Blood pressure monitoring method, smart ring, storage medium and computer program product

Abstract

The application discloses a blood pressure monitoring method, an intelligent ring, a storage medium and a computer program product, relates to the technical field of health monitoring, and the blood pressure monitoring method is applied to the intelligent ring. The intelligent ring is integrated with a heart sound signal acquisition unit and a pulse wave signal acquisition unit. The method comprises the following steps: acquiring a heart sound signal of a target heartbeat through the heart sound signal acquisition unit, and acquiring a pulse wave signal of the target heartbeat through the pulse wave signal acquisition unit; determining a pulse wave conduction time of the target heartbeat based on the heart sound signal and the pulse wave signal of the target heartbeat, and generating a blood pressure monitoring result of a wearer based on the pulse wave conduction time of the target heartbeat, wherein the blood pressure monitoring result comprises at least one of blood pressure numerical information, blood pressure classification information and risk level information. The application can improve the accuracy of the blood pressure monitoring result.

Description

Technical Field

[0001] This application relates to the field of health monitoring technology, and in particular to blood pressure monitoring methods, smart rings, storage media, and computer program products. Background Technology

[0002] Blood pressure is an important physiological parameter reflecting the functional state of the human cardiovascular system. Long-term hypertension or abnormal blood pressure fluctuations are major contributing factors to serious diseases such as stroke, coronary heart disease, heart failure, and kidney failure. Therefore, timely and accurate monitoring of one's blood pressure level is crucial for early warning of cardiovascular diseases, risk assessment, and daily health management. With increasing health awareness, the need for routine and convenient blood pressure monitoring is becoming increasingly urgent.

[0003] Currently, most smart wearable devices on the market (such as smart bracelets and smartwatches) use photoplethysmography (PPG) technology for non-invasive blood pressure monitoring. This involves illuminating the skin with a light source on the back of the device and using photoelectric sensors to receive reflected or transmitted light signals generated by changes in blood volume within blood vessels due to heartbeats. The pulse wave characteristics are then extracted to estimate blood pressure values.

[0004] However, pulse wave signals essentially reflect changes in blood volume in peripheral blood vessels, and are easily affected by factors such as the tightness of the garment, ambient light, limb movement, and individual differences in vascular elasticity. This results in a large amount of noise mixed into the signal, making it difficult to fully characterize the complex hemodynamic characteristics of the cardiovascular system, leading to a serious lack of accuracy in blood pressure monitoring results. Summary of the Invention

[0005] The main objective of this application is to provide a blood pressure monitoring method, a smart ring, a storage medium, and a computer program product, aimed at improving the accuracy of blood pressure monitoring results.

[0006] To achieve the above objectives, this application provides a blood pressure monitoring method applied to a smart ring, the smart ring integrating a heart sound signal acquisition unit and a pulse wave signal acquisition unit, the method comprising: The heart sound signal acquisition unit acquires the heart sound signal of the target heart beat, and the pulse wave signal acquisition unit acquires the pulse wave signal of the target heart beat. Based on the heart sound signal and pulse wave signal of the target heartbeat, the pulse wave conduction time of the target heartbeat is determined, and based on the pulse wave conduction time of the target heartbeat, the wearer's blood pressure monitoring results are generated, wherein the blood pressure monitoring results include at least one of blood pressure numerical information, blood pressure classification information, and risk level information.

[0007] In addition, to achieve the above objectives, this application also provides a smart ring, the smart ring comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the blood pressure monitoring method as described above.

[0008] In addition, to achieve the above objectives, this application also provides a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the blood pressure monitoring method described above.

[0009] 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 steps of the blood pressure monitoring method described above.

[0010] This application provides a blood pressure monitoring method, a smart ring, a storage medium, and a computer program product. The method is applied to a smart ring, which integrates a heart sound signal acquisition unit and a pulse wave signal acquisition unit. The method includes: acquiring the heart sound signal of a target heartbeat through the heart sound signal acquisition unit and acquiring the pulse wave signal of the target heartbeat through the pulse wave signal acquisition unit; determining the pulse wave conduction time of the target heartbeat based on the heart sound signal and pulse wave signal; and generating a blood pressure monitoring result for the wearer based on the pulse wave conduction time of the target heartbeat. The blood pressure monitoring result includes at least one of blood pressure numerical information, blood pressure classification information, and risk level information.

[0011] This application's embodiments effectively overcome the shortcomings of single pulse wave signals, such as susceptibility to interference and insufficient characterization ability, by organically fusing heart sound signals and pulse wave signals. Specifically, heart sound signals directly reflect the opening and closing of heart valves and the mechanical vibrations generated by blood flow impacting the blood vessel walls, providing key hemodynamic information such as the onset of cardiac ejection, heart sound intensity, and interval; while pulse wave signals reflect the pulsatile changes in blood volume in peripheral blood vessels. The two signals are naturally complementary in terms of time axis and physiological mechanism—there is a definite time delay from cardiac ejection to peripheral pulse wave propagation, namely the pulse transit time (PTT). This time delay is highly correlated with blood pressure levels, but a single pulse wave signal is insufficient to accurately extract the ventricular ejection initiation point. This application integrates both a heart sound signal acquisition unit and a pulse wave signal acquisition unit into a smart ring. By using the peak value of the first heart sound (S1) in the heart sound signal of the same target heart beat as the starting reference point for cardiac ejection, and combining this with the starting point of the rising limb of the pulse wave or the peak value of the pulse wave in the pulse wave signal of the same beat, the pulse wave conduction time of the target heart beat can be accurately calculated. Compared to schemes that rely solely on pulse wave signals to estimate blood pressure, the introduction of heart sound signals in this application significantly reduces the sensitivity to interference factors such as wearing tightness, ambient light, and limb movement, because heart sound signals mainly originate from internal mechanical vibrations and are less affected by external conditions.

[0012] Furthermore, this application embodiment can further utilize the rich features contained in the heart sound signal (such as heart sound intensity, splitting, murmurs, etc.) and the waveform morphology features of the pulse wave signal (such as rising slope, descending notch, amplitude, etc.) for joint analysis to construct a multi-dimensional hemodynamic parameter set. Through machine learning or regression models, these parameters can be mapped to accurate blood pressure values, while simultaneously outputting blood pressure classification information and risk level information. For example, when the heart sound signal shows increased heart sounds and the pulse wave signal shows increased peripheral vascular resistance, the abnormal trend of systolic and diastolic blood pressure can be comprehensively judged. This multimodal sensing strategy not only makes up for the limitations of a single pulse wave signal in scenarios such as peripheral vascular elasticity differences and arrhythmias, but also realizes end-to-end monitoring from cardiac pumping function to peripheral vascular resistance status, thereby significantly improving the accuracy and clinical reference value of blood pressure monitoring results. The shape of the smart ring naturally facilitates the stable acquisition of pulse wave signals from the fingertips. At the same time, the wearer only needs to press the smart ring against the precordial area to directly acquire clear and reliable heart sound signals. This method further ensures the synchronization and signal-to-noise ratio of the two signals, providing a reliable and low-cost solution to the technical problem of insufficient accuracy in blood pressure monitoring of existing smart wearable devices. 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, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a flowchart illustrating the first embodiment of the blood pressure monitoring method of this application; Figure 2 This is a flowchart illustrating the second embodiment of the blood pressure monitoring method of this application; Figure 3 A system flowchart provided for a specific embodiment of this application; Figure 4 A timing diagram comparing the waveforms of two signals provided in a specific embodiment of this application; Figure 5 A flowchart illustrating the steps of a specific embodiment of this application; Figure 6 A schematic diagram of the pressure position of the smart ring in the precordial region provided in a specific embodiment of this application; Figure 7 This is a schematic diagram of the hardware operating environment involved in the blood pressure monitoring method in this application embodiment.

[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] 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.

[0018] 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.

[0019] Currently, most smart wearable devices on the market (such as smart bracelets and smartwatches) use photoplethysmography (PPG) technology for non-invasive blood pressure monitoring. This involves illuminating the skin with a light source on the back of the device and using photoelectric sensors to receive reflected or transmitted light signals generated by changes in blood volume within blood vessels due to heartbeats. The pulse wave characteristics are then extracted to estimate blood pressure values.

[0020] However, pulse wave signals essentially reflect changes in blood volume in peripheral blood vessels, and are easily affected by factors such as the tightness of the garment, ambient light, limb movement, and individual differences in vascular elasticity. This results in a large amount of noise mixed into the signal, making it difficult to fully characterize the complex hemodynamic characteristics of the cardiovascular system, leading to a serious lack of accuracy in blood pressure monitoring results.

[0021] The main solution of this application embodiment is a blood pressure monitoring method applied to a smart ring. The smart ring integrates a heart sound signal acquisition unit and a pulse wave signal acquisition unit. The method includes: acquiring the heart sound signal of a target heart beat through the heart sound signal acquisition unit, and acquiring the pulse wave signal of the target heart beat through the pulse wave signal acquisition unit; determining the pulse wave conduction time of the target heart beat based on the heart sound signal and pulse wave signal of the target heart beat, and generating a blood pressure monitoring result for the wearer based on the pulse wave conduction time of the target heart beat. The blood pressure monitoring result includes at least one of blood pressure numerical information, blood pressure classification information, and risk level information.

[0022] This application's embodiments effectively overcome the shortcomings of single pulse wave signals, such as susceptibility to interference and insufficient characterization ability, by organically fusing heart sound signals and pulse wave signals. Specifically, heart sound signals directly reflect the opening and closing of heart valves and the mechanical vibrations generated by blood flow impacting the blood vessel walls, providing key hemodynamic information such as the onset of cardiac ejection, heart sound intensity, and interval; while pulse wave signals reflect the pulsatile changes in blood volume in peripheral blood vessels. The two signals are naturally complementary in terms of time axis and physiological mechanism—there is a definite time delay from cardiac ejection to peripheral pulse wave propagation, namely the pulse transit time (PTT). This time delay is highly correlated with blood pressure levels, but a single pulse wave signal is insufficient to accurately extract the ventricular ejection initiation point. This application integrates both a heart sound signal acquisition unit and a pulse wave signal acquisition unit into a smart ring. By using the peak value of the first heart sound (S1) in the heart sound signal of the same target heart beat as the starting reference point for cardiac ejection, and combining this with the starting point of the rising limb of the pulse wave or the peak value of the pulse wave in the pulse wave signal of the same beat, the pulse wave conduction time of the target heart beat can be accurately calculated. Compared to schemes that rely solely on pulse wave signals to estimate blood pressure, the introduction of heart sound signals in this application significantly reduces the sensitivity to interference factors such as wearing tightness, ambient light, and limb movement, because heart sound signals mainly originate from internal mechanical vibrations and are less affected by external conditions.

[0023] Furthermore, this application embodiment can further utilize the rich features contained in the heart sound signal (such as heart sound intensity, splitting, murmurs, etc.) and the waveform morphology features of the pulse wave signal (such as rising slope, descending notch, amplitude, etc.) for joint analysis to construct a multi-dimensional hemodynamic parameter set. Through machine learning or regression models, these parameters can be mapped to accurate blood pressure values, while simultaneously outputting blood pressure classification information and risk level information. For example, when the heart sound signal shows increased heart sounds and the pulse wave signal shows increased peripheral vascular resistance, the abnormal trend of systolic and diastolic blood pressure can be comprehensively judged. This multimodal sensing strategy not only makes up for the limitations of a single pulse wave signal in scenarios such as peripheral vascular elasticity differences and arrhythmias, but also realizes end-to-end monitoring from cardiac pumping function to peripheral vascular resistance status, thereby significantly improving the accuracy and clinical reference value of blood pressure monitoring results. The shape of the smart ring naturally facilitates the stable acquisition of pulse wave signals from the fingertips. At the same time, the wearer only needs to press the smart ring against the precordial area to directly acquire clear and reliable heart sound signals. This method further ensures the synchronization and signal-to-noise ratio of the two signals, providing a reliable and low-cost solution to the technical problem of insufficient accuracy in blood pressure monitoring of existing smart wearable devices.

[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 presents a blood pressure monitoring method according to a first embodiment.

[0026] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the blood pressure monitoring method of this application.

[0027] In this embodiment, the blood pressure monitoring method is applied to a smart ring, which integrates a heart sound signal acquisition unit and a pulse wave signal acquisition unit. The blood pressure monitoring method includes steps S100~S200: Step S100: The heart sound signal of the target heart beat is acquired through the heart sound signal acquisition unit, and the pulse wave signal of the target heart beat is acquired through the pulse wave signal acquisition unit. It should be noted that, in this embodiment, a smart ring refers to a ring-shaped smart device that can be worn on a finger (usually the index or middle finger), with various physiological signal sensors integrated on its inner or outer side. Compared to bracelets or watches, smart rings fit more closely and stably to the skin at the fingertips, and the fingertips have abundant arteries, which is particularly beneficial for the acquisition of high-quality pulse wave signals.

[0028] A heart sound signal acquisition unit refers to a sensor component integrated inside or on the surface of a smart ring to detect heart sound signals generated during a heartbeat. The core components of this unit are typically a high-sensitivity accelerometer, a microphone (such as a bone conduction microphone), or a piezoelectric thin-film sensor. When the wearer presses the smart ring against the precordial region (usually the left sternal border or the apex of the heart), this unit can pick up the weak vibration signals caused by the opening and closing of heart valves and the impact of blood flow on blood vessel walls through the skin, fat, and chest wall tissue, and convert them into digital heart sound signals.

[0029] The pulse wave signal acquisition unit refers to a photoplethysmography (PPG) sensor component integrated inside the smart ring (in contact with the fingertip or side) to detect changes in blood volume in the arteries of the finger (usually at the base of the finger) over time. This unit typically consists of a light-emitting diode (LED) (such as green, red, or infrared light) and a photodetector of the corresponding wavelength. When the LED illuminates the fingertip tissue, some light is absorbed by fixed tissues such as skin, muscle, and bone, while the rest is modulated by the periodically changing blood volume within the blood vessels due to heartbeats. The intensity of the reflected or transmitted light received by the photodetector then fluctuates periodically, forming a pulse wave signal after analog-to-digital conversion.

[0030] The target heartbeat refers to the physiological signal segment to be analyzed that corresponds to a complete cycle of cardiac contraction and relaxation (i.e., one cardiac cycle).

[0031] Heart sound signals refer to time-series waveforms that reflect the mechanical activity of the heart. A normal heart sound signal mainly contains two important components: the first heart sound (S1) – corresponding to the vibration produced when the mitral and tricuspid valves close, marking the beginning of ventricular systole; and the second heart sound (S2) – corresponding to the vibration produced when the aortic and pulmonary valves close, marking the beginning of ventricular diastole. The amplitude of S1 is usually higher than that of S2, and its main peak value has clear physiological significance.

[0032] A pulse wave signal is a waveform that reflects the periodic changes in arterial blood volume. A typical cardiac cycle pulse wave includes: an ascending limb (rapidly rising to the main peak, corresponding to the cardiac ejection phase), the main peak (the point of maximum systolic pressure), the dicrotic pre-wave, the dicrotic notch (a marker of aortic valve closure), and the diastolic decay segment. Among these, the position of the main peak is a key parameter for calculating pulse wave conduction time.

[0033] In this embodiment, the smart ring, through its heart sound signal acquisition unit and pulse wave signal acquisition unit, simultaneously acquires two types of signals—at least one target heartbeat—when the wearer actively presses the ring against the precordial region. The acquisition process must ensure that the two types of signals are strictly aligned on the time axis (i.e., using the same sampling clock or aligning via post-interpolation) to facilitate subsequent calculation of time delay. Since the heart sound signal originates from the mechanical activity of the heart, while the pulse wave signal originates from changes in peripheral vascular volume, there is a natural time difference between the two—the propagation delay from heart sound S1 (the start of cardiac ejection) to the peak of the fingertip pulse wave (blood reaching the fingertip) is the pulse wave conduction time (PTT). This delay is highly correlated with blood pressure levels and is the core physical basis for subsequent blood pressure estimation.

[0034] This step establishes a complete hemodynamic observation chain from cardiac pumping to peripheral vascular perfusion by simultaneously acquiring heart sound and pulse wave signals. The heart sound signal provides a precise reference for the onset of cardiac ejection, overcoming the limitation of a single pulse wave signal in determining the ventricular ejection initiation point; while the pulse wave signal provides a true time marker for blood reaching the periphery. The combination of these two signals provides a reliable signal pair for subsequent accurate calculation of pulse wave conduction time, fundamentally improving the physiological rationality and anti-interference capability of blood pressure estimation.

[0035] Step S200: Based on the heart sound signal and pulse wave signal of the target heartbeat, determine the pulse wave conduction time of the target heartbeat, and generate the wearer's blood pressure monitoring result based on the pulse wave conduction time of the target heartbeat, wherein the blood pressure monitoring result includes at least one of blood pressure numerical information, blood pressure classification information, and risk level information.

[0036] It should be noted that in this embodiment, Pulse Transit Time (PTT) refers to the time required for the pressure wave generated by the heart's ejection to propagate from the aortic valve to the peripheral measurement point (in this case, the fingertip). In this embodiment, PTT can be quantified in two common ways: one is the time interval from the peak of the main wave of heart sound S1 to the starting point of the rising limb of the pulse wave, reflecting the propagation time of the pressure wave from the heart to the initial segment of the peripheral blood vessels; the other is the time interval from the peak of the main wave of heart sound S1 to the peak of the main wave of the pulse wave, which additionally includes the propagation time of the pressure wave within the peripheral arteries from the starting point of the rising limb to the peak. Both definitions are negatively correlated with blood pressure—the higher the blood pressure, the more taut the blood vessel walls, the faster the pulse wave propagation velocity, and the shorter the PTT. Which definition is used can be determined in advance based on experimental calibration or model training; this embodiment does not limit this.

[0037] Blood pressure monitoring results refer to the structured assessment information of the wearer's current blood pressure status, ultimately output by the system based on the calculated PTT (post-traumatic stress test) and combined with individual calibration parameters or pre-trained models. This result includes at least one of the following three information formats: Blood pressure numerical information: such as the specific values ​​of systolic / diastolic blood pressure, which can be directly used for clinical records or alarms.

[0038] Blood pressure classification information: Blood pressure is classified into normal, slightly high, grade 1 hypertension, grade 2 hypertension, etc. according to internationally accepted standards, which makes it easy for users to understand intuitively.

[0039] Risk level information: This function associates abnormal blood pressure with the risk of cardiovascular events and outputs low, medium, and high risk levels, which can be used for early warning or health management recommendations.

[0040] The core of this step lies in leveraging the physiological negative correlation between PTT (post-traumatic heart rate) and blood pressure, converting the measured PTT into an estimated blood pressure value through a regression model or lookup table method. Specifically: during system initialization or periodic calibration, one or more sets of real blood pressure values ​​are obtained using a reference blood pressure monitor (such as a cuff-type monitor), and the corresponding PTT values ​​are simultaneously collected to construct a "PTT-blood pressure" mapping function (usually a linear or nonlinear regression model). During normal monitoring, for each target heartbeat's PTT detected, the system substitutes it into this mapping function to calculate the current blood pressure value. Furthermore, the blood pressure value is mapped to a preset grading interval to obtain blood pressure grading information; then, combined with the grading results and the user's historical data, a comprehensive assessment of the cardiovascular risk level is performed.

[0041] This embodiment effectively overcomes the shortcomings of single pulse wave signals, such as susceptibility to interference and insufficient characterization ability, by organically fusing heart sound signals and pulse wave signals. Heart sound signals directly reflect the opening and closing of heart valves and the onset of blood ejection, and are minimally affected by external wearing conditions (tightness, ambient light), exhibiting significantly better stability than solutions relying solely on PPG. Simultaneously, PTT itself is a comprehensive measure of the cardiovascular system's elasticity. Compared to directly extracting features from pulse wave morphology (such as rise slope and amplitude), PTT has more explicit physiological significance and lower individual variability. Therefore, blood pressure monitoring results generated based on PTT are not only more accurate but also more robust across different users and scenarios. This solution also supports the output of blood pressure classification and risk level information, making the monitoring results more readable and actionable for ordinary users, truly achieving a seamless connection from "data collection" to "health decision-making."

[0042] This embodiment effectively overcomes the shortcomings of a single pulse wave signal, such as susceptibility to interference and insufficient characterization ability, by organically fusing heart sound signals and pulse wave signals. Specifically, heart sound signals directly reflect the opening and closing of heart valves and the mechanical vibrations generated by blood flow impacting the blood vessel walls, providing key hemodynamic information such as the onset of cardiac ejection, heart sound intensity, and interval; while pulse wave signals reflect the pulsatile changes in blood volume in peripheral blood vessels. The two signals are naturally complementary in terms of time axis and physiological mechanism—there is a definite time delay from cardiac ejection to peripheral pulse wave propagation, namely the pulse transit time (PTT). This time delay is highly correlated with blood pressure levels, but a single pulse wave signal is insufficient to accurately extract the onset of ventricular ejection. This embodiment integrates both a heart sound signal acquisition unit and a pulse wave signal acquisition unit into a smart ring. It uses the peak value of the first heart sound (S1) in the heart sound signal of the same target heart beat as the starting reference point for cardiac ejection. Combined with the starting point of the rising limb of the pulse wave or the peak value of the pulse wave in the pulse wave signal of the same beat, the pulse wave conduction time of the target heart beat can be accurately calculated. Compared to schemes that rely solely on pulse wave signals to estimate blood pressure, this embodiment, by introducing heart sound signals, significantly reduces sensitivity to interference factors such as wearing tightness, ambient light, and limb movement, because heart sound signals mainly originate from internal mechanical vibrations and are less affected by external conditions.

[0043] Furthermore, this embodiment can further utilize the rich features contained in the heart sound signal (such as heart sound intensity, splitting, murmurs, etc.) and the waveform morphology features of the pulse wave signal (such as rising slope, descending notch, amplitude, etc.) for joint analysis to construct a multi-dimensional hemodynamic parameter set. Through machine learning or regression models, these parameters can be mapped to accurate blood pressure values, while simultaneously outputting blood pressure classification information and risk level information. For example, when the heart sound signal shows increased heart sounds and the pulse wave signal shows increased peripheral vascular resistance, the abnormal trend of systolic and diastolic blood pressure can be comprehensively judged. This multimodal sensing strategy not only compensates for the limitations of a single pulse wave signal in scenarios such as peripheral vascular elasticity differences and arrhythmias, but also realizes end-to-end monitoring from cardiac pumping function to peripheral vascular resistance status, thereby significantly improving the accuracy and clinical reference value of blood pressure monitoring results. The shape of the smart ring naturally facilitates the stable acquisition of pulse wave signals from the fingertips. At the same time, the wearer only needs to press the smart ring against the precordial area to directly acquire clear and reliable heart sound signals. This method further ensures the synchronization and signal-to-noise ratio of the two signals, providing a reliable and low-cost solution to the technical problem of insufficient accuracy in blood pressure monitoring of existing smart wearable devices.

[0044] In one feasible implementation, the smart ring also integrates a pressure sensor, and prior to step S100, the blood pressure monitoring method further includes steps A10-A20: Step A10: The contact pressure between the heart sound signal acquisition unit and the wearer's precordial region is detected by the pressure sensor; It should be noted that, in this embodiment, the pressure sensor refers to a miniature force-sensitive element (such as a piezoresistive, capacitive, or piezoelectric sensor) integrated inside the smart ring, adjacent to or integrally formed with the heart sound signal acquisition unit. Its function is to measure the normal contact force (usually in Newtons or grams) between the heart sound signal acquisition unit and the skin in real time when the wearer presses the smart ring against their chest. Contact pressure refers to the instantaneous pressure or total pressure value detected by the sensor between the heart sound signal acquisition unit and the wearer's precordial region.

[0045] This step is a preliminary assurance procedure to ensure the quality of heart sound signal acquisition. Heart sound signals are weak, low-frequency mechanical vibrations, with amplitudes typically in the millivolt or even microvolt range. They are highly susceptible to poor coupling media (air gaps, clothing barriers) or insufficient contact force leading to relative slippage between the sensor and the skin. When the contact pressure is too low, the heart sound signal acquisition unit cannot effectively couple the vibrational energy of the subcutaneous tissue, resulting in a very low signal-to-noise ratio of the acquired heart sound signal, or even complete submersion in ambient noise, causing subsequent PTT calculations to fail. Conversely, excessive pressure, while enhancing coupling, may compress chest wall tissue, affect breathing, or cause discomfort. This step uses a pressure sensor to continuously monitor the contact pressure value, providing an objective and quantitative basis for subsequent condition assessments.

[0046] Step A20: When the contact pressure is greater than a preset threshold, the step of acquiring the heart sound signal of the target heartbeat through the heart sound signal acquisition unit is performed.

[0047] It should be noted that in this embodiment, the preset threshold is a pre-calibrated minimum effective contact pressure value, the value of which can be experimentally determined: the minimum pressure value required for the heart sound signal acquisition unit to stably detect the S1 / S2 peak value and the signal-to-noise ratio to be greater than the preset quality threshold under typical human chest curvature and fat thickness conditions. This threshold is usually set in the range of 1 to 3 Newtons (approximately 100 to 300 grams of force), and can be adjusted according to the sensor sensitivity and the target population.

[0048] This step integrates contact quality control with signal acquisition start / stop. The system only initiates heart sound signal acquisition and recording when the contact pressure meets the preset minimum reliable coupling condition; otherwise, the system can output prompts (such as vibration or indicator lights) to guide the wearer to appropriately increase the pressure, or automatically pause acquisition and wait for the pressure to reach the target level. This design avoids acquiring low-quality signals and wasting computational resources in invalid contact states, and prevents incorrect PTT calculations due to poor signal quality, thus avoiding misleading blood pressure monitoring results.

[0049] This implementation significantly improves the effective acquisition rate and signal-to-noise ratio of heart sound signals by introducing a pressure detection and threshold judgment mechanism before acquiring heart sound signals, thus ensuring the reliability of subsequent PTT calculations from the source. This is especially important for smart rings that require active user pressure—different users have significantly different pressing habits and pressure levels. This mechanism can adaptively filter out valid data segments, not only improving the system's automation level but also enhancing the user experience.

[0050] In one feasible implementation, the step S200 above, which determines the pulse wave conduction time of the target heartbeat based on the heart sound signal and pulse wave signal of the target heartbeat, includes steps S210 to S230: Step S210: Extract the peak time of the first heart sound of the target heart beat from the heart sound signal of the target heart beat, wherein the peak time of the first heart sound of the target heart beat is the time corresponding to the main peak point of the first heart sound in the heart sound signal of the target heart beat; It should be noted that in this embodiment, the first heart sound (S1) is the waveform component with the largest amplitude and lowest frequency in the heart sound signal, corresponding to the vibration generated by the closure of the mitral and tricuspid valves during ventricular systole. The peak point of the first heart sound refers to the moment when the positive phase amplitude in the S1 waveform reaches a local maximum value, which is usually considered as the acoustic marker of the start of ventricular ejection. The peak time of the first heart sound refers to extracting the coordinates of this peak point on the signal time axis (usually expressed in milliseconds or sampling point indices), which serves as the cardiac end-time reference point for subsequent calculation of PTT.

[0051] The extraction method for the peak time of the first heart sound can employ an adaptive threshold peak detection algorithm: First, bandpass filtering is applied to the heart sound signal to suppress low-frequency interference such as respiration and friction. Then, the upper envelope of the signal is calculated (e.g., through Hilbert transform or moving window maximum value). Within a preset time window, the maximum value of the envelope is found, and the corresponding time is the peak time of the first heart sound. To improve robustness, template matching or a dual threshold strategy (coarse detection followed by precise localization) can also be used.

[0052] Step S220: Extract the peak time of the target heartbeat from the pulse wave signal of the target heartbeat, wherein the peak time of the target heartbeat is the time corresponding to the main peak point of the pulse wave in the pulse wave signal of the target heartbeat. It should be noted that, in this embodiment, the main peak point of the pulse wave refers to the positive peak point with the highest amplitude in the pulse wave waveform of each cardiac cycle, corresponding to the moment of maximum dilation of peripheral arteries during cardiac systole. The pulse wave peak time is the position of this peak point on the time axis.

[0053] Methods for extracting the peak time of a pulse wave typically include: low-pass filtering the pulse wave signal to remove high-frequency noise; and using the difference method or slope method to detect the zero-crossing point (the point where the first derivative of the signal changes from positive to negative). This point is the main peak point of the pulse wave, and the corresponding time is the peak time of the pulse wave. Alternatively, a more robust baseline correction and peak sorting algorithm can be used to find local maxima within each cardiac cycle segment, thereby determining the main peak point of the pulse wave and thus the peak time.

[0054] Step S230: Determine the pulse wave conduction time of the target heartbeat based on the peak time of the first heart sound and the peak time of the pulse wave.

[0055] It should be noted that, in this embodiment, the pulse wave conduction time (PTT) is obtained by calculating the difference between the peak pulse wave time and the peak time of the first heart sound in the same heart beat, i.e., PTT = t_PPG_peak - t_S1_peak, where t_PPG_peak is the peak pulse wave time and t_S1_peak is the peak time of the first heart sound.

[0056] This definition selects the time interval from the peak of the S1 main wave to the peak of the pulse wave main wave. The physiological basis for this is that the peak of the S1 main wave precisely corresponds to the moment the mitral valve closes, while the peak of the pulse wave main wave corresponds to the moment when the pressure reaches its peak after the blood reaches the fingertip. The difference between the two includes the entire propagation time of the pressure wave from the aortic root to the fingertip artery, as well as part of the pressure rise time. Experiments show that this PTT (post-pulse time) has a significant negative correlation with both systolic and diastolic blood pressure, and that using the peak of the main wave is more stable and robust than using only the pulse wave rising limb (which is easily affected by baseline drift and noise).

[0057] This implementation achieves accurate PTT calculation by extracting two physiological anchor points: the S1 peak time (i.e., the peak time of the first heart sound) and the pulse wave peak time (i.e., the peak time of the pulse wave). The advantages of this combination are: S1 peak point detection is minimally affected by peripheral environmental interference (heart sounds originate from within the body), providing a stable reference for the onset of ejection; pulse wave peak point detection is mature and computationally inefficient. The combination of these two methods makes PTT calculation both accurate and efficient, suitable for real-time operation on embedded platforms of smart rings with limited computing resources.

[0058] In one feasible implementation, the step of generating the wearer's blood pressure monitoring result based on the pulse wave conduction time of the target heartbeat in step S200 above includes steps S240~S260: Step S240: Extract features from the heart sound signal of the target heart beat to obtain the heart sound feature data of the target heart beat; It should be noted that, in this embodiment, heart sound feature data refers to a set of digital features extracted from heart sound signals that can quantitatively describe the mechanical activity characteristics of the heart and are related to blood pressure levels. Typical heart sound features may include, but are not limited to: S1 amplitude (reflecting ventricular contractility), the time interval from S1 to S2 (reflecting left ventricular ejection time), S2 amplitude (reflecting aortic valve closure strength), heart sound intensity ratio (S1 / S2 ratio, reflecting cardiac reserve), and heart sound envelope morphological features (such as rise slope and decay time constant). These features can be extracted through time domain analysis (peak detection, area integration) or frequency domain analysis (spectral centroid, specific frequency band energy).

[0059] Step S250: Extract features from the pulse wave signal of the target heartbeat to obtain the pulse wave feature data of the target heartbeat. It should be noted that, in this embodiment, pulse wave characteristic data refers to quantitative features extracted from the pulse wave signal that are related to changes in vascular elasticity, peripheral resistance, and blood volume. Common features include: ascending limb slope (reflecting cardiac output and arterial stiffness), main peak amplitude (reflecting maximum systolic blood volume), dicrotic notch height (reflecting the aortic valve closure time), diastolic decay rate (reflecting peripheral resistance), and waveform width indices (such as half-width at half-maximum). These features can be used individually or in combination to characterize the peripheral vascular status.

[0060] Step S260: Input the heart sound feature data, pulse wave feature data and pulse wave conduction time of the target heartbeat into the pre-trained blood pressure monitoring model, and output the wearer's blood pressure monitoring results. The blood pressure monitoring model is trained in advance using training samples with heart sound feature data, pulse wave feature data, and pulse wave conduction time as sample features and blood pressure monitoring results as sample labels.

[0061] It should be noted that in this embodiment, the blood pressure monitoring model is a regression / classification model based on machine learning or deep learning. The input to this model is the extracted heart sound feature data, pulse wave feature data, and feature vectors in three dimensions (PTT), and the output is the blood pressure monitoring result. Common model architectures include: linear regression / ridge regression (suitable for scenarios where PTT is the primary feature and the number of features is small), random forest or gradient boosting tree (capable of automatically handling feature interactions and nonlinear relationships, with good interpretability), and lightweight neural networks (suitable for edge deployment and capable of learning deep patterns).

[0062] Training Process: In the offline phase, heart sound and pulse wave signals of the same heartbeat are collected from a sufficient number of subjects in different states (resting, post-exercise, different time periods), and the actual blood pressure values ​​are measured simultaneously using reference standards (such as a mercury sphygmomanometer or oscillometric cuff). The heart sound feature data, pulse wave feature data, and PTT extracted from each heartbeat are used as sample features X, and the corresponding actual blood pressure value, blood pressure classification, and / or risk level are used as sample labels Y, forming a training set. The model parameters are trained by minimizing the prediction error (such as mean squared error or cross-entropy loss). After the model parameters stabilize, personalized fine-tuning can be performed for each wearer to make the model prediction results more closely match the wearer's individual situation.

[0063] During the inference phase (step S260), for each target heartbeat, the system concatenates its real-time extracted heart sound feature data, pulse wave feature data, and PTT into a feature vector, which is then input into a pre-trained blood pressure monitoring model. After forward calculation, the model directly outputs the blood pressure value, blood pressure classification, and / or risk level as the wearer's blood pressure monitoring result.

[0064] This implementation significantly improves the accuracy and individual adaptability of blood pressure estimation by fusing multimodal feature data (heart sound feature data, pulse wave feature data, and PTT) and employing a machine learning model for blood pressure monitoring. While a single PTT shows a linear relationship with blood pressure, factors such as individual differences in vascular elasticity and heart rate variations can introduce biases. Adding heart sound feature data (such as heart sound intensity and interval) can partially characterize the heart's pumping function, and adding pulse wave feature data can quantify peripheral vascular resistance characteristics. The complementary nature of these three elements allows the blood pressure monitoring model to more accurately predict the wearer's true blood pressure, significantly improving robustness to complex physiological scenarios (such as arrhythmias and vascular calcification), ultimately outputting blood pressure monitoring results that are closer to clinically accurate values.

[0065] Based on the above embodiments, this application proposes a second embodiment of a blood pressure monitoring method.

[0066] 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.

[0067] Please refer to Figure 2 , Figure 2 This is a flowchart illustrating the second embodiment of the blood pressure monitoring method of this application.

[0068] In this embodiment, the step of generating the wearer's blood pressure monitoring result based on the pulse wave conduction time of the target heartbeat in step S200 above may include steps S270~S280: Step S270: Determine the valid target heartbeat from multiple target heartbeats; It should be noted that, in this embodiment, a valid target heartbeat refers to a heartbeat that has been screened and whose signal quality and physiological consistency meet predetermined standards. During continuous blood pressure monitoring, due to limb movement, unstable instantaneous compressions, respiratory interference, or arrhythmias (such as premature beats), the heart sound or pulse wave signals of some target heartbeats may be severely distorted. Directly using these target heartbeats for blood pressure monitoring will introduce significant errors. Therefore, screening out a valid subset from multiple consecutive target heartbeats and then performing subsequent processing based on the valid target heartbeats is a key step in improving the stability and reliability of blood pressure monitoring results.

[0069] The following provides two implementation methods for determining effective target heartbeats that can be used in combination.

[0070] In a first feasible implementation, step S270 above includes steps S271 to S272: Step S271: Standardize the pulse wave conduction time of multiple target heartbeats to obtain the standard score of pulse wave conduction time for each target heartbeat. It should be noted that, in this embodiment, standardization refers to the process of converting a set of PTTs into a standardized score (Z-score) with zero mean and unit variance. The pulse wave conduction time standard score indicates the degree to which the PTT of that heartbeat deviates from the average level, and the unit is a multiple of the standard deviation.

[0071] Step S272: Target heartbeats whose pulse wave conduction time standard score is less than a preset standard score threshold are determined as valid target heartbeats.

[0072] It should be noted that in this embodiment, the preset standard score threshold is an empirical parameter, typically ranging from 2.0 to 3.0 (corresponding to a 99% or 99.7% confidence interval). When the absolute value of the PTT standard score of a heartbeat is less than this threshold, it indicates that its PTT value falls within the core region of a large-sample normal distribution, belonging to the statistically "normal" range, and its corresponding hemodynamic state is relatively stable. Conversely, if the PTT standard score is too high (e.g., exceeding 3), it may be an extreme value caused by abnormal heartbeats (premature beats, atrial fibrillation), motion artifacts, or signal loss, and should be considered invalid and discarded.

[0073] This implementation method, based on the principle of statistical outlier detection, can efficiently eliminate abnormal PTT values ​​caused by transient interference or ectopic heartbeats. The advantages of this method are that it does not rely on additional physiological reference signals, is computationally simple, and offers strong real-time performance, making it particularly suitable for operation in embedded devices. After removing outliers, the average PTT calculated based on the effective target heartbeat will be more stable, and blood pressure monitoring results will not fluctuate drastically due to occasional abnormal heartbeats, thereby improving the smoothness of the monitoring curve and clinical reliability.

[0074] In a second feasible implementation, step S270 above includes steps S273 to S274: Step S273: Perform cross-correlation analysis on the heart sound signals and pulse wave signals of multiple target heart beats to obtain the signal correlation coefficient of each target heart beat, wherein the signal correlation coefficient is the correlation coefficient between the heart sound signal and the pulse wave signal; It should be noted that, in this embodiment, cross-correlation analysis refers to calculating the Pearson correlation coefficient between aligned heart sound signal segments and pulse wave signal segments within the same cardiac beat. The signal correlation coefficient measures the degree of linear consistency between two signals within the same cardiac beat cycle. For true, well-synchronized heart sounds and pulse waves, due to the regularity of the heart's pumping cycle, the two should show a high positive correlation; if a signal is severely contaminated by motion artifacts or noise, the correlation coefficient will decrease significantly.

[0075] Step S274: If the signal correlation coefficient is greater than the preset correlation coefficient threshold, it is determined as a valid target heartbeat.

[0076] It should be noted that in this embodiment, the preset correlation coefficient threshold is a lower limit value calibrated experimentally, typically set to 0.6~0.8. When the signal correlation coefficient is greater than this threshold, it indicates that the two signals have good consistency in the time domain, representing that the quality of the acquired heart sound and pulse wave signals is acceptable and the time relationship between the two signals is reliable, thus the calculated PTT has physiological significance. Conversely, if the correlation coefficient is too low, it indicates that at least one signal is distorted, and the target heartbeat should be discarded.

[0077] This implementation method directly evaluates the synchronization quality of heart sound and pulse wave signals within the same heartbeat through cross-correlation analysis. Compared to the aforementioned methods based solely on PTT statistics, this method is better able to detect distortions in the signal waveform itself—for example, if the heart sound signal contains murmurs or the pulse wave is elongated by motion artifacts, even if the PTT value does not exceed the statistical range, its physiological rationality has been lost. The correlation coefficient can effectively identify and filter these distortions. This method requires high-quality hardware acquisition, but the selected effective heartbeats have higher physiological fidelity, making it particularly suitable for improving the accuracy of blood pressure monitoring in motion or complex noise environments.

[0078] Step S280: Based on the pulse wave conduction time of the effective target heartbeat, generate the wearer's blood pressure monitoring results.

[0079] It should be noted that in this embodiment, the set of effective target heartbeats selected in step S270 is considered to represent the wearer's current stable and true cardiovascular state. Based on these effective target heartbeats, the PTT (post-traumatic stress test) for blood pressure monitoring can be generated using two strategies: first, calculate the median or mean of the effective heartbeat PTT and then substitute it into the aforementioned PTT-blood pressure mapping function to obtain the blood pressure value; second, calculate the PTT independently for each effective heartbeat and input it into the blood pressure monitoring model, then smooth the blood pressure value output by the model in the time domain (e.g., by moving average). After determining the blood pressure value, the blood pressure classification can be determined by combining it with blood pressure classification standards, and further, the wearer's user history data and personal health information can be combined to comprehensively assess the cardiovascular risk level.

[0080] In one feasible implementation, step S280 above includes steps S281 to S283: Step S281: Extract features from the heart sound signals of the effective target heart beats to obtain the heart sound feature data of the effective target heart beats; Step S282: Extract features from the pulse wave signal of the effective target heartbeat to obtain the pulse wave feature data of the effective target heartbeat; Step S283: Input the heart sound feature data, pulse wave feature data and pulse wave conduction time of the effective target heartbeat into the pre-trained blood pressure monitoring model, and output the wearer's blood pressure monitoring results; The blood pressure monitoring model is trained in advance using training samples with heart sound feature data, pulse wave feature data, and pulse wave conduction time as sample features and blood pressure monitoring results as sample labels.

[0081] It should be noted that this implementation method is basically the same as steps S240-S260 in the first embodiment in terms of technical principles and operational details. The only difference is that this implementation method only processes effective target heartbeats, not all heartbeats. Therefore, the definitions and descriptions of heart sound characteristic data, pulse wave characteristic data, blood pressure monitoring models, and training processes can be directly referred to the preceding text.

[0082] This implementation method limits feature extraction and model inference to the effective set of heartbeats, eliminating the interference of abnormal or low-quality heartbeats on blood pressure estimation. This results in higher accuracy and repeatability of the final blood pressure monitoring results, making it particularly suitable for long-term continuous monitoring scenarios. It can effectively avoid accidental false alarms caused by user momentary actions or poor contact.

[0083] This embodiment significantly improves the robustness of multi-heartbeat fusion processing by introducing an effective target heartbeat screening mechanism before generating blood pressure monitoring results. Whether using a PTT statistical outlier detection method or a quality assessment method based on heart sound-pulse wave cross-correlation analysis, the core idea is to extract the signal segments that best represent the true physiological state from the raw data stream and eliminate contaminated data. This reduces unnecessary interference to downstream models and lowers the false alarm rate, enabling the smart ring to output stable and reliable blood pressure monitoring results even in dynamic, uncontrolled real-world usage environments, further enhancing the practical value and clinical reliability of this technical solution.

[0084] To facilitate understanding of the above embodiments / implementations of this application, a specific embodiment is provided: In this specific implementation, the smart ring integrates the following core components: Bone conduction sensor (i.e., heart sound signal acquisition unit): Located on the fingertip side of the ring. Unlike traditional air conduction microphones, this specific embodiment uses a bone conduction sensor, specifically designed to acquire PCG (Phonocardiogram) signals (i.e., heart sound signals) transmitted through the bones. This sensor is sensitive to low-frequency mechanical vibrations, effectively filtering out ambient noise transmitted through the air, and clearly capturing the first heart sound (S1) and the second heart sound (S2).

[0085] PPG (Photoplethysmography) sensor: Includes a light-emitting diode and a photodetector, and is placed on the side or base of the finger of the ring to collect PPG signals (i.e., pulse wave signals) from the fingertip.

[0086] Clock Module: To ensure the accuracy of blood pressure calculation, this specific embodiment introduces a clock module with sub-millisecond precision. This clock module provides a unified time reference for signal acquisition throughout the system, resolving the time synchronization issue between PCG and PPG signals caused by differences in sampling rates and transmission paths.

[0087] Control module: includes analog front-end, analog-to-digital converter and microprocessor, used to execute the algorithm flow of subsequent blood pressure monitoring.

[0088] Communication module: Used to transmit blood pressure monitoring results to external terminals (such as mobile phones or watches). like Figure 3 As shown, in this specific embodiment, after acquiring the PPG and PCG signals based on the aforementioned sensors, the first step is to perform time synchronization processing on the acquired PPG and PCG signals to ensure that the two signals are strictly aligned on the time axis; then, feature extraction is performed on the synchronized signals, such as... Figure 4 As shown, the peak point of the first heart sound S1 is identified and located from the PCG signal, and the peak point P of the pulse wave within the same cardiac cycle is identified from the PPG signal. Then, based on the extracted feature points, the time difference between the peak point of S1 and point P is calculated to obtain the pulse wave conduction time (PTT). After that, an artifact suppression step is performed. Through phase matching algorithm and outlier removal strategy, abnormal heartbeat cycles caused by user movement or poor contact are identified and removed, retaining high-quality PTT and feature data. Then, the filtered PTT and extracted feature data are input into a pre-trained blood pressure model (i.e., blood pressure monitoring model) to predict the wearer's blood pressure value or risk level, thus completing a complete monitoring process.

[0089] In this specific embodiment, the specific procedure for blood pressure monitoring is as follows: Figure 5 As shown: Step S1: Signal acquisition and pressure feedback; like Figure 6 As shown, users can place the finger wearing the smart ring (such as the index finger) firmly against the precordial region. Specifically, the smart ring can be pressed at position 1 (sternum), position 2 (left third rib), or position 3 (between positions 1 and 2). The bone conduction sensor collects PCG signals through bone conduction, while the PPG sensor simultaneously collects PPG signals from the fingertip.

[0090] Preferred Solution: To ensure signal acquisition quality, the smart ring also integrates a pressure sensor. This sensor monitors the contact pressure between the bone conduction sensor and the wearer's precordial region in real time. When the contact pressure is less than a preset threshold (e.g., 1-3 Newtons), the system prompts the user to increase the pressure via an app or vibration to ensure effective coupling of the bone conduction signal. Conversely, when the contact pressure exceeds the preset threshold, the system indicates that blood pressure monitoring requirements have been met and no further pressure increase is necessary. Furthermore, when the contact pressure exceeds a preset upper pressure limit, the system alerts the user that excessive pressure may interfere with accurate PCG signal acquisition and requires a reduction in pressure. The preset upper pressure limit is greater than the preset threshold.

[0091] Step S2: High-precision time synchronization processing Since the sampling rates of PCG and PPG signals are usually different (e.g., PCG has a higher sampling rate and PPG has a lower sampling rate), and the two signals have an inherent delay in the transmission link, this embodiment implements the following synchronization mechanism: Unified labeling: Using a clock module, each sampling point is timestamped the instant the sensor acquires the sampling data.

[0092] Data alignment: Upsample the PPG signal with a lower sampling rate to make it have the same sampling rate as the PCG signal.

[0093] Inherent Delay Calibration: Before the product leaves the factory, the inherent delay of the two data receiving clocks is calibrated and adjusted using an external standard signal source. By calculating calibration parameters, time deviations caused by differences in the transmission link are eliminated, ensuring absolute alignment of the two signals on the time axis.

[0094] Step S3: Feature point extraction and PTT calculation; like Figure 6 As shown, after filtering and denoising the synchronized PCG and PPG signals, feature extraction is used to determine the main peak point S1 of the first heart sound and the main peak point P of the pulse wave for the same beat weight. The time interval between the two is PTT. PCG Feature Extraction: Identify the peak time of the first heart sound S1 (i.e., the peak time of the first heart sound) from the PCG signal. In this embodiment, the S1 peak is selected as the starting reference point for the mechanical contraction and pumping of blood by the heart. This is more accurate than the R wave peak of the electrocardiogram in reflecting the actual time when blood is ejected into the aorta, thereby eliminating the error caused by the pre-ejection phase.

[0095] PPG feature extraction: Identify the peak time of the pulse wave within the same heartbeat cycle (i.e., the peak time of the pulse wave) from the PPG signal.

[0096] PTT calculation: Calculate the time difference between the two to obtain the pulse wave conduction time based on PCG and PPG.

[0097] Step S4: Phase matching and artifact suppression; To improve the resistance to motion interference, this embodiment introduces a combined PCG-PPG phase matching algorithm: Cross-correlation analysis: Cross-correlation analysis is performed on the PCG envelope and PPG waveform of multiple consecutive heartbeat cycles to calculate the signal correlation coefficient of each heartbeat cycle, and heartbeat cycles with excessively low signal correlation coefficients are eliminated.

[0098] Anomaly removal: When the PTT of a certain heartbeat cycle deviates significantly from the average PTT (e.g., more than 3 times the standard deviation), or when the amplitude of the PCG / PPG signal changes abruptly, the heartbeat cycle is determined to be motion artifact interference and is removed.

[0099] Auxiliary verification: The physiological time constraint relationship between the second heart sound and the PPG diphthous wave is used to further verify the correctness of the feature point matching.

[0100] Step S5: Generate blood pressure monitoring results.

[0101] The calculated PTT, along with the extracted PCG feature data (i.e., heart sound feature data) and PPG feature data (i.e., pulse wave feature data), are input into a pre-trained blood pressure monitoring model (such as support vector regression or neural network). The model outputs the user's blood pressure value, blood pressure classification, and / or risk level, where the blood pressure value includes both systolic and diastolic blood pressure.

[0102] This specific embodiment uses a bone conduction sensor to replace the traditional ECG (Electrocardiogram) electrode and uses the main wave peak point of S1 as a reference point to replace the R wave peak point of ECG, thus eliminating the systematic error of the pre-ejection stage from the physical source and significantly improving the accuracy of blood pressure monitoring results.

[0103] In addition, this application also provides a smart ring, 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, which are executed by the at least one processor to enable the at least one processor to perform the blood pressure monitoring method in the above embodiments.

[0104] The following is for reference. Figure 7 The diagram shows a structural schematic suitable for implementing the smart ring of the present application embodiments. Figure 7 The smart ring shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of this application.

[0105] like Figure 7 As shown, the smart ring may include a processing device 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 smart ring. The processing device 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: an input device 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display, speaker, vibrator, etc.; a storage device 1003 including, for example, magnetic tape, hard disk, etc.; and a communication device 1009. The communication device 1009 allows the smart ring to communicate wirelessly or wiredly with other devices to exchange data. Although the diagram shows a smart ring 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.

[0106] 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.

[0107] The smart ring provided in this application, employing the blood pressure monitoring method described in the above embodiments, can improve the accuracy of blood pressure monitoring results. Compared with the prior art, the beneficial effects of the smart ring provided in this application are the same as those of the blood pressure monitoring method provided in the above embodiments, and other technical features of the smart ring are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0108] 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.

[0109] In addition, this application also provides a storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the blood pressure monitoring method in the above embodiments.

[0110] 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 with 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.

[0111] The aforementioned storage medium may be included in the smart ring; or it may exist independently and not be assembled into the smart ring.

[0112] The aforementioned storage medium carries one or more programs. When the aforementioned one or more programs are executed by the smart ring, the smart ring, which integrates a heart sound signal acquisition unit and a pulse wave signal acquisition unit, acquires the heart sound signal of the target heartbeat through the heart sound signal acquisition unit and the pulse wave signal acquisition unit; based on the heart sound signal and pulse wave signal of the target heartbeat, it determines the pulse wave conduction time of the target heartbeat and generates the wearer's blood pressure monitoring results based on the pulse wave conduction time of the target heartbeat. The blood pressure monitoring results include at least one of blood pressure numerical information, blood pressure classification information, and risk level information.

[0113] 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 local area networks (LANs) or wide area networks (WANs), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0114] 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.

[0115] 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.

[0116] The storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described blood pressure monitoring method, thereby improving the accuracy of blood pressure monitoring results. Compared with the prior art, the beneficial effects of the medium provided in this application are the same as those of the blood pressure monitoring method provided in the above embodiments, and will not be repeated here.

[0117] In addition, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the blood pressure monitoring method described above.

[0118] The computer program product provided in this application can improve the accuracy of blood pressure monitoring results. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the blood pressure monitoring method provided in the above embodiments, and will not be repeated here.

[0119] 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.

[0120] 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 monitoring blood pressure, characterized in that, The method is applied to a smart ring, which integrates a heart sound signal acquisition unit and a pulse wave signal acquisition unit. The method includes: The heart sound signal acquisition unit acquires the heart sound signal of the target heart beat, and the pulse wave signal acquisition unit acquires the pulse wave signal of the target heart beat. Based on the heart sound signal and pulse wave signal of the target heartbeat, the pulse wave conduction time of the target heartbeat is determined, and based on the pulse wave conduction time of the target heartbeat, the wearer's blood pressure monitoring results are generated, wherein the blood pressure monitoring results include at least one of blood pressure numerical information, blood pressure classification information, and risk level information.

2. The blood pressure monitoring method as described in claim 1, characterized in that, The smart ring also integrates a pressure sensor. Before the step of acquiring the heart sound signal of the target heartbeat through the heart sound signal acquisition unit, the method further includes: The pressure sensor detects the contact pressure between the heart sound signal acquisition unit and the wearer's precordial region. When the contact pressure is greater than a preset threshold, the step of acquiring the heart sound signal of the target heartbeat through the heart sound signal acquisition unit is performed.

3. The blood pressure monitoring method as described in claim 1, characterized in that, The step of determining the pulse wave conduction time of the target heartbeat based on the heart sound signal and pulse wave signal of the target heartbeat includes: The peak time of the first heart sound of the target heart beat is extracted from the heart sound signal of the target heart beat. The peak time of the first heart sound of the target heart beat is the time corresponding to the main peak point of the first heart sound in the heart sound signal of the target heart beat. The peak time of the target heartbeat is extracted from the pulse wave signal of the target heartbeat. The peak time of the target heartbeat is the time corresponding to the main peak point of the pulse wave in the pulse wave signal of the target heartbeat. The pulse wave conduction time of the target heartbeat is determined based on the peak time of the first heart sound and the peak time of the pulse wave.

4. The blood pressure monitoring method as described in claim 1, characterized in that, The step of generating the wearer's blood pressure monitoring results based on the pulse wave conduction time of the target heartbeat includes: Identify the effective target heartbeat from multiple target heartbeats; Based on the pulse wave conduction time of the effective target heartbeat, the wearer's blood pressure monitoring results are generated.

5. The blood pressure monitoring method as described in claim 4, characterized in that, The step of determining the valid target heartbeat from multiple target heartbeats includes: The pulse wave conduction time of multiple target heartbeats is standardized to obtain the standard score of pulse wave conduction time for each target heartbeat. A target heartbeat whose pulse wave conduction time standard score is less than a preset standard score threshold is determined as a valid target heartbeat.

6. The blood pressure monitoring method as described in claim 4, characterized in that, The step of determining the valid target heartbeat from multiple target heartbeats includes: Cross-correlation analysis was performed on the heart sound signals and pulse wave signals of multiple target heart beats to obtain the signal correlation coefficient of each target heart beat, wherein the signal correlation coefficient is the correlation coefficient between the heart sound signal and the pulse wave signal; Signals with a correlation coefficient greater than a preset correlation coefficient threshold are identified as valid target heartbeats.

7. The blood pressure monitoring method according to any one of claims 4 to 6, characterized in that, The step of generating the wearer's blood pressure monitoring results based on the pulse wave conduction time of the effective target heartbeat includes: Feature extraction is performed on the heart sound signals of the effective target heart beats to obtain the heart sound feature data of the effective target heart beats; Feature extraction is performed on the pulse wave signal of the effective target heartbeat to obtain the pulse wave feature data of the effective target heartbeat; The heart sound characteristic data, pulse wave characteristic data, and pulse wave conduction time of the effective target heart beat are input into the pre-trained blood pressure monitoring model, and the blood pressure monitoring results of the wearer are output. The blood pressure monitoring model is trained in advance using training samples with heart sound feature data, pulse wave feature data, and pulse wave conduction time as sample features and blood pressure monitoring results as sample labels.

8. A smart ring, characterized in that, The smart ring includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the blood pressure monitoring method as described in any one of claims 1 to 7.

9. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the blood pressure monitoring method as described in any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the blood pressure monitoring method as described in any one of claims 1 to 7.

Images