amplitude judgment threshold used to identify the second target heart sound valley value in S2. Its value is dynamically generated based on the second heart sound envelope valley value As2env_min, ensuring that the second target heart sound valley value in S2 can be accurately identified under different signal intensities.

[0166] It should be noted that the fourth adjustment coefficient δ is a dimensionless proportionality coefficient between 0 and 1, used to proportionally attenuate the second heart sound envelope valley value As2env_min to generate the fourth threshold th4.

[0167] For example: th4 = δ × As2env_min; The initial value of δ ranges from 0.15 to 0.25, with 0.2 being preferred.

[0168] The core of this strategy is to set a reasonable recognition threshold downward based on the absolute value of the minimum overall energy of S2 (i.e., the second heart sound envelope valley value), so as to ensure that the identified valley value is significantly different from the background noise and can also accommodate the amplitude differences caused by the different transmission paths of multiple vibration sources inside S2.

[0169] It is easy to understand that, in addition to the method of determining the fourth threshold based on the fourth adjustment coefficient provided in this example, the fourth adjustment threshold can also be determined by other methods based on the second heart sound envelope valley value. For details, please refer to the implementation method of determining the first threshold and the second threshold described in the first embodiment above. This example will not elaborate on it here.

[0170] Step D40: Detect whether there is a fourth target heart sound valley point within a preset time after the peak point of the first heart sound of the second heart sound, wherein the fourth target heart sound valley point is greater than the fourth threshold and the fourth target heart sound valley point is out of phase with the peak point of the first heart sound of the second heart sound; It should be noted that, in this example, the fourth target heart sound valley point refers to a local minimum value that appears within a preset time (e.g., within 50ms) after the first candidate peak point (i.e., the first heart sound peak point) in S2 and meets the following conditions: the amplitude is greater than th4 (excluding abnormal noise), the phase is opposite to the first candidate peak point in S2 (the peak value is positive and the valley value is negative), and it follows immediately in time, conforming to the decay response law of mechanical vibration.

[0171] Step D50: If there is a fourth target heart sound valley point, then the first heart sound peak point of the second heart sound is determined as the first valid heart sound peak point in the second heart sound.

[0172] The operation steps and technical effects of this example are similar to those of steps A10 to A50 in step S432 of the second embodiment, and will not be repeated here. Those skilled in the art can refer to the examples corresponding to steps A10 to A50 in the second embodiment.

[0173] Furthermore, in a feasible implementation, after step D40 described above, the heart sound analysis method may further include steps D60-D70: Step D60: If there is no fourth target heart sound valley value, the fourth adjustment coefficient is reduced by the fourth preset step value to obtain the updated fourth adjustment coefficient. Then, the step of adjusting the second heart sound envelope valley value based on the fourth adjustment coefficient is returned to be executed until the fourth target heart sound valley value is detected within a preset time after the first heart sound peak point of the second heart sound, or until the fourth threshold is less than the preset value. It should be noted that in this embodiment, the fourth preset step value Δδ is a small proportional increment (e.g., 0.02) used to gradually reduce the δ value, thereby reducing th4, and ultimately gradually relaxing the screening conditions for the fourth target heart sound valley point, increasing the sensitivity to weak amplitude valley points in S2, until the fourth target heart sound valley point is screened out from the S2 interval, or the screening conditions are relaxed to the limit (i.e., the fourth threshold is less than the preset value), determining that there is no fourth target heart sound valley point. The preset value is a preset empirical lower limit used to prevent the identification thresholds (including the first threshold, second threshold, third threshold, and fourth threshold) from being excessively relaxed, leading to false noise reception.

[0174] Step D70: If, until the fourth threshold is less than the preset value, no fourth target heart sound valley point is detected within the preset time after the first heart sound peak point of the second heart sound, then discard the current target heart beat, take other heart beats as new target heart beats, and return to the step of obtaining the target physiological signal of the target heart beat.

[0175] The operation steps and technical effects of this embodiment are similar to those of steps A60 to A70 in step S432 of the second embodiment, and will not be described in detail here. Those skilled in the art can refer to the embodiments corresponding to steps A60 to A70 in the second embodiment.

[0176] Step S463: If no peak point greater than the second threshold is detected in the fourth time interval, the second adjustment coefficient is reduced by the second preset step value to obtain the updated second adjustment coefficient. Then, the process returns to the step of adjusting the second heart sound envelope peak based on the second adjustment coefficient to obtain the second threshold, until a peak point greater than the second threshold is detected in the fourth time interval, or until the second threshold is less than the preset value. It should be noted that in this embodiment, the second preset step value Δβ is a small proportional decrease (such as 0.01) used to gradually reduce the β value, thereby reducing th2, and finally gradually relaxing the screening conditions for heart sound peak points, increasing the sensitivity to weak amplitude peak points in S2, until heart sound peak points are screened out from the S2 interval, or the screening conditions are relaxed to the limit (that is, the second threshold is less than the preset value), and it is determined that there are no heart sound peak points.

[0177] In this implementation, dynamic threshold reduction of the fourth threshold is achieved: when the initial threshold is too high and no peak can be detected, the system automatically lowers the threshold and attempts to find effective signals at a lower intensity, which reflects the adaptive capability of the algorithm.

[0178] Step S464: If no peak point greater than the second threshold is detected in the fourth time interval until the second threshold is less than the preset value, then discard the current target heartbeat, take other heartbeats as new target heartbeats, and return to the step of obtaining the target physiological signal of the target heartbeat.

[0179] The operation steps and technical effects of this embodiment are similar to those of steps S431 to S434 in step S430 of the second embodiment, and will not be described in detail here. Those skilled in the art can refer to the embodiments corresponding to steps S431 to S434 in the second embodiment.

[0180] Step S470: Based on the first valid heart sound peak point in the second heart sound, determine the fifth time interval; based on the peak points in the fifth time interval that are greater than the second threshold, identify the second valid heart sound peak point in the second heart sound; wherein, the fifth time interval is the third preset time interval after the first valid heart sound peak point in the second heart sound.

[0181] It should be noted that, in this embodiment, the third preset time interval refers to a time window formed by extending a third preset duration (e.g., 30-60 ms) from the first valid heart sound peak point in S2, used to search for the secondary vibration event in S2—pulmonary valve closure P2. The third preset duration can be a preset value to ensure that the fifth time interval effectively covers the pulmonary valve closure event in S2.

[0182] It should also be noted that, in this embodiment, the second effective heart sound peak point in the second heart sound refers to the positive local maximum point in S2 within the fifth time interval, which represents the vibration event of pulmonary valve closure P2.

[0183] This embodiment employs a "progressive window positioning" strategy: starting from the previous valid heart sound peak point, the search range for the next valid heart sound peak point is limited. This avoids blindly scanning across the entire S2 interval and misidentifying peaks, improving the accuracy of identifying valid heart sound peak points corresponding to P2 events in the S2 component, and ensuring the temporal logic correctness of the S2 multi-peak structure. This implementation simulates the propagation sequence of cardiac mechanical activity, improving the physiological consistency of multi-peak identification.

[0184] In one feasible implementation, the step of identifying the second valid heart sound peak point in the second heart sound based on the peak point greater than the second threshold within the fifth time interval in step S470 above may include steps S471 to S474: Step S471: Detect peak points that are greater than the second threshold within the fifth time interval; Step S472: If a peak point greater than the second threshold is detected in the fifth time interval, then the second valid heart sound peak point in the second heart sound is determined based on the largest peak point in the fifth time interval. For example, in one feasible implementation, the step of determining the second valid heart sound peak point in the second heart sound based on the largest peak point within the fifth time interval in step S472 above may include steps E10~E30: Step E10: Take the largest peak point in the fifth time interval as the second heart sound peak point in the second heart sound. Step E20: Detect whether there is a fifth target heart sound valley point within a preset time after the peak point of the second heart sound. The fifth target heart sound valley point is greater than the fourth threshold and the phase of the fifth target heart sound valley point is opposite to that of the peak point of the second heart sound. It should be noted that in this example, the fifth target heart sound valley point refers to a local minimum value that appears within a preset time (e.g., within 50ms) after the second heart sound peak point in S2 and meets the following conditions: the amplitude is greater than th4 (excluding abnormal noise), the phase is opposite to the second heart sound peak point in S2 (the peak value is positive and the valley value is negative), and it follows immediately in time, which conforms to the decay response law of mechanical vibration.

[0185] In step E30, if there is a fifth target heart sound valley point, then the second heart sound peak point of the second heart sound is determined as the second effective heart sound peak point in the second heart sound.

[0186] The operation steps and technical effects of this example are similar to those of steps A10 to A50 in step S432 of the second embodiment, and will not be repeated here. Those skilled in the art can refer to the examples corresponding to steps A10 to A50 in the second embodiment.

[0187] Furthermore, in a feasible implementation, after step E20 described above, the heart sound analysis method may further include steps E40 to E50: Step E40: If there is no fifth target heart sound valley value, the fourth adjustment coefficient is reduced by the fourth preset step value to obtain the updated fourth adjustment coefficient. Then, the step of adjusting the second heart sound envelope valley value based on the fourth adjustment coefficient is returned to be executed until a fifth target heart sound valley value is detected within a preset time after the second heart sound peak value of the second heart sound, or until the fourth threshold is less than the preset value. In step E50, if no fifth target heart sound valley point is detected within a preset time after the second heart sound peak point until the fourth threshold is less than the preset value, then discard the current target heart beat, take other heart beats as new target heart beats, and return to the step of obtaining the target physiological signal of the target heart beat.

[0188] The operation steps and technical effects of this embodiment are similar to those of steps A60 to A70 in step S432 of the second embodiment, and will not be described in detail here. Those skilled in the art can refer to the embodiments corresponding to steps A60 to A70 in the second embodiment.

[0189] Step S473: If no peak point greater than the second threshold is detected in the fifth time interval, the second adjustment coefficient is reduced by the second preset step value to obtain the updated second adjustment coefficient. Then, the process returns to the step of adjusting the second heart sound envelope peak based on the second adjustment coefficient to obtain the second threshold, until a peak point greater than the second threshold is detected in the fifth time interval, or until the second threshold is less than the preset value. Step S474: If no peak point greater than the second threshold is detected in the fifth time interval until the second threshold is less than the preset value, then discard the current target heartbeat, take other heartbeats as new target heartbeats, and return to the step of obtaining the target physiological signal of the target heartbeat.

[0190] The operation steps and technical effects of this embodiment are similar to those of steps S431 to S434 in step S430 of the second embodiment, and will not be described in detail here. Those skilled in the art can refer to the embodiments corresponding to steps S431 to S434 in the second embodiment.

[0191] like Figure 4 As shown, in this specific embodiment, synchronous signal acquisition is first performed in step S110, and the acquired raw physiological signals are preprocessed to extract the target physiological signals of the target heartbeat, including the target electrocardiogram signal ECG, the target heart sound signal PCG, and the target pulse signal PPG. Then, Hilbert transform is performed on the target physiological signals in step S330 to obtain the heart sound envelope signal. Next, feature point identification is performed on the heart sound envelope signal in steps S340 and S350 to obtain the first heart sound envelope peak value and the second heart sound envelope peak value. Subsequently, dynamic threshold adjustment is performed in steps S400 to S470 to complete the identification of the first peak point and the selection of effective peak values. Finally, three effective heart sound peak points in the first heart sound and two effective heart sound peak points in the second heart sound are determined.

[0192] It should be noted that the above embodiments / implementations are only used to assist in understanding this application and do not constitute a limitation on the heart sound analysis method of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0193] In addition, please refer to Figure 5 , Figure 5 This is a schematic diagram of the electronic device structure of the hardware operating environment involved in the heart sound analysis method in the embodiments of this application.

[0194] This application also provides an electronic device, which includes: 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 perform the steps of the heart sound analysis method in the above embodiments.

[0195] The following is for reference. Figure 5 The diagram illustrates a structural schematic of an electronic device suitable for implementing the embodiments of this application. The electronic device in the embodiments of this application may include, but is not limited to, mobile phones, tablets, desktop computers, vehicle terminals, wearable devices, and any electronic device capable of performing the above functions. Figure 5The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0196] like Figure 5 As shown, the electronic device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory 1002 or a program loaded from a storage device 1003 into a random access memory 1004. The random access memory 1004 also stores various programs and data required for the operation of the electronic device. The processing unit 1001, the read-only memory 1002, and the random access memory 1004 are interconnected via a bus 1005. An input / output interface 1006 is also connected to the bus. Typically, the following systems can be connected to the input / output interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. The communication device 1009 allows the electronic device to communicate wirelessly or wiredly with other devices to exchange data. Although the diagrams show electronic devices with various systems, it should be understood that it is not required to implement or have all of the systems shown. More or fewer systems may be implemented alternatively.

[0197] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0198] The electronic device provided in this application, employing the heart sound analysis method described in the above embodiments, can solve the technical problem of how to more accurately analyze heart sound signals in low signal-to-noise ratio environments, thereby improving the practicality and reliability of heart sound signals in health monitoring and clinical decision support. Compared with the prior art, the beneficial effects of the electronic device provided in this application are the same as those of the heart sound analysis method provided in the above embodiments, and other technical features of the electronic device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0199] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0200] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of protection of the above claims.

[0201] In addition, this application also provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to perform the steps of the heart sound analysis method in the above embodiments.

[0202] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, optical fibers, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (Radio Frequency), etc., or any suitable combination thereof.

[0203] The aforementioned computer-readable storage medium may be included in an electronic device or may exist independently without being assembled into an electronic device.

[0204] The aforementioned computer-readable storage medium carries one or more programs. When the aforementioned one or more programs are executed by an electronic device, the electronic device causes the electronic device to: acquire the target physiological signal of the target heartbeat, wherein the target physiological signal includes a target electrocardiogram (ECG) signal, a target heart sound signal, and a target pulse signal; identify the R-wave peak point of the target ECG signal and take the time corresponding to the R-wave peak point as a first time point; identify the trough point of the target pulse signal and take the time corresponding to the trough point as a second time point; segment the target heart sound signal based on the first time point and the second time point to obtain a first heart sound and a second heart sound, and determine the first heart sound envelope peak value corresponding to the first heart sound and the second heart sound envelope peak value corresponding to the second heart sound; determine a first threshold based on the first heart sound envelope peak value and a second threshold based on the second heart sound envelope peak value; identify three valid heart sound peak points in the first heart sound according to the first threshold and two valid heart sound peak points in the second heart sound according to the second threshold; perform heart sound analysis based on the three valid heart sound peak points in the first heart sound and the two valid heart sound peak points in the second heart sound to obtain heart sound analysis results.

[0205] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0206] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0207] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0208] The computer-readable storage medium provided in this application stores computer-readable program instructions (i.e., a computer program) for performing the steps of the heart sound analysis method in the above embodiments. It solves the technical problem of how to more accurately analyze heart sound signals in low signal-to-noise ratio environments, thereby improving the practicality and reliability of heart sound signals in health monitoring and clinical decision support. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the heart sound analysis method provided in the above embodiments, and will not be repeated here.

[0209] Furthermore, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the heart sound analysis method described above.

[0210] The computer program product provided in this application solves the technical problem of how to more accurately analyze heart sound signals in low signal-to-noise ratio environments, thereby improving the practicality and reliability of heart sound signals in health monitoring and clinical decision support. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the heart sound analysis method provided in the above embodiments, and will not be repeated here.

[0211] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for analyzing heart sounds, characterized in that, include: Acquire the target physiological signals of the target heartbeat, wherein the target physiological signals include the target electrocardiogram signal, the target heart sound signal, and the target pulse signal; Identify the peak point of the R wave in the target electrocardiogram signal and take the time corresponding to the peak point as the first time point; identify the trough point of the target pulse signal and take the time corresponding to the trough point as the second time point; Based on the first time point and the second time point, the target heart sound signal is segmented to obtain the first heart sound and the second heart sound, and the first heart sound envelope peak value corresponding to the first heart sound and the second heart sound envelope peak value corresponding to the second heart sound are determined. Based on the first heart sound envelope peak value, a first threshold is determined, and based on the second heart sound envelope peak value, a second threshold is determined; based on the first threshold, three valid heart sound peak points in the first heart sound are identified, and based on the second threshold, two valid heart sound peak points in the second heart sound are identified; Heart sound analysis is performed based on the three effective heart sound peak points in the first heart sound and the two effective heart sound peak points in the second heart sound to obtain the heart sound analysis results.

2. The heart sound analysis method as described in claim 1, characterized in that, The steps for acquiring the target physiological signal of the target heartbeat include: Simultaneously acquire raw physiological signals, including raw electrocardiogram signals, raw pulse signals, and raw heart sound signals; The original physiological signal is preprocessed to obtain a preprocessed physiological signal. The preprocessing includes at least one of filtering, normalization, and noise reduction. The preprocessed physiological signal includes a preprocessed electrocardiogram signal, a preprocessed pulse signal, and a preprocessed heart sound signal. Based on the preprocessed physiological signal, the target physiological signal of the target heartbeat is obtained.

3. The heart sound analysis method as described in claim 1, characterized in that, The step of segmenting the target heart sound signal based on the first time point and the second time point to obtain the first heart sound and the second heart sound includes: The target heart sound signal between the first time point and the second time point is taken as the first heart sound; and, The heart sound signal of the target heart sound signal between the second time point and the third time point is taken as the second heart sound; The third time point is the time point that continues for a preset duration after the second time point. The preset duration is determined based on the product of the duration of the cardiac cycle and a preset coefficient factor, and the preset coefficient factor ranges from 0.35 to 0.

45.

4. The heart sound analysis method as described in claim 1, characterized in that, The steps of determining the first heart sound envelope peak value corresponding to the first heart sound and the second heart sound envelope peak value corresponding to the second heart sound include: Perform a Hilbert transform on the target heart sound signal to obtain the heart sound envelope signal; The peak value of the heart sound envelope signal between the first time point and the second time point is taken as the first heart sound envelope peak value; and... The peak value between the second time point and the third time point is taken as the second heart sound envelope peak value; The third time point is the time point that continues for a preset duration after the second time point. The preset duration is determined based on the product of the duration of the cardiac cycle and a preset coefficient factor, and the preset coefficient factor ranges from 0.35 to 0.

45.

5. The heart sound analysis method as described in claim 4, characterized in that, The steps of determining a first threshold based on the first heart sound envelope peak value and determining a second threshold based on the second heart sound envelope peak value include: The peak value of the first heart sound envelope is adjusted based on a first adjustment coefficient to obtain a first threshold, wherein the first threshold is less than the peak value of the first heart sound envelope; and The second heart sound envelope peak value is adjusted based on the second adjustment coefficient to obtain a second threshold value, wherein the second threshold value is less than the second heart sound envelope peak value.

6. The heart sound analysis method as described in claim 5, characterized in that, The steps for identifying three valid heart sound peak points in the first heart sound based on the first threshold include: Based on the peak points in the first heart sound that are greater than the first threshold within the first time interval, identify the first valid heart sound peak point in the first heart sound. Based on the first valid heart sound peak point in the first heart sound, a second time interval is determined, and based on the peak points in the second time interval that are greater than the first threshold, the second valid heart sound peak point in the first heart sound is identified, wherein the second time interval is the first preset time interval after the first valid heart sound peak point in the first heart sound; Based on the second valid heart sound peak point in the first heart sound, a third time interval is determined. Based on the peak points in the third time interval that are greater than the first threshold, the third valid heart sound peak point in the first heart sound is identified. The third time interval is the second preset time interval after the second valid heart sound peak point in the first heart sound.

7. The heart sound analysis method as described in claim 6, characterized in that, The step of identifying the first valid heart sound peak point in the first heart sound based on peak points greater than the first threshold within a first time interval in the first heart sound includes: Detect peak points that exceed the first threshold within the first time interval; If a peak point greater than the first threshold is detected within the first time interval, the first valid heart sound peak point in the first heart sound is determined based on the largest peak point within the first time interval. If it is detected that there is no peak point greater than the first threshold in the first time interval, the first adjustment coefficient is reduced by a first preset step value to obtain an updated first adjustment coefficient. Then, the step of adjusting the first heart sound envelope peak value based on the first adjustment coefficient to obtain the first threshold is returned to be executed until a peak point greater than the first threshold is detected in the first time interval, or until the first threshold is less than a preset value. If no peak point greater than the first threshold is detected within the first time interval until the first threshold is less than the preset value, then the current target heartbeat is discarded, other heartbeats are taken as new target heartbeats, and the process returns to the step of obtaining the target physiological signal of the target heartbeat.

8. The heart sound analysis method as described in claim 7, characterized in that, The heart sound envelope signal includes an upper heart sound envelope signal and a lower heart sound envelope signal. The step of determining the first valid heart sound peak point in the first heart sound based on the largest peak point within the first time interval includes: The largest peak point within the first time interval is taken as the first heart sound peak point in the first heart sound. The heart sound envelope signal at the heart sound valley value between the first time point and the second time point is taken as the first heart sound envelope valley value, wherein the first heart sound envelope valley value belongs to the heart sound lower envelope signal; The first heart sound envelope trough value is adjusted based on the third adjustment coefficient to obtain a third threshold, wherein the third threshold is less than the first heart sound envelope trough value; After detecting the peak value of the first heart sound, there is a first target heart sound valley value within a preset time period, wherein the first target heart sound valley value is greater than the third threshold, and the first target heart sound valley value is out of phase with the peak value of the first heart sound. If the first target heart sound valley point exists, then the first heart sound peak point of the first heart sound is determined as the first valid heart sound peak point in the first heart sound.

9. The heart sound analysis method as described in claim 8, characterized in that, After detecting whether a first target heart sound trough exists within a preset time period after the peak value of the first heart sound, the method further includes: If the first target heart sound valley value does not exist, the third adjustment coefficient is reduced by a third preset step value to obtain the updated third adjustment coefficient. Then, the step of adjusting the first heart sound envelope valley value based on the third adjustment coefficient is returned to be executed until the first target heart sound valley value is detected within a preset time after the first heart sound peak value of the first heart sound is detected, or until the third threshold is less than the preset value. If, even if the third threshold is less than the preset value, no first target heart sound valley point is detected within a preset time after the first heart sound peak point, then the current target heart beat is discarded, and other heart beats are taken as new target heart beats, and the process returns to the step of obtaining the target physiological signal of the target heart beat.

10. The heart sound analysis method as described in claim 8 or 9, characterized in that, The step of identifying the second valid heart sound peak point in the first heart sound based on peak points greater than the first threshold within the second time interval includes: Detect peak points that exceed the first threshold within the second time interval; If a peak point greater than the first threshold is detected within the second time interval, then the second valid heart sound peak point in the first heart sound is determined based on the largest peak point within the second time interval. If it is detected that there is no peak point greater than the first threshold in the second time interval, the first adjustment coefficient is reduced by a first preset step value to obtain the updated first adjustment coefficient. Then, the step of adjusting the first heart sound envelope peak value based on the first adjustment coefficient to obtain the first threshold is returned to be executed until a peak point greater than the first threshold is detected in the second time interval, or until the first threshold is less than the preset value. If no peak point greater than the first threshold is detected in the second time interval until the first threshold is less than the preset value, then the current target heartbeat is discarded, other heartbeats are taken as new target heartbeats, and the process returns to the step of obtaining the target physiological signal of the target heartbeat.

11. The heart sound analysis method as described in claim 8 or 9, characterized in that, The step of identifying the third valid heart sound peak point in the first heart sound based on peak points greater than the first threshold within the third time interval includes: Detect peak points that are greater than the first threshold within the third time interval; If a peak point greater than the first threshold is detected within the third time interval, then the third valid heart sound peak point in the first heart sound is determined based on the largest peak point within the third time interval. If it is detected that there is no peak point greater than the first threshold in the third time interval, the first adjustment coefficient is reduced by a first preset step value to obtain an updated first adjustment coefficient. Then, the step of adjusting the first heart sound envelope peak value based on the first adjustment coefficient to obtain the first threshold is returned to be executed until a peak point greater than the first threshold is detected in the third time interval, or until the first threshold is less than a preset value. If no peak point greater than the first threshold is detected in the third time interval until the first threshold is less than the preset value, then the current target heartbeat is discarded, other heartbeats are taken as new target heartbeats, and the process returns to the step of obtaining the target physiological signal of the target heartbeat.

12. The heart sound analysis method as described in claim 10, characterized in that, Based on the largest peak point within the second time interval, the second valid heart sound peak point in the first heart sound is determined, including: The largest peak point in the second time interval is taken as the second heart sound peak point in the first heart sound; After detecting the second peak point of the first heart sound, does a second target heart sound valley point exist within a preset time period? The second target heart sound valley point is greater than the third threshold, and the second target heart sound valley point is out of phase with the second peak point of the first heart sound. If the second target heart sound valley point exists, then the second heart sound peak point of the first heart sound is determined as the second valid heart sound peak point in the first heart sound.

13. The heart sound analysis method as described in claim 12, characterized in that, After detecting whether a second target heart sound trough exists within a preset time period after the peak point of the second heart sound of the first heart sound, the method further includes: If the second target heart sound valley point does not exist, the third adjustment coefficient is reduced by a third preset step value to obtain the updated third adjustment coefficient. Then, the step of adjusting the first heart sound envelope valley value based on the third adjustment coefficient is returned to be executed until the second heart sound peak point of the first heart sound is detected and there is a second target heart sound valley point within a preset time, or until the third threshold is less than the preset value. If, even if the third threshold is less than the preset value, no second target heart sound valley point is detected within a preset time after the second heart sound peak point of the first heart sound, then the current target heart beat is discarded, and other heart beats are taken as new target heart beats, and the process returns to the step of obtaining the target physiological signal of the target heart beat.

14. The heart sound analysis method as described in claim 11, characterized in that, Based on the largest peak point within the third time interval, the third valid heart sound peak point in the first heart sound is determined, including: The largest peak point within the third time interval is taken as the third heart sound peak point in the first heart sound; After detecting the third peak point of the first heart sound, does a third target heart sound valley point exist within a preset time period? The third target heart sound valley point is greater than the third threshold, and the third target heart sound valley point is out of phase with the third peak point of the first heart sound. If the third target heart sound valley point exists, then the third heart sound peak point of the first heart sound is determined as the second valid heart sound peak point in the first heart sound.

15. The heart sound analysis method as described in claim 14, characterized in that, After detecting whether a third target heart sound trough exists within a preset time period after the third heart sound peak of the first heart sound, the method further includes: If the third target heart sound valley point does not exist, the third adjustment coefficient is reduced by a third preset step value to obtain the updated third adjustment coefficient. Then, the step of adjusting the first heart sound envelope valley value based on the third adjustment coefficient is returned to be executed until the third heart sound peak point of the first heart sound is detected and a third target heart sound valley point exists within a preset time, or until the third threshold is less than the preset value. If, even if the third threshold is less than the preset value, no third target heart sound valley point is detected within a preset time after the third heart sound peak point, then the current target heart beat is discarded, and other heart beats are taken as new target heart beats, and the process returns to the step of obtaining the target physiological signal of the target heart beat.

16. The heart sound analysis method as described in claim 5, characterized in that, The step of identifying two valid heart sound peak points in the second heart sound based on the second threshold includes: Based on the peak points in the fourth time interval of the second heart sound that are greater than the second threshold, the first valid heart sound peak point in the second heart sound is identified. Based on the first valid heart sound peak point in the second heart sound, a fifth time interval is determined. Based on the peak points in the fifth time interval that are greater than the second threshold, the second valid heart sound peak point in the second heart sound is identified. The fifth time interval is the third preset time interval after the first valid heart sound peak point in the second heart sound.

17. An electronic device, characterized in that, The electronic device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the heart sound analysis method as described in any one of claims 1 to 16.

Images