Vascular sclerosis evaluation device, system and method based on multi-physiological parameter fusion and dynamic load test

Abstract

The application relates to the technical field of wearable medical devices, and particularly provides a blood vessel hardening evaluation device, system and method based on multi-physiological parameter fusion and dynamic load testing, which comprises a wearing body and a processing system integrated on the wearing body, the processing system comprising: a signal synchronous acquisition module for synchronously acquiring PPG signals, ECG signals and motion measurement signals; a physiological parameter estimation module for calculating first-type physiological parameters based on the PPG signals and the ECG signals; a comprehensive index generation module for generating a blood vessel health comprehensive index of a user according to the first-type physiological parameters and static personal information of the user; a dynamic function evaluation module for generating a dynamic recovery index of the user; and an evaluation data output module for outputting final evaluation data based on the blood vessel health comprehensive index and / or the dynamic recovery index. The device realizes non-invasive, dynamic and high-credibility system evaluation of the blood vessel health condition under the condition of convenient wearing.

Description

Technical Field

[0001] This application relates to the field of wearable medical device technology, specifically to a device, system, and method for assessing vascular sclerosis based on multi-physiological parameter fusion and dynamic load testing. Background Technology

[0002] Arteriosclerosis is a key pathophysiological basis for the development of cardiovascular diseases, and its early assessment is of great significance for the prevention of cardiovascular and cerebrovascular events. An ideal assessment method should be non-invasive, convenient, capable of continuous monitoring, and able to comprehensively reflect the structural and functional status of blood vessels.

[0003] However, existing assessment technologies have significant shortcomings in achieving the above objectives. On the one hand, clinical gold standards, such as carotid-femoral pulse wave velocity and blood flow-mediated vasodilation function testing, usually rely on large-scale specialized equipment, are complex to operate, and can only be performed as single, static measurements within medical institutions, making it impossible to achieve daily continuous monitoring and dynamic functional assessment.

[0004] On the other hand, while current consumer wearable devices that integrate photoplethysmography (PPG) sensors can continuously monitor parameters such as heart rate, their assessment capabilities are limited. These devices typically provide only a single or a few isolated parameter indicators, offering a one-sided assessment and generally lacking the ability to quantitatively assess the dynamic response and recovery capabilities of the vascular system under standardized loads.

[0005] Furthermore, in the field of clinical research, although it has been recognized that integrating multiple physiological parameters can provide a more comprehensive assessment of vascular health, such approaches usually require the combination of multiple large devices, making the systems complex, costly, and data processing cumbersome, and difficult to translate into convenient solutions for personal daily use. Summary of the Invention

[0006] To overcome the shortcomings of the prior art, this application provides a device, system and method for assessing arteriosclerosis based on the fusion of multiple physiological parameters and dynamic load testing, which can achieve a systematic, dynamic and highly reliable comprehensive assessment of the degree of arteriosclerosis and functional elasticity of human blood vessels in a convenient and imperceptible daily wearable device.

[0007] The first aspect of this application provides a vascular sclerosis assessment device based on multi-physiological parameter fusion and dynamic load testing, including a wearable body and a processing system integrated on the wearable body, the processing system comprising:

[0008] The signal synchronization acquisition module is used to simultaneously acquire PPG signals, ECG signals, and motion measurement signals; The physiological parameter estimation module is used to calculate at least one type of physiological parameter for assessing vascular function based on the synchronously acquired PPG signal and ECG signal. The comprehensive index generation module is used to generate a comprehensive vascular health index for the user based on the first type of physiological parameters and the user's static personal information. The dynamic functional assessment module based on motion measurement is used to guide users to perform standardized exercises through a preset interactive method, and after the standardized exercises are completed, to monitor the recovery process of the first type of physiological parameters and / or the second type of physiological parameters, and to monitor the exercise status during the standardized exercises based on the motion measurement signals; and to generate the user's dynamic recovery index based on the recovery process. The assessment data output module is used to output final assessment data reflecting the degree of vascular hardening and functional status based on the comprehensive vascular health index and / or the dynamic recovery index.

[0009] In an optional embodiment, the signal synchronization acquisition module includes: At least one optical sensor is used to acquire PPG signals; At least one pair of bioelectric electrodes are used to acquire ECG signals; At least one motion measurement sensor is used to acquire motion measurement signals; The synchronous acquisition unit is used to control the optical sensor and the bioelectric electrode to synchronously acquire the PPG signal and the ECG signal at a sampling frequency of not less than 125Hz, and to control the sampling frequency of the motion measurement sensor to be not less than 10Hz.

[0010] In an optional embodiment, the optical sensor includes a single-wavelength PPG sensor that emits green light.

[0011] In an optional embodiment, the first type of physiological parameters includes one or more combinations of pulse wave velocity, pulse transit time, blood pressure variability, baroreflex sensitivity, and nocturnal blood pressure pattern; the second type of physiological parameters includes heart rate.

[0012] In an optional embodiment, monitoring the recovery process of the first type of physiological parameter and / or the second type of physiological parameter, and generating a dynamic recovery index for the user based on the recovery process, includes: Monitor the recovery process of pulse wave velocity (a first type of physiological parameter) and / or heart rate (a second type of physiological parameter), and obtain the recovery curves of the pulse wave velocity and / or the heart rate. The dynamic recovery index is generated based on the time required for the recovery curve of the pulse wave conduction velocity and / or the heart rate to return to the baseline level, or based on the slope of the recovery curve of the pulse wave conduction velocity and / or the heart rate.

[0013] In an optional embodiment, the interaction method includes tactile feedback from the wearable body or audiovisual prompts from a remote server connected to the wearable body; the standardized movement includes isometric grip movement or stepping in place movement.

[0014] In an optional embodiment, the comprehensive index generation module has a built-in preset weighted fusion model or a trained machine learning model, including random forest, gradient boosting tree or neural network model.

[0015] In an optional embodiment, the dynamic function evaluation module is further configured to feed back the dynamic recovery index to the comprehensive index generation module; The comprehensive index generation module is also used to optimize and update the comprehensive vascular health index based on the physiological parameters, the static personal information, and the dynamic recovery index.

[0016] In an optional embodiment, the arteriosclerosis assessment device further includes an automatic alarm module disposed on the wearable body; the automatic alarm module is used to generate an alarm signal based on at least one of micro-vibration detection, voice command recognition or music pattern recognition, and / or automatically generate an alarm signal when the first type of physiological parameter, the second type of physiological parameter or the dynamic recovery index exceeds the corresponding preset threshold.

[0017] The second aspect of this application provides a vascular sclerosis assessment system based on multi-physiological parameter fusion and dynamic load testing, including the vascular sclerosis assessment device based on multi-physiological parameter fusion and dynamic load testing as described above, and a remote server connected in communication with the vascular sclerosis assessment device. The remote server is used to receive, store, analyze, and visualize at least one of the final assessment data, the comprehensive vascular health index, and the dynamic recovery index.

[0018] A third aspect of this application provides a method for assessing arteriosclerosis based on multi-physiological parameter fusion and dynamic stress testing, applied to the arteriosclerosis assessment device based on multi-physiological parameter fusion and dynamic stress testing as described above, the method comprising: Simultaneously acquire PPG signals, ECG signals, and motion measurement signals; Based on the synchronously acquired PPG and ECG signals, at least one type I physiological parameter for assessing vascular function is calculated. Based on the first type of physiological parameters and the user's static personal information, a comprehensive vascular health index for the user is generated. The system guides users to perform standardized exercises through a preset interactive method and monitors their exercise status during the standardized exercise process based on the exercise measurement signals; and after the standardized exercise is completed, it monitors the recovery process of the first type of physiological parameters and / or the second type of physiological parameters, and generates the user's dynamic recovery index based on the recovery process. Based on the comprehensive vascular health index and / or the dynamic recovery index, the final assessment data reflecting the degree of vascular hardening and functional status is output.

[0019] This application has at least the following beneficial effects: This application provides a device, system, and method for assessing vascular sclerosis based on multi-physiological parameter fusion and dynamic load testing. The device includes a wearable body and a processing system integrated into the wearable body. The processing system includes: a signal synchronization acquisition module for simultaneously acquiring PPG signals, ECG signals, and motion measurement signals; a physiological parameter estimation module for calculating at least one type I physiological parameter for assessing vascular function based on the synchronously acquired PPG and ECG signals; a comprehensive index generation module for generating a comprehensive vascular health index for the user based on the type I physiological parameters and the user's static personal information; a dynamic function assessment module for guiding the user through standardized exercises via a preset interactive method, monitoring the recovery process of the type I and / or type II physiological parameters after the standardized exercises, and generating a dynamic recovery index for the user based on the recovery process; and an assessment data output module for outputting final assessment data reflecting the degree of vascular sclerosis and functional status based on the comprehensive vascular health index and / or the dynamic recovery index. This device achieves a systematic, dynamic, and highly reliable comprehensive assessment of the degree of vascular sclerosis and functional elasticity in a convenient and imperceptible everyday wearable environment. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A schematic diagram of the overall architecture of the arteriosclerosis assessment device based on multi-physiological parameter fusion and dynamic load testing provided in the embodiments of this application; Figure 2 A schematic diagram of the structure of the arteriosclerosis assessment device based on multi-physiological parameter fusion and dynamic load testing provided in the embodiments of this application; Figure 3 A schematic diagram of the modules of the arteriosclerosis assessment device based on multi-physiological parameter fusion and dynamic load testing provided in the embodiments of this application; Figure 4 The overall flowchart of the arteriosclerosis assessment method based on multi-physiological parameter fusion and dynamic load testing provided in the embodiments of this application is shown below. Figure 5 A schematic diagram of the three-stage dynamic testing of the arteriosclerosis assessment method based on multi-physiological parameter fusion and dynamic load testing provided in the embodiments of this application; Figure 6 A comparison of ideal recovery and aging recovery curves for the arteriosclerosis assessment method based on multi-physiological parameter fusion and dynamic load testing provided in this application embodiment; Figure 7 The flowchart illustrates the dynamic testing control process of the arteriosclerosis assessment method based on multi-physiological parameter fusion and dynamic load testing provided in this application embodiment.

[0022] Figure label: 1-Wearable device; 2-Signal synchronization acquisition module; 3-Physiological parameter estimation module; 4-Comprehensive index generation module; 5-Dynamic function assessment module based on motion measurement; 6-Assessment data output module; 7-Microprocessor; 8-Wireless communication module; 9-Power supply; 21-Optical sensor; 22-Bioelectric electrode; 23-Synchronous acquisition unit. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0024] Example 1 like Figure 1-3 As shown, this application provides a vascular sclerosis assessment device based on multi-physiological parameter fusion and dynamic load testing, including a wearable body 1 and a processing system integrated on the wearable body 1. The processing system includes: Signal synchronous acquisition module 2 is used to synchronously acquire PPG signals, ECG signals and motion measurement signals; Physiological parameter estimation module 3 is used to calculate at least one type I physiological parameter for assessing vascular function based on synchronously acquired PPG and ECG signals. The comprehensive index generation module 4 is used to generate a comprehensive vascular health index for users by integrating the first type of physiological parameters and the user's static personal information. The motion measurement-based dynamic functional assessment module 5 is used to guide users to perform standardized exercises through preset interactive methods, monitor the exercise status according to motion measurement signals to meet preset load conditions, and monitor the recovery process of the first type of physiological parameters and / or the second type of physiological parameters after the standardized exercise is completed, and generate the user's dynamic recovery index based on the recovery process. The assessment data output module 6 is used to output final assessment data reflecting the degree of vascular hardening and functional status based on the comprehensive vascular health index and / or dynamic recovery index.

[0025] This application utilizes a signal synchronous acquisition module 2 to simultaneously acquire PPG and ECG signals, effectively avoiding parameter calculation deviations caused by signal acquisition delays. This ensures more accurate and reliable estimation of first-type physiological parameters such as pulse wave transit time, providing a high-quality data foundation for vascular elasticity assessment. Furthermore, by integrating first-type physiological parameters with the user's static personal information through a comprehensive index generation module 4, an individualized comprehensive evaluation of vascular health is achieved. This overcomes the limitations of traditional single-parameter assessments, which are susceptible to interference from factors such as age and underlying diseases, making the assessment results more targeted and valuable. Moreover, by guiding users through standardized exercises and monitoring the recovery process of their physiological parameters through a dynamic function assessment module 5, the application objectively reflects the regulatory and recovery capabilities of blood vessels under load, exhibiting high sensitivity, especially in detecting early vascular function decline. This compensates for the shortcomings of static measurements in reflecting dynamic vascular function. The final output covers two dimensions: the degree of vascular hardening and functional status. This not only helps users achieve convenient and continuous vascular health monitoring in daily scenarios but also provides medical institutions with more comprehensive and hierarchical auxiliary diagnostic information, demonstrating outstanding practical value and promising prospects in early screening, risk assessment, and health management of cardiovascular diseases.

[0026] Furthermore, the signal synchronization acquisition module 2 includes: At least one optical sensor 21 is used to acquire PPG signals; At least one pair of bioelectric electrodes 22 are used to acquire ECG signals; At least one motion measurement sensor is used to acquire motion measurement signals; The synchronous acquisition unit 23 is used to control the optical sensor 21 and the bioelectric electrode 22 to synchronously acquire PPG signals and ECG signals at a sampling frequency of not less than 125Hz, and to control the sampling frequency of the motion measurement sensor to be not less than 10Hz.

[0027] This application, through the cooperation of optical sensor 21, bioelectric electrode 22, motion measurement sensor, and synchronous acquisition unit 23, can capture the morphological details and feature points of PPG waveforms, ECG waveforms, and motion signals with greater precision. On the one hand, by using a sampling frequency of no less than 125Hz for PPG and ECG signals, the calculation accuracy of time parameters such as pulse wave propagation time is significantly improved, thereby reducing estimation errors caused by insufficient sampling. On the other hand, by using a sampling frequency of no less than 10Hz for the motion measurement sensor, low-frequency to mid-frequency motion state information of the user can be effectively acquired, providing synchronous and reliable reference data for subsequent motion artifact suppression or standardized exercise load monitoring. At the same time, through a strict synchronous acquisition mechanism, the time alignment between PPG signals, ECG signals, and motion measurement signals is ensured at the hardware level, eliminating interference introduced by relative delays between signals, making the estimation of physiological parameters and dynamic function assessment based on multi-signal joint analysis more accurate and reliable.

[0028] Specifically, the optical sensor 21 includes a single-wavelength PPG sensor that emits green light. During the dynamic load test and the subsequent recovery phase, users' limbs often exhibit unavoidable subtle movements. Compared to traditional red or infrared PPG sensors, the green light sensor is less sensitive to these motion artifacts and can maintain a more stable signal baseline under mild activity interference, thus ensuring the continuity and authenticity of the recovery process data upon which the dynamic function assessment module 5 relies. Secondly, the higher signal-to-noise ratio directly improves the accuracy of pulse wave feature point detection (such as waveform peaks, troughs, and inflection points). This provides higher-quality raw data input for the physiological parameter estimation module 3 to calculate accurate pulse wave conduction time, reflected wave enhancement index, and other first-type physiological parameters, indirectly enhancing the reliability of the comprehensive vascular health index.

[0029] In a specific hardware embodiment, the wearable body 1 is a smart ring, which adopts a biocompatible titanium alloy inner ring and a medical-grade polymer outer coating structure. The processing system includes a microprocessor 7, and a physiological parameter estimation module 3, a comprehensive index generation module 4, a dynamic function evaluation module 5, and an evaluation data output module 6 are integrated into the microprocessor 7. The optical sensor 21 is an integrated green LED and photodiode module with a wavelength of 530nm, which serves as a PPG sensor and is placed on the inner wall of the ring near the root of the metacarpophalangeal joint to take advantage of the rich blood supply and relatively small motion artifacts in this location. The at least one pair of bioelectric electrodes 22 are two medical-grade stainless steel electrodes, serving as the working electrode and the reference electrode, respectively; the working electrode is located on the inner wall of the ring in contact with the ventral side of the finger, and the reference electrode is located on the outer wall of the ring. The user needs to briefly touch the reference electrode with another finger (such as the thumb) to form a circuit, thereby constituting a single-lead ECG acquisition. The synchronous acquisition unit 23 uses a dedicated integrated chip, such as the TIAFE4900, which contains two independent analog-to-digital conversion channels and is specifically designed for hardware-level synchronization of PPG and ECG signals to ensure strict alignment of sampling times. In this embodiment, the system is set to a sampling frequency of 250Hz to fully capture the details of the PPG waveform and meet the millisecond-level accuracy requirement for pulse wave initiation point detection based on pulse wave propagation velocity. The motion measurement sensor is a microelectromechanical triaxial accelerometer, fixedly mounted on the circuit board of the wearable device, aligning the sensitive axis of the motion measurement sensor with the direction of the user's limb movement. The sampling frequency can be configured via registers to 10Hz, 25Hz, or 50Hz, and outputs a digital triaxial acceleration value.

[0030] Furthermore, the first category of physiological parameters includes one or more combinations of pulse wave velocity, pulse transit time, blood pressure variability, baroreflex sensitivity, and nocturnal blood pressure patterns; the second category of physiological parameters includes heart rate. This device enables a comprehensive, multi-level quantitative assessment of the user's vascular health status, from "static structure" to "dynamic function," and from "local characteristics" to "systemic regulation." The final assessment data output is not only more clinically convincing but also provides precise and multi-dimensional decision-making support for individualized health interventions and risk management, greatly enhancing the practical depth and application prospects of this wearable device in preventive medicine and health management.

[0031] Specifically, the selected first-class physiological parameters, such as pulse wave velocity, pulse transit time, blood pressure variability, baroreflex sensitivity, and nocturnal blood pressure pattern, comprehensively evaluate the vascular structure and function from different and key physiological dimensions, including vascular elasticity, short-term autonomic nervous system regulation, cardiovascular reflex arc integrity, and circadian rhythm physiological patterns. This multi-indicator combination overcomes the limitations of relying solely on pulse wave velocity, enabling the assessment to reflect both the degree of structural sclerosis and to reveal functional regulatory abnormalities and neuroendocrine imbalances, especially demonstrating higher detection sensitivity for early and subclinical vascular dysfunction.

[0032] The selected Class I physiological parameters, such as heart rate, have important physiological basis and practical significance. Heart rate recovery dynamics is a sensitive indicator reflecting autonomic nervous system balance and the overall stress and recovery capacity of the cardiovascular system. Analyzing it in conjunction with the recovery process of Class I parameters can more comprehensively depict the overall functional recovery status of the cardiovascular system after load, thus allowing the dynamic recovery index to contain richer physiological information.

[0033] Monitoring the recovery process of Type I and / or Type II physiological parameters, and generating a dynamic recovery index for the user based on the recovery process, includes: Monitor the recovery process of pulse wave velocity (a first-order physiological parameter) and / or heart rate (a second-order physiological parameter) to obtain recovery curves of pulse wave velocity and / or heart rate; A dynamic recovery index is generated based on the time required for the recovery curves of pulse wave velocity and / or heart rate to return to baseline levels, or based on the slope of the recovery curves of pulse wave velocity and / or heart rate.

[0034] This application significantly improves the objectivity, accuracy, and clinical applicability of dynamic assessment of vascular function by focusing on the recovery curves of two core physiological parameters, pulse wave velocity and / or heart rate, and quantifying the dynamic recovery index based on the time required to recover to baseline levels or the slope of the recovery curve.

[0035] The interaction methods include tactile feedback from the wearable device 1 or audiovisual prompts from a remote server connected to the wearable device 1; the standardized movements include isometric grip movements or stepping in place movements. By clearly defining the human-computer interaction methods and standardized movement forms, this application effectively solves the key operational challenges of dynamic function assessment in practical applications, and significantly improves the standardization of the testing process, user compliance, and the repeatability and comparability of the assessment results.

[0036] In a specific hardware embodiment, in the above-described smart ring embodiment, the tactile feedback of the wearable body 1 is provided by a vibration motor integrated in the smart ring, which guides the user to start and end the isometric grip movement, thereby achieving standardized dynamic load testing without the assistance of external devices.

[0037] The comprehensive index generation module 4 has a built-in preset weighted fusion model or a trained machine learning model, including random forest, gradient boosting tree or neural network model.

[0038] In one embodiment, the comprehensive index generation module 4 has a built-in preset weighted fusion model to assign reasonable weights to different first-type physiological parameters and static personal information, thereby achieving highly interpretable and logically transparent index calculation and ensuring that the evaluation results have a stable clinical significance basis.

[0039] In another embodiment, the comprehensive index generation module 4 incorporates a trained machine learning model, including random forests, gradient boosting trees, or neural network models. This enables the device to handle complex, nonlinear physiological relationships: such models can learn from large amounts of historical labeled data and uncover potential deep correlation patterns between multiple input features and vascular health status, thereby generating a more accurate and sensitive comprehensive assessment index that surpasses traditional linear formulas. This not only significantly improves the accuracy of identifying early vascular dysfunction or hardening risk but also enhances the index's adaptability to individual-specific physiological patterns, making the assessment results more personalized.

[0040] In other embodiments, a pre-defined weighted fusion model may be used first, followed by a machine learning model, to efficiently generate a comprehensive vascular health index with limited local computing resources.

[0041] The dynamic function assessment module 5 is also used to feed back the dynamic recovery index to the comprehensive index generation module 4; the comprehensive index generation module 4 is also used to optimize and update the comprehensive vascular health index based on physiological parameters, static personal information, and the dynamic recovery index. This application achieves a significant upgrade to the comprehensive vascular health assessment model by establishing a closed-loop data feedback and optimization mechanism between the dynamic function assessment module 5 and the comprehensive index generation module 4, enabling the assessment results to possess advanced characteristics of dynamic evolution, self-correction, and high individualization. In the above system, the smart ring communicates with the user's mobile APP (as a near-end server or gateway) via a wireless communication module 8, such as Bluetooth 5.2, and then uploads the data to a cloud-based remote server for in-depth analysis and visualization.

[0042] Furthermore, the arteriosclerosis assessment device based on multi-physiological parameter fusion and dynamic load testing also includes an automatic alarm module, which is installed on the wearable body. The automatic alarm module generates an alarm signal based on at least one triggering method selected from micro-vibration detection, voice command recognition, or music pattern recognition, and / or automatically generates an alarm signal when the first type of physiological parameter, the second type of physiological parameter, or the dynamic recovery index exceeds the corresponding preset threshold. This application improves the safety of high-risk groups by integrating the arteriosclerosis assessment device with an emergency alarm function; moreover, the multi-modal triggering method takes into account both user-initiated assistance and automatic early warning of physiological abnormalities, avoiding the risk of failure of a single mechanism when the user is unable to operate or the device misjudges.

[0043] Specifically, for triggering methods based on micro-vibration detection, the automatic alarm module may include one or more MEMS accelerometers or dedicated vibration sensors. The sensitivity of the vibration sensor is configured to detect user-preset tapping patterns, such as two consecutive rapid taps or long-press vibrations. When the detected vibration time-domain characteristics (number of pulses, interval time) or frequency-domain characteristics (main frequency range 10Hz~100Hz) match the preset alarm triggering mode, the microcontroller generates an alarm signal.

[0044] For triggering methods based on voice command recognition, the automatic alarm module can include a microphone, a voice activity detector, and locally stored wake words. After detecting preset keywords such as "emergency alarm" or "I need help," an alarm signal can be generated without connecting to the network.

[0045] For music pattern recognition-based triggering, the automatic alarm module includes an audio acquisition unit and an audio fingerprint matching unit. It pre-stores specific music clips, such as the acoustic feature vectors of an emergency contact's phone ringtone or a specific alarm tone. By comparing the acquired ambient audio with the pre-stored music clips in real time, an alarm signal is generated when the matching degree exceeds a threshold. The threshold can be set to 85%.

[0046] In addition, the automatic alarm module continuously monitors the first type of physiological parameters, the second type of physiological parameters output by the physiological parameter estimation module, or the dynamic recovery index output by the dynamic function assessment module, and compares them with their respective preset safety thresholds. An alarm is automatically triggered when any parameter exceeds the threshold and the duration exceeds the set confirmation window. For example, an alarm is automatically triggered if the pulse wave velocity in the first type of physiological parameter exceeds 9 m / s, or if the heart rate in the second type of physiological parameter exceeds 85% of the maximum heart rate.

[0047] The alarm signals can be generated in the following ways: by driving the buzzer on the wearable device to make a sound, by causing the vibration motor to vibrate in a specific mode, by sending a notification message containing current physiological data and a timestamp to the paired mobile phone via Bluetooth or Wi-Fi, or by directly dialing a preset emergency phone number.

[0048] A system for assessing arteriosclerosis based on multi-physiological parameter fusion and dynamic stress testing includes an arteriosclerosis assessment device based on multi-physiological parameter fusion and dynamic stress testing, and a remote server connected to the arteriosclerosis assessment device. The remote server is used to receive, store, analyze, and visualize at least one of the final assessment data, a comprehensive vascular health index, and a dynamic recovery index. This application, by introducing a remote server, greatly expands the depth and intelligence of data processing, supporting not only the mining of complex historical trends and the comparison of population data, but also continuous optimization and personalized training of the built-in machine learning model, enabling the entire system's assessment algorithm to have dynamic evolution capabilities. The remote server includes mobile terminals and cloud-based systems.

[0049] like Figure 4-7 As shown, a method for assessing arteriosclerosis based on multi-physiological parameter fusion and dynamic stress testing is applied to an arteriosclerosis assessment device based on multi-physiological parameter fusion and dynamic stress testing. The method includes: Simultaneously acquire PPG signals, ECG signals, and motion measurement signals; Based on the synchronously acquired PPG and ECG signals, at least one type I physiological parameter for assessing vascular function is calculated. Based on the first type of physiological parameters and the user's static personal information, a comprehensive vascular health index for the user is generated. The system guides users to perform standardized exercises through a preset interactive method and monitors the exercise status during the standardized exercise process based on the exercise measurement signals; and after the standardized exercise is completed, it monitors the recovery process of the first type of physiological parameters and / or the second type of physiological parameters and generates the user's dynamic recovery index based on the recovery process. Based on the comprehensive vascular health index and / or dynamic recovery index, the final assessment data reflecting the degree of vascular hardening and functional status is output.

[0050] This invention generates a comprehensive vascular health index by simultaneously acquiring PPG and ECG signals and integrating various static physiological parameters such as pulse wave velocity with user personal information. Simultaneously, it uses standardized exercise load testing to monitor the dynamic recovery process of physiological parameters, generating a dynamic recovery index. This approach combines static assessment with dynamic functional testing, achieving a comprehensive, multi-level quantitative assessment of the degree of vascular sclerosis and regulatory capacity. Its beneficial effects include: significantly improving the sensitivity and accuracy of early vascular dysfunction detection; making assessment results more personalized and predictive through machine learning models and closed-loop feedback mechanisms; ensuring ease of operation and repeatability of results through standardized interactive design; and ultimately constructing a complete system from data acquisition and intelligent analysis to visualized management, providing an efficient and reliable solution for early screening of cardiovascular disease risks, health trend tracking, and personalized health management.

[0051] Specifically, step 1: synchronously acquiring PPG and ECG signals includes: The raw PPG and ECG signals acquired by the signal synchronization acquisition module 2 are bandpass filtered (PPG: 0.5Hz-15Hz; ECG: 0.67Hz-40Hz) to remove baseline drift and high-frequency noise, and an algorithm based on accelerometer signals is used to further reduce motion artifacts, providing a clean signal for subsequent accurate analysis.

[0052] Step 2: Based on the synchronously acquired PPG and ECG signals, at least one type I physiological parameter for assessing vascular function is calculated, specifically including: Based on synchronously acquired PPG and ECG signals, one or more combinations of pulse wave conduction velocity, blood pressure variability, baroreflex sensitivity, and nocturnal blood pressure patterns are calculated.

[0053] For pulse wave velocity, PWV = distance / PAT = estimated PWV of the arm-finger pathway.

[0054] PAT, or Pulse Arrival Time, is calculated by detecting the peak of the R-wave in the ECG signal and the start of the pulse wave (the inflection point where the waveform begins to rise rapidly) in the PPG signal within a single cardiac cycle. The time difference between the peak of the R-wave in the ECG and the start of the PPG in the corresponding heartbeat is then calculated.

[0055] The distance is preset by the user through the accompanying mobile application (APP). In one example, the arm length (the length from the acromion to the tip of the middle finger) entered by the user is used directly as the estimated distance. If the user does not provide the arm length, the estimated distance is calculated from the height using a statistical relationship model between height and the length of the vascular path from the aorta to the peripheral measurement point (in this case, the finger) (e.g., path length ≈ height × a fixed empirical coefficient).

[0056] Substituting the PAT (time) obtained through signal processing and the "distance" (length) estimated through user information into the formula PWV = distance / PAT, the result is the "PWV estimate of the arm-finger pathway", that is, the pulse wave conduction velocity.

[0057] For blood pressure variability (BPV) and nocturnal blood pressure patterns, the trends of systolic and diastolic blood pressure are continuously estimated from the PPG waveform using a pre-trained neural network model (trained with thousands of PPG and cuff blood pressure data pairs). Based on this, the standard deviation of blood pressure over 24 hours is calculated as an indicator of blood pressure variability (BPV). Nighttime sleep periods are identified (either through user-defined markings or automatic determination based on activity levels), and the percentage decrease in nighttime average blood pressure relative to daytime average blood pressure is calculated as a quantitative indicator of the nocturnal blood pressure pattern.

[0058] For the baroreflex sensitivity (BRS), a sequential method is used. From continuous blood pressure trend signals and cardiac cycle signals derived from the ECG-R interval, sequences in which systolic blood pressure changes in the same or opposite direction as the subsequent RR interval are identified (for example, if blood pressure rises more than 3 times consecutively, the RR interval also prolongs consecutively). The regression slopes of these sequences are calculated, and the average value is taken as the estimate of the baroreflex sensitivity (BRS).

[0059] Step 3: Based on the first type of physiological parameters and the user's static personal information, a comprehensive vascular health index for the user is generated, specifically including: In one example, the user's Vascular Health Index (VHI) is calculated using the following weighted formula after normalization with the user's age based on parameters such as pulse wave velocity (PWV), blood pressure variability (BPV), nocturnal dip, and baroreflex sensitivity (BRS): VHI=0.35×(1-Norm_PWV)+0.25×(1-Norm_BPV)+0.20×Norm_NocturnalDip+0.15×Norm_BRS+0.05×(1-Norm_Age) Here, Norm_× represents the normalized value that maps the corresponding original parameter to the interval [0,1]. The mapping relationship is predefined based on statistical data from a large-scale healthy population. For example, PWV is normalized to 0.1 at 6 m / s and to 0.9 at 10 m / s; the percentage decrease in nighttime blood pressure is normalized to 1.0 at 15% and to 0.1 at 0%. Ultimately, the closer the VHI value is to 1, the healthier and more elastic the blood vessels are.

[0060] In another example, normalized PWV, BPV, percentage decrease in nighttime blood pressure, BRS, age, sex, and resting heart rate are input features into a lightweight embedded neural network model. This model directly outputs a VHI value between 0 and 1. The model requires training on a massive dataset with gold-standard labels (such as neck-femoral PWV) in the cloud, and then deployed to the ring's microprocessor 7 after quantization and compression.

[0061] Step 4: Guide the user to perform standardized exercises through a preset interactive method, and monitor the exercise status during the standardized exercises based on the exercise measurement signals; and after the standardized exercises are completed, monitor the recovery process of the first type of physiological parameters and / or the second type of physiological parameters, and generate the user's dynamic recovery index based on the recovery process. Specifically, this includes: like Figure 5-7 As shown, the corresponding user starts the "Vascular Function Test" mode on the APP, guides the user to sit still, and continuously collects and records parameters such as PWV and heart rate in a resting state for 2-3 minutes as a safety baseline and benchmark for subsequent calculations.

[0062] Vibration prompts guide users to perform a standardized, low- to moderate-intensity physiological stress response. In a preferred embodiment, isometric grip exercise is employed: the ring or app instructs the user to exercise continuously at 30%-40% of their maximum grip strength. During this process, the system monitors the user's heart rate in real time. The core control logic is: when the algorithm detects in real time that the user's heart rate has reached and stabilized within their preset safe target heart rate range (e.g., 60%-70% of the maximum heart rate estimated based on age), it is determined that an optimal and safe physiological stress state has been reached.

[0063] Once the above heart rate conditions are met, the system will immediately notify the user to stop exercising and remain quiet via continuous vibration. The load phase will automatically end, and the system will immediately enter an active recovery monitoring period of up to 5 minutes. During this period, the system will collect signals at a higher frequency (e.g., once per second) and calculate PWV and heart rate, obtaining their recovery curves over time.

[0064] The dynamic recovery index for users is generated by extracting key kinetic parameters from the recovery curve. For example, the PWV recovery half-life (the time required to recover to half of baseline, T1 / 2_PWV) and the initial slope of heart rate recovery (the rate of heart rate decline in the first 60 seconds, HRR_Slope) are extracted. The dynamic recovery index is calculated using the following weighted formula: Dynamic recovery index = 0.6 × (1 - Norm_T1 / 2_PWV) + 0.4 × Norm_HRR_Slope.

[0065] Here, Norm_T1 / 2_PWV and Norm_HRR_Slope represent the values ​​after normalization of the original T1 / 2_PWV and HRR_Slope parameters, respectively. This index directly quantifies the functional elasticity of the vascular system in recovering to homeostasis after physiological stress.

[0066] Alternatively, a standardized exercise can be stepping in place. The ring's built-in accelerometer is used to count and maintain the rhythm (e.g., instructing the user to maintain 120 steps per minute) for 3 minutes, and the recovery monitoring metrics and methods are the same as those described in the grip strength test example above.

[0067] For example, such as Figure 6 As shown, simulated data curves visually compare the core differences between individuals with good vascular function and those with vascular aging / hardening during the dynamic test recovery period: PWV (Pulse Wave Velocity) curve: Rapid recovery (ideal state): The curve shows a steep downward trend. After exercise stops, the PWV value drops rapidly, usually approaching or reaching the resting baseline level within 3-4 minutes. This reflects good vascular elasticity and compliance, enabling the blood vessels to quickly recover their original shape after being subjected to blood flow impact.

[0068] Slow recovery (aging state): The curve declines gradually and with a lag. The PWV value remains at a high level during the recovery period, and is still significantly higher than the baseline after 5 minutes. This directly reflects vascular stiffness, poor buffering capacity, and difficulty in quickly restoring the pressure wave transmission speed to normal after stress.

[0069] Heart rate recovery curve: Rapid recovery: The curve shows a sharp and rapid drop after the movement stops (mainly driven by the rapid recovery of vagal tone in the autonomic nervous system), indicating that the heart and autonomic nervous system are highly efficient and responsive to physiological stress.

[0070] Slow recovery: A slow and insufficient decline in the curve indicates impaired or delayed autonomic nervous system regulation (especially the reactivation of the parasympathetic nervous system), which is a sign of aging cardiovascular function and increased risk.

[0071] This systematically reveals the correlation between arteriosclerosis and the decline of neuroregulatory function. By quantifying the characteristics of these recovery curves (such as half-life, area under the curve, and initial slope), an objective "dynamic recovery index" can be calculated, thus transforming the abstract concept of "vascular elasticity" into a measurable scientific indicator.

[0072] Step 5: Based on the comprehensive vascular health index and / or dynamic recovery index, output the final assessment data reflecting the degree of vascular hardening and functional status, specifically including: The final assessment data = 0.7 × VHI + 0.3 × Dynamic Recovery Index. Where VHI is the comprehensive vascular health index generated in step 3, and the Dynamic Recovery Index is the result generated in step 4. This total score, sub-parameters, and trend charts are transmitted via Bluetooth to a mobile app for visualization and can generate personalized health recommendations.

[0073] Those skilled in the art will understand that the modules or steps described above in this application can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computing device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0074] Note that the above description is merely a preferred embodiment and the technical principles employed in this application. Those skilled in the art will understand that this application is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of this application. Therefore, although this application has been described in detail through the above embodiments, this application is not limited to the above embodiments. Many other equivalent embodiments may be included without departing from the concept of this application, and the scope of this application is determined by the scope of the appended claims.

[0075] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A device for assessing arteriosclerosis based on multi-physiological parameter fusion and dynamic load testing, characterized in that, Includes a wearable body and a processing system integrated on the wearable body, the processing system including: The signal synchronization acquisition module is used to simultaneously acquire PPG signals, ECG signals, and motion measurement signals; The physiological parameter estimation module is used to calculate at least one type of physiological parameter for assessing vascular function based on the synchronously acquired PPG signal and ECG signal. The comprehensive index generation module is used to generate a comprehensive vascular health index for the user based on the first type of physiological parameters and the user's static personal information. The dynamic functional assessment module based on motion measurement is used to guide users to perform standardized exercises through a preset interactive method, and monitor the exercise status during the standardized exercise process according to the motion measurement signal; and after the standardized exercise is completed, monitor the recovery process of the first type of physiological parameters and / or the second type of physiological parameters, and generate the user's dynamic recovery index based on the recovery process. The assessment data output module is used to output final assessment data reflecting the degree of vascular hardening and functional status based on the comprehensive vascular health index and / or the dynamic recovery index.

2. The arteriosclerosis assessment device based on multi-physiological parameter fusion and dynamic load testing according to claim 1, characterized in that, The signal synchronization acquisition module includes: At least one optical sensor is used to acquire PPG signals; At least one pair of bioelectric electrodes are used to acquire ECG signals; At least one motion measurement sensor is used to acquire motion measurement signals; The synchronous acquisition unit is used to control the optical sensor and the bioelectric electrode to synchronously acquire the PPG signal and the ECG signal at a sampling frequency of not less than 125Hz, and to control the sampling frequency of the motion measurement sensor to be not less than 10Hz.

3. The arteriosclerosis assessment device based on multi-physiological parameter fusion and dynamic load testing according to claim 2, characterized in that, The optical sensor includes a single-wavelength PPG sensor that emits green light.

4. The arteriosclerosis assessment device based on multi-physiological parameter fusion and dynamic load testing according to claim 1, characterized in that, The first category of physiological parameters includes one or more combinations of pulse wave velocity, pulse transit time, blood pressure variability, baroreflex sensitivity, and nocturnal blood pressure pattern; the second category of physiological parameters includes heart rate.

5. The arteriosclerosis assessment device based on multi-physiological parameter fusion and dynamic load testing according to claim 1, characterized in that, Monitoring the recovery process of the first type of physiological parameter and / or the second type of physiological parameter, and generating a dynamic recovery index for the user based on the recovery process, includes: Monitor the recovery process of pulse wave velocity (a first type of physiological parameter) and / or heart rate (a second type of physiological parameter), and obtain the recovery curves of the pulse wave velocity and / or the heart rate. The dynamic recovery index is generated based on the time required for the recovery curve of the pulse wave conduction velocity and / or the heart rate to return to the baseline level, or based on the slope of the recovery curve of the pulse wave conduction velocity and / or the heart rate.

6. The arteriosclerosis assessment device based on multi-physiological parameter fusion and dynamic load testing according to claim 1, characterized in that, The interaction methods include tactile feedback from the wearable device or audiovisual prompts from a remote server connected to the wearable device; the standardized movements include isometric grip movements or stationary stepping movements.

7. The arteriosclerosis assessment device based on multi-physiological parameter fusion and dynamic load testing according to claim 1, characterized in that, The comprehensive index generation module has a built-in preset weighted fusion model or a trained machine learning model, including random forest, gradient boosting tree or neural network model.

8. The arteriosclerosis assessment device based on multi-physiological parameter fusion and dynamic load testing according to claim 1, characterized in that, The dynamic function evaluation module is also used to feed the dynamic recovery index back to the comprehensive index generation module; The comprehensive index generation module is also used to optimize and update the comprehensive vascular health index based on the physiological parameters, the static personal information, and the dynamic recovery index.

9. The arteriosclerosis assessment device based on multi-physiological parameter fusion and dynamic load testing according to claim 1, characterized in that, The arteriosclerosis assessment device also includes an automatic alarm module, which is disposed on the wearable body. The automatic alarm module is used to generate an alarm signal based on at least one of the triggering methods of micro-vibration detection, voice command recognition, or music pattern recognition, and / or automatically generate an alarm signal when the first type of physiological parameter, the second type of physiological parameter, or the dynamic recovery index exceeds the corresponding preset threshold.

10. A system for assessing arteriosclerosis based on multi-physiological parameter fusion and dynamic load testing, characterized in that, The device includes a vascular sclerosis assessment device based on multi-physiological parameter fusion and dynamic load testing as described in any one of claims 1-9, and a remote server connected to the vascular sclerosis assessment device. The remote server is used to receive, store, analyze, and visualize at least one of the final assessment data, the comprehensive vascular health index, and the dynamic recovery index.

11. A method for assessing arteriosclerosis based on multi-physiological parameter fusion and dynamic load testing, characterized in that, The method, applied to the arteriosclerosis assessment device based on multi-physiological parameter fusion and dynamic load testing as described in any one of claims 1-9, comprises: Simultaneously acquire PPG signals, ECG signals, and motion measurement signals; Based on the synchronously acquired PPG and ECG signals, at least one type I physiological parameter for assessing vascular function is calculated. Based on the first type of physiological parameters and the user's static personal information, a comprehensive vascular health index for the user is generated. The system guides users to perform standardized exercises through a preset interactive method and monitors the exercise status during the standardized exercise process based on the exercise measurement signals; and after the standardized exercise is completed, it monitors the recovery process of the first type of physiological parameters and / or the second type of physiological parameters and generates the user's dynamic recovery index based on the recovery process. Based on the comprehensive vascular health index and / or the dynamic recovery index, the final assessment data reflecting the degree of vascular hardening and functional status is output.

Images