Multi-parameter coordinated cardiovascular health monitoring system and method

Abstract

The application provides a multi-parameter cooperative cardiovascular health monitoring system and method, which comprises: a millimeter wave radar acquisition module for acquiring multi-dimensional physiological signals of a target user; a three-level signal processing module for carrying out noise reduction processing on the multi-dimensional physiological signals and carrying out hierarchical feature extraction on the noise-reduced multi-dimensional physiological signals to obtain multi-dimensional physiological feature parameters; a four-order population adaptation AI evaluation module for carrying out population subdivision on the target user to obtain a subdivided population, dynamically adjusting the weight proportion of each feature parameter in the multi-dimensional physiological feature parameters according to the subdivided population to obtain dynamically adjusted feature parameters, and carrying out blood pressure calculation, blood vessel hardening calculation and blood vessel risk prediction according to the dynamically adjusted feature parameters to obtain a blood vessel health monitoring result. The application embodiment further improves the cardiovascular health monitoring accuracy and the application range through software optimization and measurement technology expansion.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of biomedical engineering, and particularly relates to a multi-parameter cooperative cardiovascular health monitoring system and method. BACKGROUND

[0002] As one of the primary causes of death and disability worldwide, cardiovascular disease poses a serious threat to human health due to its high incidence, concealment and suddenness. With the aggravation of population aging and changes in lifestyle, the incidence of the disease is gradually extending to young people. Early screening, real-time monitoring and risk warning are key means to reduce the mortality and morbidity of cardiovascular disease.

[0003] In the existing cardiovascular health monitoring process, different populations are monitored by using a unified threshold, which leads to low accuracy of cardiovascular health monitoring. SUMMARY

[0004] The purpose of the embodiment of the present application is to provide a multi-parameter cooperative cardiovascular health monitoring system and method to solve the problem of low accuracy of existing cardiovascular health monitoring.

[0005] The embodiment of the present application is implemented as follows: a multi-parameter cooperative cardiovascular health monitoring system comprises:

[0006] A millimeter wave radar acquisition module is configured to acquire multi-dimensional physiological signals of a target user.

[0007] A three-level signal processing module is configured to perform noise reduction processing on the multi-dimensional physiological signals and perform hierarchical feature extraction on the multi-dimensional physiological signals after noise reduction to obtain multi-dimensional physiological feature parameters.

[0008] A four-order population adaptation AI evaluation module is configured to perform population segmentation on the target user to obtain a segmented population, dynamically adjust the weight proportion of each feature parameter in the multi-dimensional physiological feature parameters according to the segmented population to obtain dynamically adjusted feature parameters, and perform blood pressure calculation, blood vessel hardening calculation and blood vessel risk prediction according to the dynamically adjusted feature parameters to obtain a blood vessel health monitoring result.

[0009] The embodiment of the present application can effectively collect the multi-dimensional physiological signals of the target user through the millimeter wave radar acquisition module, improve the quality of the multi-dimensional physiological signals through noise reduction processing of the multi-dimensional physiological signals, effectively extract the multi-dimensional physiological feature parameters through hierarchical feature extraction of the noise-reduced multi-dimensional physiological signals, improve the diversity of the physiological feature parameters, prevent the phenomenon of low cardiovascular health monitoring accuracy caused by single collection parameter, effectively classify the target user through population segmentation of the target user, dynamically adjust the weight proportion of each feature parameter in the multi-dimensional physiological feature parameters based on the segmented population, effectively improve the accuracy of the multi-dimensional physiological feature parameters for different populations, and further improve the accuracy of the cardiovascular health monitoring result and the cardiovascular health monitoring accuracy.

[0010] Preferably, the millimeter wave radar acquisition module comprises:

[0011] The radar array unit adopts a millimeter wave radar with a multi-transmit multi-receive antenna layout, is used for synchronously collecting the pulse waves, heart rates and respiration signals of the carotid artery of the neck, the aorta of the chest, the radial artery of the wrist and the posterior tibial artery of the ankle of the target user, and the body surface temperature field distribution data and hemodynamic correlation signals of the neck, the chest, the wrist and the ankle, and obtains the multi-dimensional physiological signals.

[0012] The adaptive signal adjustment unit is used for automatically adjusting the radar transmission frequency, signal gain and sampling period according to the blood vessel depth, skin thickness and fat coverage rate of the monitored part, so as to adapt to the target user with different body types.

[0013] The multi-scene calibration unit is used for obtaining the monitoring posture of the target user, determining preset scene calibration parameters according to the monitoring posture, and eliminating the interference of the scene posture on signal acquisition according to the preset scene calibration parameters.

[0014] Preferably, the three-level signal processing module comprises:

[0015] The multi-dimensional noise reduction unit adopts a combination algorithm of wavelet transform, Kalman filtering and independent component analysis, performs noise reduction processing on the multi-dimensional physiological signals, and eliminates the influence of body motion interference, environmental noise and temperature drift on the multi-dimensional physiological signals.

[0016] The hierarchical feature extraction unit comprises a basic physiological layer, a pulse wave feature layer and a derived parameter layer.

[0017] The basic physiological layer is used for extracting the heart rate, respiration rate and body surface temperature field temperature difference in the noise-reduced multi-dimensional physiological signals.

[0018] a pulse wave feature layer, configured to extract a rising edge slope, a falling edge slope, a rising time from a valley to a first peak, a falling time from the first peak to the valley, a second peak amplitude, a second peak delay time, and a pulse wave conduction time difference in the multi-dimensional physiological signal after noise reduction;

[0019] a derived parameter layer, configured to calculate a heart rate variability, a pulse pressure, and a pulse pressure index;

[0020] The multi-dimensional physiological feature parameters include a heart rate, a respiratory rate, a body surface temperature field temperature difference, a rising edge slope, a falling edge slope, a rising time, a falling time, a second peak amplitude, a second peak delay time, a pulse wave conduction time difference, a heart rate variability, a pulse pressure, and a pulse pressure index.

[0021] Preferably, the fourth-order population adaptation AI evaluation module comprises:

[0022] a population segmentation unit, configured to obtain population attribute information of the target user, and perform population segmentation on the target user according to the population attribute information to obtain the segmented population, wherein the population attribute information includes age, gender, body shape, underlying diseases, and medication history;

[0023] a dynamic weighting unit, configured to dynamically adjust a weight proportion of each feature parameter in the multi-dimensional physiological feature parameters according to the segmented population to obtain the dynamically adjusted feature parameters;

[0024] a multi-model fusion calculation unit, configured to calculate a systolic pressure and a diastolic pressure by a gradient boosting tree algorithm according to a pulse wave feature, a heart rate variability, and the population attribute information in the dynamically adjusted feature parameters to obtain a target blood pressure, wherein the pulse wave feature includes a rising edge slope and a second peak amplitude;

[0025] configured to output a vascular hardening index by a random forest algorithm according to a pulse wave conduction time difference, a second peak delay time, a pulse pressure index, and a body surface temperature field temperature difference in the dynamically adjusted feature parameters;

[0026] configured to obtain blood pressure and vascular hardening data of the target user within a preset time range, and perform vascular risk prediction according to the blood pressure and vascular hardening data to obtain a vascular risk prediction trend;

[0027] a cross-validation and optimization unit, configured to perform consistency verification and population matching verification on the target blood pressure and the vascular hardening index to obtain a double verification result;

[0028] The vascular health monitoring result includes the target blood pressure, the vascular hardening index, the vascular risk prediction trend, and the double verification result.

[0029] Preferably, the multi-parameter collaborative cardiovascular health monitoring system further includes a full-cycle management module, which includes:

[0030] An encrypted data storage unit, supporting local storage and cloud backup, is used to store the vascular health monitoring results;

[0031] The trend analysis and reporting unit is used to generate personalized health reports for different population segments.

[0032] A tiered early warning unit is used to determine the risk level based on preset differentiated early warning thresholds for the segmented population, and to push intervention suggestions based on the risk level:

[0033] The doctor-patient collaborative interaction unit is used for data sharing, remote parameter adjustment, and online consultation between the target user and the doctor.

[0034] The lifestyle intervention support unit is used to record the target user's diet, monitor exercise, and remind them of medication.

[0035] Preferably, the radar array unit adjusts the antenna spacing through detachable antenna components to adapt to different monitoring needs.

[0036] Preferably, the full-cycle management module also includes a family sharing function unit, which allows family members to view the elderly person's monitoring data and early warning information, enabling remote care.

[0037] Another objective of this invention is to provide a multi-parameter collaborative cardiovascular health monitoring method, comprising:

[0038] Physiological signals are simultaneously collected from different parts of the target user to obtain multi-dimensional physiological signals, and noise reduction processing is performed on the multi-dimensional physiological signals.

[0039] Hierarchical feature extraction is performed on the denoised multidimensional physiological signal to obtain multidimensional physiological feature parameters, and the target user is further segmented to obtain segmented user groups;

[0040] Based on the segmented population, the weight ratio of each feature parameter in the multi-dimensional physiological feature parameters is dynamically adjusted to obtain the dynamically adjusted feature parameters;

[0041] Based on the dynamically adjusted characteristic parameters, blood pressure calculation, arteriosclerosis calculation, and vascular risk prediction are performed to obtain vascular health monitoring results.

[0042] Preferably, physiological signals are simultaneously collected from different parts of the target user to obtain multi-dimensional physiological signals, including:

[0043] acquire the blood vessel depth, skin thickness, and fat coverage of the monitored part of the target user, and determine an adjustment parameter according to the blood vessel depth, skin thickness, and fat coverage;

[0044] adjust the radar transmission frequency, signal gain, and sampling period of the millimeter wave radar according to the adjustment parameter, and synchronously collect the pulse waves, heart rate, and respiration signals of the carotid artery of the neck, the aorta of the chest, the radial artery of the wrist, and the posterior tibial artery of the ankle of the target user, and the body surface temperature field distribution data and hemodynamic correlation signals of the neck, chest, wrist, and ankle according to the adjusted millimeter wave radar, to obtain the multi-dimensional physiological signals;

[0045] acquire the monitoring posture of the target user, determine a preset scene calibration parameter according to the monitoring posture, and perform parameter calibration on the multi-dimensional physiological signals according to the preset scene calibration parameter.

[0046] Preferably, hierarchical feature extraction is performed on the multi-dimensional physiological signals after noise reduction to obtain multi-dimensional physiological feature parameters, including:

[0047] extract the heart rate, respiration rate, and body surface temperature field temperature difference in the multi-dimensional physiological signals after noise reduction, and extract the rising edge slope, falling edge slope, rising time from the valley to the first peak, falling time from the first peak to the valley, second peak amplitude, second peak appearance delay time, and pulse wave conduction time difference in the multi-dimensional physiological signals after noise reduction;

[0048] calculate the heart rate variability, pulse pressure, and pulse pressure index according to the heart rate, respiration rate, body surface temperature field temperature difference, rising edge slope, falling edge slope, rising time from the valley to the first peak, falling time from the first peak to the valley, second peak amplitude, second peak appearance delay time, and pulse wave conduction time difference, to obtain the multi-dimensional physiological feature parameters. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1 is a structural schematic diagram of a multi-parameter cooperative cardiovascular health monitoring system provided by the first embodiment of the present application;

[0050] Figure 2 is a structural schematic diagram of a millimeter wave radar acquisition module provided by the first embodiment of the present application;

[0051] Figure 3 is a structural schematic diagram of a three-level signal processing module provided by the first embodiment of the present application;

[0052] Figure 4 is a structural schematic diagram of a four-order population adaptation AI evaluation module provided by the first embodiment of the present application;

[0053] Figure 5is a structural schematic diagram of a full-cycle management module provided by the first embodiment of the present application.

[0054] Figure 6 is a flow chart of a multi-parameter collaborative cardiovascular health monitoring method provided by the second embodiment of the present application.

[0055] Figure 7 is a structural schematic diagram of a terminal device provided by the third embodiment of the present application. DETAILED DESCRIPTION

[0056] In order to make the objectives, technical solutions, and advantages of the present application clearer, the present application is further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0057] In order to illustrate the technical solutions of the present application, the following specific embodiments are used for illustration.

[0058] Embodiment One

[0059] Please refer to Figure 1 is a schematic diagram of a multi-parameter collaborative cardiovascular health monitoring system 100 provided by the first embodiment of the present application, comprising a millimeter wave radar acquisition module 10, a three-stage signal processing module 11, a four-order crowd adaptation AI evaluation module 12, and a full-cycle management module 13.

[0060] The millimeter wave radar acquisition module 10 is used to acquire multi-dimensional physiological signals of a target user.

[0061] Optionally, please refer to Figure 2 The millimeter wave radar acquisition module 10 comprises:

[0062] The radar array unit 101 adopts a millimeter wave radar with a multi-transmit multi-receive antenna layout, is used to synchronously acquire pulse waves, heart rates, and breathing signals of at least three parts of the carotid artery of the neck, the aorta of the chest, the radial artery of the wrist, and the posterior tibial artery of the ankle of the target user, and surface temperature field distribution data and hemodynamic correlation signals (vessel pulsation amplitude, pulsation cycle variability) of the neck, chest, wrist, and ankle, to obtain the multi-dimensional physiological signals; preferably, the radar array unit 101 adjusts the antenna spacing through a detachable antenna assembly to adapt to different monitoring requirements, and the radar array unit 101 can also measure other physiological signals through microwaves.

[0063] The adaptive signal adjustment unit 102 is used for automatically adjusting radar transmission frequency, signal gain and sampling period according to the blood vessel depth (such as carotid artery deep, radial artery shallow) of the monitored part, the skin thickness (such as thin skin of children, thick skin of the elderly), and the fat coverage (such as thick fat of obese people), so as to adapt to different target users of different body types, support different users of different body types in a conventional use distance range, and penetrate through conventional thickness clothes, medical dressings or hair;

[0064] The multi-scene calibration unit 103 is used for acquiring the monitoring posture of the target user, determining preset scene calibration parameters according to the monitoring posture, and eliminating the interference of the scene posture on signal collection according to the preset scene calibration parameters. The multi-scene calibration unit 103 can use the signal reflection intensity data of the existing radar to monitor the fitting angle of the sensor and the skin in real time (the fitting angle is deduced through the change of reflectivity, and the precision is ±1°); when the angle deviation exceeds ±15°, the signal gain (adjustment range ±20%), the sampling frequency (500Hz / 1000Hz switching) and the feature extraction weight are automatically adjusted to compensate for the signal attenuation caused by the angle deviation, and to ensure that the monitoring accuracy variation coefficient under different wearing postures is less than or equal to 5%.

[0065] In the embodiment, the radar array unit 101 adopts a multi-transmission and multi-reception antenna layout, synchronously collects signals of three or more parts such as the neck (carotid artery, reflecting central blood vessel state), chest (aorta, reflecting large blood vessel elasticity), wrist (radial artery) and ankle (posterior tibial artery, reflecting peripheral blood vessel state), breaks through the limitation of traditional “single part collection”, and can obtain more comprehensive blood vessel physiological characteristics; the collection parameters include not only pulse wave, heart rate and respiration, but also newly added body surface temperature field distribution (reflected by radar signal phase change inversion, reflecting local blood flow) and hemodynamic correlation signal (blood vessel pulsation amplitude and pulsation cycle variability, assisting in judging blood vessel wall elasticity), and the parameter dimension is expanded from the traditional three to six;

[0066] The adaptive signal adjustment unit 102 dynamically adjusts the radar parameters according to the physiological differences of the monitoring parts, such as reducing the transmission power for the shallow neck blood vessels of children (<2mm), and increasing the signal gain for the thick fat of obese people on the chest (>3cm), so as to ensure the signal quality of different groups of people; the multi-scene calibration unit 103 eliminates the posture interference of sitting, lying and standing postures, such as adjusting the radar angle to match the blood vessel direction when standing, and reducing the sampling period to reduce the breathing interference when lying.

[0067] The three-level signal processing module 11 is used for performing noise reduction processing on the multi-dimensional physiological signals, and performing hierarchical feature extraction on the noise-reduced multi-dimensional physiological signals to obtain multi-dimensional physiological feature parameters.

[0068] Optionally, referring to Figure 3 , the three-level signal processing module 11 comprises:

[0069] The multi-dimensional denoising unit 111 adopts a wavelet transform, Kalman filtering and independent component analysis combined algorithm to perform denoising processing on the multi-dimensional physiological signal, eliminate the influence of body motion interference (such as hand shaking, body displacement), environmental noise and temperature drift on the multi-dimensional physiological signal, and output a pure physiological signal.

[0070] The hierarchical feature extraction unit 112 includes a basic physiological layer 113, a pulse wave feature layer 114 and a derived parameter layer 115.

[0071] The basic physiological layer 113 is configured to extract a heart rate, a respiratory rate, and a body surface temperature field temperature difference (such as a neck and wrist temperature difference, a chest and ankle temperature difference) in the multi-dimensional physiological signal after denoising.

[0072] The pulse wave feature layer 114 is configured to extract an uplink slope, a downlink slope, an uplink time from a valley to a first peak, a downlink time from the first peak to the valley, a second peak amplitude, a second peak delay time, and a pulse wave conduction time difference (a difference in pulse wave arrival time between different parts) in the multi-dimensional physiological signal after denoising.

[0073] The derived parameter layer 115 is configured to calculate a heart rate variability (HRV, including SDNN, SDANN and RMSSD indexes), a pulse pressure (systolic pressure-diastolic pressure) and a pulse pressure index (PPI=pulse pressure / systolic pressure).

[0074] The multi-dimensional physiological feature parameters include a heart rate, a respiratory rate, a body surface temperature field temperature difference, an uplink slope, a downlink slope, an uplink time, a downlink time, a second peak amplitude, a second peak delay time, a pulse wave conduction time difference, a heart rate variability, a pulse pressure and a pulse pressure index.

[0075] Preferably, the hierarchical feature extraction unit 112 further includes a subtype feature extraction sublayer, which extracts two types of subtype specific features based on existing pulse wave, temperature field and hemodynamic data: ① atherosclerotic feature (pulse wave uplink slope variation coefficient, blood flow velocity fluctuation amplitude, pulse pressure index stability); and ② calcified hardening feature (pulse wave second peak amplitude proportion, temperature field temperature difference variation coefficient, PWV fluctuation coefficient), each of which has 3 items, and a total of 6 subtype features, which are included in the multi-dimensional physiological feature parameter set.

[0076] Further, the embodiment is provided with a parameter library dedicated to elderly users: ① noise reduction parameters (wavelet noise reduction threshold for elderly users is reduced by 20% compared to young users to avoid loss of weak signals); ② signal amplification parameters (signal preamplification factor for elderly users is increased by 30% to match the signal attenuation characteristics of elderly users); and ③ feature compensation coefficients (a compensation model is constructed based on age, and the pulse amplitude, blood flow velocity and other features of users over 60 years old are multiplied by an age-related coefficient of 1.1-1.3).

[0077] In the embodiment, the multi-dimensional noise reduction unit 111 adopts a combination algorithm of "wavelet transform (to remove high-frequency noise) + Kalman filter (to smooth signal fluctuations) + independent component analysis (to separate body motion and physiological signals)", which solves the problem of poor effect of traditional single filtering, and the body motion interference removal rate is increased to 98% and the signal-to-noise ratio is increased by 40%.

[0078] The hierarchical feature extraction unit 112 adopts a "basic-feature-derivative" three-level extraction logic: the basic physiological layer 113 ensures the integrity of the core vital signs, the pulse wave feature layer 114 excavates the key morphological information of vascular hardening (such as the shorter the second peak appearance delay time, the more severe the vascular hardening), and the derivative parameter layer 115 calculates the HRV, pulse pressure and other clinical evaluation indicators, forming a complete transformation from "raw signal to clinical parameters" and providing rich feature support for subsequent evaluation.

[0079] Optionally, a "third peak feature" extraction is added (existing in some population, commonly seen in elderly or users with underlying diseases), which is used to optimize the evaluation accuracy of special populations, such as the greater the third peak amplitude of the elderly population, the more severe the vascular hardening.

[0080] The four-order population adaptation AI evaluation module 12 is used for population segmentation of the target user to obtain a segmented population, dynamically adjusts the weight proportion of each feature parameter in the multi-dimensional physiological feature parameters according to the segmented population to obtain a dynamically adjusted feature parameter, and performs blood pressure calculation, vascular hardening calculation and vascular risk prediction according to the dynamically adjusted feature parameter to obtain a vascular health monitoring result.

[0081] Optionally, please refer to Figure 4 , the four-order population adaptation AI evaluation module 12 includes:

[0082] The population subunit 121 is configured to obtain population attribute information of the target user, and perform population subunit on the target user according to the population attribute information to obtain the subunit population. The population attribute information includes age, gender, body shape, underlying disease, and medication history. Specifically, the population attribute information includes age (children <14 years old, youth 14-44 years old, middle-aged 45-59 years old, and the elderly ≥60 years old), gender, body shape (BMI stratification: lean <18.5, normal 18.5-23.9, overweight 24-27.9, and obese ≥28), underlying disease (hypertension, diabetes, coronary heart disease, etc.), and medication history (such as whether to take antihypertensive drugs or lipid-lowering drugs). The user is divided into 24 types of subunit population (4 ages × 2 genders × 3 body shapes).

[0083] The dynamic weighting unit 122 is configured to dynamically adjust the weight proportion of each feature parameter in the multi-dimensional physiological feature parameter according to the subunit population to obtain the dynamically adjusted feature parameter.

[0084] The child population: HRV weight 35%, body surface temperature field temperature difference weight 25%, and pulse wave rising edge slope weight 20%;

[0085] The elderly population: pulse wave transit time difference weight 40%, second peak amplitude weight 25%, and pulse pressure index weight 20%;

[0086] The obese population: body surface temperature field temperature difference weight 30%, pulse wave falling edge slope weight 25%, and pulse pressure weight 20%;

[0087] The underlying disease population: according to the disease type adjustment, such as increasing the pulse pressure weight to 30% for the hypertension patients and increasing the HRV weight to 30% for the diabetes patients.

[0088] The multi-model fusion calculation unit 123 includes a blood pressure calculation submodel, a blood vessel hardening calculation submodel, and a risk prediction submodel.

[0089] The blood pressure calculation submodel is configured to calculate systolic pressure and diastolic pressure by a gradient boosting tree algorithm according to the pulse wave features (rising edge slope and second peak amplitude), heart rate variability, and the population attribute information in the dynamically adjusted feature parameter to obtain the target blood pressure. The pulse wave features include the rising edge slope and the second peak amplitude.

[0090] The blood vessel hardening calculation submodel is configured to output a blood vessel hardening index (including three core indexes of β index, PWVβ, and AI) by a random forest algorithm according to the pulse wave transit time difference (calculating the PWV correlation value), second peak appearance delay time, pulse pressure index, and body surface temperature field temperature difference in the dynamically adjusted feature parameter.

[0091] The risk prediction sub-model is configured to obtain blood pressure and blood vessel hardening data of the target user within a preset time range, and perform blood vessel risk prediction based on the blood pressure and blood vessel hardening data to obtain a blood vessel risk prediction trend.

[0092] The cross-validation and optimization unit is configured to perform consistency verification (for example, a Pearson correlation coefficient of a systolic pressure and a PWV correlation value is greater than or equal to 0.85) and population matching verification (for example, a blood vessel hardening index of an old population should be higher than that of a young population) on the target blood pressure and the blood vessel hardening index, to obtain a double verification result, and if the double verification result does not meet the requirement, the dynamic weight is adjusted until the result converges.

[0093] The blood vessel health monitoring result includes the target blood pressure, the blood vessel hardening index, the blood vessel risk prediction trend, and the double verification result.

[0094] Preferably, the multi-model fusion calculation unit 123 further includes a progression rate prediction sub-model, which is configured to predict a blood vessel hardening progression rate in the future 6-12 months (for example, "the blood vessel hardening index is expected to increase by 5.2% in the next 6 months") based on historical monitoring data (blood vessel hardening index, blood pressure, PWV, and pulse pressure index) of the user in the past 3-6 months, using a light-weight LSTM model (with a parameter scale less than or equal to 500,000), and output a progression risk level (low risk: <5% / 6 months; medium risk: 5%-10% / 6 months; high risk: >10% / 6 months) and a targeted intervention suggestion (for example, a high-risk suggestion "visit a doctor and adjust a treatment plan within 1 month").

[0095] The multi-model fusion calculation unit 123 can also access historical clinical data (for example, carotid IMT values and blood lipid reports) of the user to calibrate the AI model periodically and improve long-term evaluation accuracy.

[0096] In this embodiment, the population subdivision unit 121 divides the user into 24 types of subdivided populations (4 ages x 2 genders x 3 body types), and further refines them in combination with basic diseases and medication history to solve the poor adaptability problem caused by the existing "coarse classification", for example, "old people" are subdivided into "old hypertensive obese men", "old normal body type women", and the like, and each type of population corresponds to exclusive model parameters.

[0097] The dynamic weighting unit 122 assigns different weights to different populations based on the weight model trained based on 100,000+ clinical samples: the child population focuses on HRV (reflecting cardiovascular development) and body surface temperature field (reflecting peripheral blood flow), the old population focuses on pulse wave conduction (reflecting large blood vessel hardening) and second peak characteristics (reflecting reflected wave intensity), and the basic disease population focuses on parameters related to diseases (for example, high blood pressure patients focus on pulse pressure), thereby avoiding "one-size-fits-all" weight distribution and improving evaluation accuracy by 35% compared with a unified model.

[0098] The multi-model fusion calculation unit 123 adopts a "blood pressure-vascular hardening-risk prediction" three-model linkage: the blood pressure calculation sub-model combines pulse wave characteristics and population attributes to solve the error problem of traditional pulse wave-based blood pressure estimation (error reduced from ±5 mmHg to ±2.5 mmHg); the vascular hardening calculation sub-model fuses four types of core features (conduction time difference, second peak, pulse pressure index, temperature field), outputs beta index, PWV beta, and AI three indexes, and comprehensively evaluates the degree of vascular hardening; the risk prediction sub-model predicts trends based on historical data and warns of the risk of aggravation of vascular hardening 15 days in advance;

[0099] The cross-validation and optimization unit ensures the reliability of the results through double verification: "blood pressure-vascular hardening consistency verification" avoids single index error (such as high systolic pressure but normal PWV, which may indicate temporary blood pressure elevation rather than vascular hardening), and "population characteristics-computation result matching verification" avoids model abnormalities (such as children's vascular hardening index should not be higher than that of young people), and automatically adjusts the weight when verification fails, with a result convergence rate of 99%.

[0100] Please refer to Figure 5 , the full-cycle management module 13 comprises:

[0101] The encrypted data storage unit 131 supports local storage and cloud backup (in line with medical data security standards) and is used to store the vascular health monitoring results. It can store more than 1 year of monitoring data and generate data reports according to "day / week / month / season"; preferably, the encrypted data storage unit 131 includes a clinical standard export sub-module that supports exporting user monitoring data (vascular hardening index, blood pressure, PWV, monitoring time, subtype identification results, progression risk level) into a hospital commonly used format (Excel / PDF). The exported data is arranged according to the clinical report specification, including index name, value, normal reference range, abnormal prompt, monitoring period, etc., and is adapted to the hospital HIS system data import format.

[0102] The trend analysis and reporting unit 132 is used to generate individualized health reports for different subpopulations:

[0103] Children's report: focuses on HRV changes and body surface temperature field distribution to evaluate cardiovascular development status;

[0104] Young report: focuses on blood pressure fluctuation rules and vascular hardening index base value to indicate the impact of bad living habits;

[0105] Middle-aged report: analyzes the increase of vascular hardening index and blood pressure morning peak phenomenon to warn cardiovascular risk;

[0106] Old report: tracks the trend of vascular hardening index change and blood pressure circadian rhythm, and correlates with the risk of cardiovascular and cerebrovascular events;

[0107] Preferably, the trend analysis and report unit 132 includes a personalized information collection sub-module that collects information such as combined diseases (e.g., diabetes, coronary heart disease), exercise frequency, dietary preferences, etc. of the user; and dynamically generates a targeted intervention plan in combination with existing monitoring data and population attributes: ① Combined diabetes users: focus on recommending "low-sugar diet (daily carbohydrates ≤200g) + moderate-intensity exercise (fast walking / tai chi) 1 hour after meals"; ② Lack of exercise users: recommend "fast walking 3 times a week, 30 minutes each time, with heart rate controlled at (170-age) times / min"; ③ Hypertension combined with vascular sclerosis users: suggest "blood pressure control target < 130 / 80 mmHg, prefer long-acting antihypertensive drugs.

[0108] The hierarchical early warning unit 133 is used to determine the risk level according to the preset differentiated early warning threshold of the subpopulation (e.g., 140 mmHg for the systolic pressure early warning threshold of elderly hypertensive patients, 130 mmHg for young normal population; 14 for the beta index early warning threshold of elderly vascular sclerosis, and 12 for young people), trigger the sound and light early warning (ringtone + pop-up window) on the mobile phone when the threshold is exceeded, and push intervention suggestions according to the risk level: The collaborative early warning rules are set in the hierarchical early warning unit 133, all existing monitoring indicators (vascular sclerosis index, blood pressure, PWV, pulse pressure index, temperature field temperature difference) are integrated, and three types of early warning trigger conditions are set: ① At least 2 core indicators (vascular sclerosis index, PWV, blood pressure) are simultaneously out of range; ② A single core indicator shows a rising trend for 3 consecutive times (each time the rising amplitude is ≥3%); ③ The core indicator change rate exceeds the threshold (e.g., the vascular sclerosis index increases by ≥10% in 1 month); any one of the conditions triggers the sound and light early warning, and the false alarm rate is ≤8%.

[0109] Low risk: lifestyle adjustment suggestions (e.g., low-salt diet, aerobic exercise);

[0110] Medium risk: prompt medical examination (e.g., carotid ultrasound, blood lipid detection);

[0111] High risk: emergency contact with family members or doctors, and push the navigation of nearby medical institutions;

[0112] The doctor-patient collaborative interaction unit 134 is used for data sharing, remote parameter adjustment, and online consultation between the target user and the doctor;

[0113] Data sharing: the user can authorize the doctor to view the historical monitoring data and AI reports, support PDF export or direct synchronization to the hospital HIS system;

[0114] Remote parameter adjustment: the doctor can remotely adjust the weight distribution (e.g., increase the pulse wave transmission time difference weight) or early warning threshold of the AI model according to the patient's condition (e.g., new coronary heart disease diagnosis);

[0115] Online consultation: integrated doctor online consultation portal, supporting users to initiate consultation for abnormal reports;

[0116] Life intervention assistance unit 135 for recording diet, monitoring exercise and reminding medication for the target user;

[0117] Diet record: support voice / text input daily diet (such as "breakfast 2 steamed buns + 1 cup of milk"), APP automatically estimates carbohydrate intake, and associates blood pressure changes;

[0118] Exercise monitoring: synchronize mobile phone motion sensor data (steps, exercise duration), analyze the short-term impact of exercise on vascular stiffness index (such as the change of vascular stiffness index 2 hours after exercise);

[0119] Medication reminder: set medication alarm according to user medication history, record blood pressure / vascular stiffness changes after medication, and assist in evaluating drug efficacy;

[0120] Preferably, the life intervention assistance unit 135 further comprises a medication record and evaluation subunit, which supports users to manually input or associate electronic prescriptions to obtain medication information (drug name, dosage, medication time); based on existing monitoring data such as vascular stiffness index, blood pressure, PWV before and after medication, a dynamic evaluation model is constructed to calculate the improvement rate of indicators after medication (such as "vascular stiffness index decrease rate = (before medication-after medication) / before medication x 100%"), generate medication effect report (including improvement level: significant improvement / good / average / ineffective), and synchronize to the trend analysis and report unit 132.

[0121] Further, the life intervention assistance unit 135 further comprises an execution tracking sub-module, which is used for: ①exercise tracking: synchronize mobile phone motion sensor data, record daily exercise duration and steps, and judge whether the exercise target in the intervention plan is completed; ②diet record reminder: push diet record reminder at fixed time every day (such as after dinner), support voice quick input; ③intervention completion rate statistics: generate intervention plan completion rate report every week (such as "this week, exercise target completion rate is 75%, diet record completion rate is 80%"), and adjust subsequent intervention suggestions according to the completion.

[0122] Family sharing function unit 136 for family members to view the monitoring data and warning information of the old people, and realize remote care.

[0123] Specifically, the encrypted data storage unit 131 meets the medical data security standards, supports data storage for more than 1 year and "daily / weekly / monthly / quarterly" report generation, which is convenient for users and doctors to trace long-term changes;

[0124] The trend analysis and report unit 132 generates differentiated reports for different populations: children focus on evaluating cardiovascular development (such as whether HRV is within the normal range), young people report on unhealthy habits (such as blood pressure fluctuations caused by staying up late), middle-aged people report early warnings of vascular hardening (such as PWVβ increase > 5% per month), and old people report the risk of cardiovascular and cerebrovascular events (such as AI > 60% increases the risk of stroke), and the readability of the report is improved by 60% compared to traditional data lists.

[0125] The hierarchical early warning unit 133 presets 24 types of differentiated thresholds for subpopulations, such as 140 mmHg for elderly hypertensive obese men, 130 mmHg for young normal women, to avoid false positives / negatives caused by "uniform thresholds"; push intervention suggestions according to risk levels, low-risk lifestyle adjustments (such as "daily salt < 5g"), medium-risk medical examinations (such as "suggested carotid ultrasound"), and high-risk emergency response (such as contacting family members and pushing hospital navigation);

[0126] The doctor-patient interactive unit 134 breaks down the "user-doctor" data barrier, allowing doctors to remotely view historical data and adjust AI model weights (such as increasing HRV in vascular hardening assessment for diabetic patients), and users can initiate online consultations to achieve "home monitoring-doctor guidance" linkage; The life intervention assistance unit 135 correlates monitoring data with diet, exercise, and medication, such as recording "blood pressure increased by 10 mmHg 2 hours after high-salt diet", to help users identify unhealthy habits and improve health management initiative.

[0127] The family health record sub-module supports users adding family members (with authorization), correlating monitoring data of family members, automatically analyzing the incidence trend of vascular hardening within the family (such as "three people in the family have atherosclerosis, with a high genetic risk"), generating a family risk report, and supporting sharing of early warning information among family members (high-risk notification to family members simultaneously).

[0128] Each module works together through low-power wireless communication (Bluetooth / BLE / Wi-Fi) to form a full closed-loop system of "signal collection-processing-individualized calculation-risk early warning-long-term management" without human intervention to complete multi-parameter synchronous monitoring and health management.

[0129] Based on the present embodiment:

[0130] More comprehensive parameter dimensions: 6 types of core parameters are collected, the vascular hardening evaluation features are expanded from 3 to 8, and the evaluation dimensions cover "central-peripheral blood vessels", "morphology-function-metabolism", with an accuracy improvement of 40% compared to existing technologies;

[0131] Wider population adaptability: Covers 24 subgroups of population + users with underlying diseases, with an evaluation error of <5% for children, a missed diagnosis rate of <8% for the elderly, an adaptability improvement of 50% for obese people, and a coverage rate of 98% for the whole population;

[0132] Wider protection scope: The claims do not involve specific numerical values, and the core protection is the innovative method of "multi-parameter + population AI model", which avoids being evaded by fine-tuned parameters and improves protection by 60%;

[0133] More complete management functions: Upgraded from single monitoring to "collection-analysis-warning-intervention-cooperation" whole-cycle management, with a user compliance improvement of 70% and a meeting of the long-term management needs of chronic diseases;

[0134] Higher clinical value: The output of blood pressure, vascular stiffness index, and risk prediction results meets clinical standards and can be used as an auxiliary basis for doctor's diagnosis, reducing unnecessary outpatient examinations.

[0135] For example, Sample 1: An elderly hypertensive obese male (68 years old, BMI 29, taking antihypertensive drugs) is monitored:

[0136] The radar module is placed at a regular distance in front of the user, and the user attributes (elderly male, BMI 29, 10-year history of hypertension, taking amlodipine) are input through the mobile phone APP;

[0137] Signal collection and processing:

[0138] Collection site: neck, chest, wrist;

[0139] Raw signal: pulse wave (rising slope 0.4V / s, second peak amplitude 0.6V, second peak delay time 0.15s), heart rate 72 beats / min, respiration 18 times / min, body surface temperature field (neck 36.2℃, wrist 35.1℃, temperature difference 1.1℃);

[0140] Noise reduction processing: Eliminate body motion interference (user's slight hand tremor), signal-to-noise ratio improved to 35dB;

[0141] Feature extraction: HRV (SDNN=28ms, RMSSD=15ms), pulse pressure=138-86=52mmHg, pulse pressure index=52 / 138≈0.38;

[0142] Four-order AI evaluation:

[0143] Population segmentation: divided into "elderly hypertensive obese male group";

[0144] Dynamic weighting: pulse wave transit time difference weight 40%, second peak amplitude weight 25%, pulse pressure weight 30%, HRV weight 5%;

[0145] Multi-model calculation:

[0146] Blood pressure: systolic pressure 138 mmHg, diastolic pressure 86 mmHg (error ± 2 mmHg, compared with cuff blood pressure meter);

[0147] Vascular stiffness index: beta index 13.5, PWV beta = 1180 cm / s, AI = 58%;

[0148] Risk prediction: vascular stiffness index may increase to beta = 14.0 in the next 15 days (close to the warning threshold 14.5);

[0149] Cross-validation: correlation coefficient of systolic pressure and PWV beta = 0.88, meeting the verification requirements;

[0150] Mobile phone APP management:

[0151] Report generation: prompt "vascular stiffness index close to the warning threshold of the elderly hypertensive and obese group, blood pressure control meets the standard (after taking medicine)";

[0152] Early warning intervention: push the medium-risk suggestion "visit a doctor for blood lipid recheck within 1 week and adjust the antihypertensive drug dose";

[0153] Doctor-patient cooperation: after the doctor checks the data, remotely adjusts the pulse pressure weight to 35%, and leaves a message "suggest adding statin drugs to reduce blood lipids";

[0154] Life record: the user inputs "dinner with high-salt sausages", and the APP prompts "blood pressure may rise the next day, suggest monitoring morning peak blood pressure".

[0155] Example 2: monitoring of a child (8 years old, BMI 17.5, no underlying diseases)

[0156] The radar module uses a detachable antenna specially designed for children (small interval), and the APP inputs the user's attributes (child male, BMI 17.5, no diseases);

[0157] Signal acquisition and processing:

[0158] Acquisition site: neck, wrist;

[0159] Feature extraction: HRV (SDNN = 65 ms, RMSSD = 40 ms), body surface temperature field temperature difference 0.5℃, pulse wave rising slope 0.9V / s (children have good vascular elasticity, with a large slope);

[0160] AI evaluation:

[0161] Population segmentation: "child normal male group", HRV weight 35%, temperature field temperature difference weight 25%;

[0162] Calculation result: blood pressure 92 / 60 mmHg, vascular stiffness index (β=6.2, PWVβ=850 cm / s, both in the normal range for children);

[0163] APP management:

[0164] Report: "cardiovascular development is normal, and HRV indicators are excellent";

[0165] Intervention: push the suggestion of "ensure 1 hour of outdoor exercise every day to promote vascular development";

[0166] Family sharing: parents can view data in real time and receive no abnormal warnings.

[0167] Preferably, the embodiment further provides a vascular stiffness evaluation index system, including but not limited to:

[0168] I. Vascular stiffness evaluation index based on pulse wave

[0169] (1) Reflection wave enhancement index (AI) - evaluation based on the "second peak" of the pulse wave;

[0170] (2) Systolic / diastolic pressure estimation based on pulse wave (auxiliary evaluation of stiffness);

[0171] II. Vascular stiffness evaluation index based on blood pressure parameters

[0172] (1) Pulse pressure (PP) and pulse pressure index (PPI);

[0173] (2) Systolic / diastolic pressure ratio (SBP / DBP);

[0174] III. Vascular stiffness-related indicators based on ECG and cardiac function

[0175] (1) Heart rate variability (HRV) related indicators (indirectly reflect vascular stiffness);

[0176] (2) Left ventricular ejection fraction (LVEF) and vascular stiffness correlation (indirect evaluation);

[0177] IV. Blood pressure / pulse wave measurement methods at different sites (related to vascular stiffness evaluation);

[0178] (1) Four-limb blood pressure measurement (comparative evaluation of peripheral vascular stiffness);

[0179] (2) Carotid ultrasound-related indicators (direct observation of vascular stiffness);

[0180] V. Vascular stiffness-related indicators based on heart sounds

[0181] (1) Second heart sound (S2) splitting and intensity change;

[0182] (2) The presence of the fourth heart sound (S4);

[0183] Six. ECG-based vascular stiffness-related indicators

[0184] (1) Left ventricular hypertrophy (LVH)-related electrocardiogram indicators;

[0185] (2) Myocardial ischemia-related electrocardiogram changes (T wave, ST segment abnormalities);

[0186] Seven. Heart function (ultrasound cardiogram-based) -based vascular stiffness-related indicators

[0187] (1) Left ventricular diastolic function indicators;

[0188] (2) Left ventricular mass index (LVMI);

[0189] (3) Aortic root internal diameter (AO) and aortic elasticity;

[0190] Eight. Blood pressure dynamic monitoring-based vascular stiffness-related indicators

[0191] (1) Abnormal circadian rhythm of blood pressure (non-dipper blood pressure);

[0192] (2) Abnormal morning blood pressure peak (morning hypertension);

[0193] Nine. Blood biochemical-based vascular stiffness-related indicators

[0194] (1) Lipid metabolism-related indicators;

[0195] (2) Inflammation and oxidative stress indicators;

[0196] (3) Glucose metabolism and insulin resistance indicators;

[0197] Ten. Imaging-based vascular stiffness-related indicators

[0198] (1) Coronary CT angiography (CCTA) related indicators;

[0199] (2) Aortic CTA / MRI-related indicators;

[0200] (3) Vascular endothelial function ultrasound (FMD);

[0201] Eleven. Emerging technology-based vascular stiffness-related indicators

[0202] (1) Endothelial progenitor cell (EPC) count;

[0203] (2) microRNA (microRNA) detection;

[0204] Eleven. Emerging technology-based vascular stiffness-related indicators

[0205] (i) Endothelial progenitor cell (EPC) count;

[0206] (ii) microRNA detection;

[0207] Twelve, Quantitative indices based on vascular calcification

[0208] (i) Aortic calcification score (AAC);

[0209] (ii) Ankle vascular calcification (AVC);

[0210] Thirteen, Advanced indices based on hemodynamics

[0211] (i) Central arterial pressure (CAP);

[0212] (ii) Arterial compliance (AC);

[0213] Fourteen, Assessment based on clinical signs

[0214] (i) Vascular bruits;

[0215] (ii) Femoral artery pulse delay;

[0216] Fifteen, Comprehensive risk score based on multiple indices

[0217] (i) Cardiovascular risk score (e.g., ASCVD risk score).

[0218] In this embodiment, based on the multi-site millimeter wave radar array layout, the "center-peripheral blood vessels" are covered, the more comprehensive blood vessel physiological characteristics are obtained, the blood vessel hardening evaluation accuracy is improved by 25%, based on the multi-parameter synchronous acquisition (including temperature field), the blood flow and metabolism related parameters are supplemented, the problem of single traditional parameter is solved, the early vascular hardening recognition sensitivity is improved by 30%, based on the adaptive signal adjustment unit 102, the special groups such as children, the elderly and the obese are adapted, the signal acquisition success rate is improved from 80% to 98%, based on the multi-dimensional combination noise reduction algorithm, the signal signal-to-noise ratio is improved by 40%, the body motion interference rejection rate is 98%, the problem of poor effect of traditional single filtering is solved, based on the three-level hierarchical feature extraction logic, the "original signal-clinical parameter" complete transformation is formed, 8 types of features are provided for AI evaluation, and the error caused by feature missing is avoided, based on the 24 types of subdivided population division, the problem of poor adaptability of traditional "rough classification" is solved, the evaluation error of the child group is reduced from 15% to 5%, the misdiagnosis rate of the elderly group is reduced from 20% to 8%, based on the population dynamic weighting model, the "one-size-fits-all" weight is avoided, the evaluation accuracy is improved by 35% compared with the unified model, the blood pressure calculation error of the obese group is reduced from ±4mmHg to ±2.5mmHg, based on the three-module linkage AI calculation (blood pressure-hardening-risk), the blood pressure error is ±2.5mmHg, the vascular hardening index error is ±0.3, and the risk prediction accuracy is 85%, which meets the clinical reference standard, based on the population difference early warning and report, the false positive rate is reduced by 45%, the report readability is improved by 60%, the user is more easy to understand the health status, based on the doctor-patient cooperation + life intervention closed loop, the "monitoring" is upgraded to "management", the user compliance is improved by 70%, and the long-term management demand of chronic diseases is met.

[0219] Preferably, the multi-parameter cooperative cardiovascular health monitoring 100 further comprises a site-to-site signal time difference calculation module and a human parameter association module. The site-to-site signal time difference calculation module accurately captures the arrival time of the same pulse wave at different sites (time resolution ≤0.1 μs) by using the existing pulse wave signals of the neck, chest, wrist and ankle collection points, and automatically calculates the conduction time difference of the carotid artery-wrist and carotid artery-ankle. The human parameter association module is used to support user input of height and arm length, automatically estimate the physiological distance between the collection sites, and combine the conduction time difference to derive the pulse wave conduction velocity (PWV) in real time. The PWV is taken as a core feature and included in the multi-dimensional physiological feature parameters.

[0220] In this embodiment, the multi-dimensional physiological signals of the target user can be effectively collected by the millimeter wave radar collection module 10, the quality of the multi-dimensional physiological signals can be improved by carrying out noise reduction processing on the multi-dimensional physiological signals, the multi-dimensional physiological feature parameters can be effectively extracted by carrying out hierarchical feature extraction on the multi-dimensional physiological signals after noise reduction, the diversity of the physiological feature parameters is improved, the phenomenon of low cardiovascular health monitoring accuracy caused by single collection parameter is prevented, the target user can be effectively classified by carrying out population segmentation on the target user, based on the segmented population, the weight proportion of each feature parameter in the multi-dimensional physiological feature parameters can be dynamically adjusted, the accuracy of the multi-dimensional physiological feature parameters is effectively improved for different populations, and the accuracy of the cardiovascular health monitoring result is improved, and the cardiovascular health monitoring accuracy is improved.

[0221] Embodiment two

[0222] Please refer to Figure 6 is the flow chart of the multi-parameter cooperative cardiovascular health monitoring method provided by the second embodiment of the present application. The multi-parameter cooperative cardiovascular health monitoring method can be applied to any device or system. The multi-parameter cooperative cardiovascular health monitoring method comprises the following steps:

[0223] In step S10, the physiological signals of different parts of the target user are synchronously collected to obtain multi-dimensional physiological signals, and the multi-dimensional physiological signals are subjected to noise reduction processing.

[0224] In this embodiment, the multi-dimensional physiological signals are subjected to noise reduction processing by using a combination algorithm of wavelet transform, Kalman filtering and independent component analysis, so as to eliminate the influence of body motion interference, environmental noise and temperature drift on the multi-dimensional physiological signals.

[0225] Optionally, the physiological signals of different parts of the target user are synchronously collected to obtain multi-dimensional physiological signals, which comprises the following steps:

[0226] The blood vessel depth, skin thickness and fat coverage of the monitored part of the target user are obtained, and the adjustment parameters are determined according to the blood vessel depth, skin thickness and fat coverage.

[0227] The radar transmission frequency, signal gain and sampling period of the millimeter wave radar are adjusted according to the adjustment parameters, and the pulse wave, heart rate and respiration signal of the carotid artery of the neck, the aorta of the chest, the radial artery of the wrist and the posterior tibial artery of the ankle of the target user, and the body surface temperature field distribution data and hemodynamic correlation signal of the neck, chest, wrist and ankle are synchronously collected by the millimeter wave radar after adjustment, to obtain the multi-dimensional physiological signals.

[0228] Obtaining a monitoring posture of the target user, determining a preset scene calibration parameter according to the monitoring posture, and performing parameter calibration on the multi-dimensional physiological signal according to the preset scene calibration parameter.

[0229] Step S20, performing hierarchical feature extraction on the multi-dimensional physiological signal after noise reduction to obtain multi-dimensional physiological feature parameters, and performing crowd segmentation on the target user to obtain a segmented crowd.

[0230] Optionally, the hierarchical feature extraction on the multi-dimensional physiological signal after noise reduction to obtain multi-dimensional physiological feature parameters comprises:

[0231] Extracting the heart rate, respiratory rate, and body surface temperature field temperature difference in the multi-dimensional physiological signal after noise reduction, and extracting the rising edge slope, falling edge slope, rising time from valley to first peak, falling time from first peak to valley, second peak amplitude, second peak appearance delay time, and pulse wave conduction time difference in the multi-dimensional physiological signal after noise reduction.

[0232] According to the heart rate, respiratory rate, body surface temperature field temperature difference, rising edge slope, falling edge slope, rising time from valley to first peak, falling time from first peak to valley, second peak amplitude, second peak appearance delay time, and pulse wave conduction time difference, the heart rate variability, pulse pressure, and pulse pressure index are calculated to obtain the multi-dimensional physiological feature parameters.

[0233] Step S30, according to the segmented crowd, dynamically adjusting the weight proportion of each feature parameter in the multi-dimensional physiological feature parameters to obtain a dynamically adjusted feature parameter.

[0234] Among them, the crowd attribute information of the target user is obtained, and the target user is segmented according to the crowd attribute information to obtain the segmented crowd, and the weight proportion of each feature parameter in the multi-dimensional physiological feature parameters is dynamically adjusted according to the segmented crowd to obtain a dynamically adjusted feature parameter, and the crowd attribute information includes age, gender, body shape, underlying disease, and medication history.

[0235] Step S40, according to the dynamically adjusted feature parameter, performing blood pressure calculation, blood vessel hardening calculation, and blood vessel risk prediction to obtain a blood vessel health monitoring result.

[0236] The systolic pressure and the diastolic pressure are calculated through a gradient boosting tree algorithm according to the pulse wave feature, heart rate variability and crowd attribute information in the dynamically adjusted characteristic parameters, and the target blood pressure is obtained, the pulse wave feature includes a rising slope and a second peak amplitude; the vascular hardening index is output through a random forest algorithm according to the pulse wave transmission time difference, the second peak appearance delay time, the pulse pressure index and the body surface temperature field temperature difference in the dynamically adjusted characteristic parameters; the blood pressure and the vascular hardening data of the target user within a preset time range are obtained, and the vascular risk prediction is performed according to the blood pressure and the vascular hardening data, so that the vascular risk prediction trend is obtained, the consistency verification and the crowd matching verification are performed on the target blood pressure and the vascular hardening index, and the double verification results are obtained, and the vascular health monitoring result includes the target blood pressure, the vascular hardening index, the vascular risk prediction trend and the double verification results.

[0237] In the embodiment, the quality of the multi-dimensional physiological signals is improved through noise reduction processing, the multi-dimensional physiological characteristic parameters are effectively extracted through hierarchical feature extraction on the noise-reduced multi-dimensional physiological signals, the diversity of the physiological characteristic parameters is improved, the phenomenon of low cardiovascular health monitoring accuracy caused by single collection parameter is prevented, the target user is effectively classified through crowd segmentation, the weight proportion of each characteristic parameter in the multi-dimensional physiological characteristic parameters is dynamically adjusted based on the segmented crowd, the accuracy of the multi-dimensional physiological characteristic parameters is effectively improved for different crowds, and the accuracy of the vascular health monitoring result is improved, and the cardiovascular health monitoring accuracy is improved.

[0238] Embodiment three

[0239] Figure 7 is a structural block diagram of a terminal device 2 provided by the third embodiment of the present application. As shown in the figure, Figure 7 The terminal device 2 of the embodiment includes a processor 20, a memory 21, and a computer program 22 stored in the memory 21 and executable on the processor 20, such as a multi-parameter cooperative cardiovascular health monitoring method program. The processor 20 implements the steps in each of the above-mentioned multi-parameter cooperative cardiovascular health monitoring method embodiments when executing the computer program 22.

[0240] For example, the computer program 22 can be divided into one or more modules, which are stored in the memory 21 and executed by the processor 20 to complete the present application. The one or more modules can be a series of computer program instruction segments that can complete a specific function, which are used to describe the execution process of the computer program 22 in the terminal device 2. The terminal device can include, but is not limited to, the processor 20 and the memory 21.

[0241] The processor 20 can be a central processing unit (CPU), and can also be other general-purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0242] The memory 21 can be an internal storage unit of the terminal device 2, such as a hard disk or a memory of the terminal device 2. The memory 21 can also be an external storage device of the terminal device 2, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the memory 21 can include both the internal storage unit and the external storage device of the terminal device 2. The memory 21 is used to store the computer program and other programs and data required by the terminal device. The memory 21 can also be used to temporarily store data that has been output or is to be output.

[0243] In addition, each functional module in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0244] The integrated module, if implemented in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Among them, the computer readable storage medium can be non-volatile or volatile. Based on such understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. The computer program can be executed by a processor to implement the steps of each method embodiment described above. Among them, the computer program includes computer program code, which can be in the form of source code, object code, executable file or some intermediate form, etc. The computer readable storage medium can include any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable storage medium does not include electric carrier signal and telecommunication signal.

[0245] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A multi-parameter collaborative cardiovascular health monitoring system, characterized in that, The system includes: Millimeter-wave radar acquisition module, used to acquire multi-dimensional physiological signals of target users; The three-level signal processing module is used to perform noise reduction processing on the multi-dimensional physiological signals and to perform hierarchical feature extraction on the noise-reduced multi-dimensional physiological signals to obtain multi-dimensional physiological feature parameters. The fourth-order population adaptation AI assessment module is used to segment the target users to obtain segmented populations. Based on the segmented populations, the weight ratio of each feature parameter in the multi-dimensional physiological feature parameters is dynamically adjusted to obtain dynamically adjusted feature parameters. Based on the dynamically adjusted feature parameters, blood pressure calculation, arteriosclerosis calculation and vascular risk prediction are performed to obtain vascular health monitoring results. The three-level signal processing module includes: The multi-dimensional noise reduction unit uses a combination of wavelet transform, Kalman filtering and independent component analysis algorithms to denoise the multi-dimensional physiological signals, eliminating the influence of body movement interference, environmental noise and temperature drift on the multi-dimensional physiological signals; The hierarchical feature extraction unit includes a basic physiological layer, a pulse wave feature layer, and a derived parameter layer; The basic physiological layer is used to extract heart rate, respiratory rate, and body surface temperature field difference from the noise-reduced multi-dimensional physiological signals. The pulse wave feature layer is used to extract the rising slope, falling slope, rising time from the trough to the first peak, falling time from the first peak to the trough, second peak amplitude, second peak appearance delay time, and pulse wave conduction time difference from the multidimensional physiological signal after noise reduction. A derived parameter layer is used to calculate heart rate variability, pulse pressure, and pulse pressure index; The multidimensional physiological characteristic parameters include heart rate, respiratory rate, body surface temperature field temperature difference, rising slope, falling slope, rising time, falling time, second peak amplitude, second peak appearance delay time, pulse wave conduction time difference, heart rate variability, pulse pressure, and pulse pressure index. The fourth-order population adaptation AI assessment module includes: The audience segmentation unit is used to obtain the audience attribute information of the target user and segment the target user according to the audience attribute information to obtain the segmented audience. The audience attribute information includes age, gender, body type, underlying diseases, and medication history. A dynamic weighting unit is used to dynamically adjust the weight ratio of each feature parameter in the multi-dimensional physiological feature parameters according to the segmented population, so as to obtain the dynamically adjusted feature parameters. The multi-model fusion computing unit is used to calculate systolic blood pressure and diastolic blood pressure based on the pulse wave features, heart rate variability and population attribute information in the dynamically adjusted feature parameters, and to obtain the target blood pressure by using the gradient boosting tree algorithm. The pulse wave features include the rising slope and the second peak amplitude. The system is used to output the arteriosclerosis index by using a random forest algorithm based on the pulse wave conduction time difference, second peak appearance delay time, pulse pressure index, and body surface temperature field temperature difference among the dynamically adjusted feature parameters. This is used to acquire the blood pressure and arteriosclerosis data of the target user within a preset time range, and to predict vascular risk based on the blood pressure and arteriosclerosis data to obtain a vascular risk prediction trend. The cross-validation and optimization unit is used to perform consistency verification and population matching verification on the target blood pressure and the arteriosclerosis index to obtain dual verification results. The vascular health monitoring results include the target blood pressure, the arteriosclerosis index, the vascular risk prediction trend, and the dual verification results.

2. The multi-parameter collaborative cardiovascular health monitoring system as described in claim 1, characterized in that, The millimeter-wave radar acquisition module includes: The radar array unit employs a millimeter-wave radar with a multi-transmit and multi-receive antenna layout. It is used to simultaneously acquire pulse waves, heart rate, and respiratory signals from the carotid artery in the upper neck, the aorta in the chest, the radial artery in the wrist, and the posterior tibial artery in the ankle of the target user, as well as surface temperature field distribution data and hemodynamic correlation signals of the neck, chest, wrist, and ankle, to obtain the multi-dimensional physiological signals. An adaptive signal conditioning unit is used to automatically adjust the radar transmission frequency, signal gain, and sampling period according to the blood vessel depth, skin thickness, and fat coverage of the monitored area, so as to adapt to the target users of different body types. A multi-scenario calibration unit is used to acquire the monitoring posture of the target user, determine preset scenario calibration parameters based on the monitoring posture, and eliminate the interference of scenario posture on signal acquisition based on the preset scenario calibration parameters.

3. The multi-parameter collaborative cardiovascular health monitoring system as described in claim 1, characterized in that, The multi-parameter collaborative cardiovascular health monitoring system also includes a full-cycle management module, which includes: An encrypted data storage unit, supporting local storage and cloud backup, is used to store the vascular health monitoring results; The trend analysis and reporting unit is used to generate personalized health reports for different population segments. A tiered early warning unit is used to determine the risk level based on preset differentiated early warning thresholds for the segmented population, and to push intervention suggestions based on the risk level: The doctor-patient collaborative interaction unit is used for data sharing, remote parameter adjustment, and online consultation between the target user and the doctor. The lifestyle intervention support unit is used to record the target user's diet, monitor exercise, and remind them of medication.

4. The multi-parameter collaborative cardiovascular health monitoring system as described in claim 2, characterized in that, The radar array unit adjusts the antenna spacing through detachable antenna components to adapt to different monitoring needs.

5. The multi-parameter collaborative cardiovascular health monitoring system as described in claim 3, characterized in that, The full-cycle management module also includes a family sharing function unit, which allows family members to view the elderly person's monitoring data and early warning information, enabling remote care.

Images