Non-invasive vital sign monitoring system and method

Abstract

The invention relates to the technical field of vital sign monitoring, and discloses a non-invasive vital sign monitoring system and method.The system comprises a basic vital sign monitoring module, an extended vital sign monitoring module, a hemodynamic monitoring module and a monitoring and early warning module; the basic vital sign monitoring module is used for monitoring breathing related indexes and stress indexes; an oxyhemoglobin saturation monitoring unit of the extended vital sign monitoring module is used for acquiring a PPG signal through a photoelectric volume sensor and monitoring oxyhemoglobin saturation according to the PPG signal; the hemodynamics monitoring module is used for carrying out cuff-free blood pressure monitoring on the basis of pulse transmission time delay and monitoring output quantity per stroke and cardiac output quantity according to the electrocardiogram signal, the PPG signal and the pulse transmission time delay; and the monitoring and early warning module is used for determining the comprehensive health index of the patient according to the monitoring data of the basic vital sign monitoring module, the extended vital sign monitoring module and the hemodynamics monitoring module, and carrying out prediction and early warning on the physiological state of the patient.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of vital sign monitoring, in particular to a non-invasive vital sign monitoring system and method. BACKGROUND

[0002] In clinical and home care, continuous monitoring of patient status is essential for early detection of physiological failure. Traditional devices usually rely on intermittent measurement (such as cuff blood pressure) or single parameter monitoring, which cannot provide the basis for timely intervention for patients. Existing vital sign monitoring systems have realized respiratory and cardiac monitoring based on electrocardiogram, accelerometer and acoustic sensor, and can calculate intelligent respiratory index and stress index, but cannot perform continuous measurement, and cannot monitor hemodynamic parameters (such as stroke volume and cardiac output) and metabolic parameters (such as blood oxygen saturation and temperature), resulting in inaccurate monitoring results and inability to comprehensively monitor patient status. SUMMARY

[0003] The purpose of the present disclosure is to provide a non-invasive vital sign monitoring system and method to at least solve the technical problem that the existing vital sign monitoring system relies on intermittent measurement or single parameter monitoring, cannot comprehensively and accurately monitor vital signs, and cannot provide the basis for timely intervention for patients.

[0004] To solve the above technical problems, the present disclosure provides a non-invasive vital sign monitoring system, comprising: The non-invasive vital sign monitoring system comprises: a basic vital sign monitoring module configured to acquire a respiration-related signal through at least one of an electrocardiogram sensor, a three-axis accelerometer and an acoustic sensor, and monitor respiration-related indicators and stress indicators according to the respiration-related signal; an extended vital sign monitoring module comprising a blood oxygen saturation monitoring unit configured to acquire a PPG signal through a photoelectric volume sensor and monitor blood oxygen saturation according to the PPG signal; a hemodynamic monitoring module comprising a cuffless blood pressure monitoring unit and a hemodynamic parameter monitoring unit, the cuffless blood pressure monitoring unit being configured to perform cuffless blood pressure monitoring based on a pulse transmission time delay, wherein the pulse transmission time delay is determined according to an electrocardiogram signal acquired by the electrocardiogram sensor and a PPG signal acquired by the photoelectric volume sensor; the hemodynamic parameter monitoring unit being configured to monitor stroke volume and cardiac output according to the electrocardiogram signal, the PPG signal and the pulse transmission time delay; The monitoring and early warning module is configured to determine a comprehensive health index of the patient according to the monitoring data of the basic vital sign monitoring module, the extended vital sign monitoring module and the hemodynamic monitoring module, and to predict and warn the physiological state of the patient.

[0005] In some embodiments, the cuffless blood pressure monitoring unit calculates blood pressure according to a preset blood pressure estimation model, wherein the blood pressure estimation model comprises: a basic linear model: wherein BP is the blood pressure of the patient, PTT is the pulse transit time delay, and is a calibration constant of the patient. an advanced nonlinear model: wherein BP is the blood pressure of the patient, PTT is the pulse transit time delay, A and B are calibration constants of the patient, and A, B and n are determined by patient-specific calibration.

[0006] In some embodiments, the hemodynamic parameter monitoring unit calculates stroke volume and cardiac output according to a preset hemodynamic parameter estimation model, wherein the hemodynamic parameter estimation model comprises: a hemodynamic parameter basic model: , , wherein CO is the cardiac output, HR is the heart rate, and SV is the stroke volume, and is a calibration constant;a hemodynamic parameter advanced model: , , wherein .

[0007] In some embodiments, the hemodynamic parameter monitoring unit calculates blood oxygen saturation by dual-wavelength photoplethysmography.

[0008] In some embodiments, the extended vital sign monitoring module further comprises a temperature measurement unit configured to monitor skin temperature by a temperature sensor and to calibrate the skin temperature by a preset temperature calibration formula, wherein the temperature calibration formula is: T = C × V + D, wherein V is the output voltage of the temperature sensor, and C and D are temperature calibration constants.

[0009] In some embodiments, the monitoring and early warning module generates a basic comprehensive health index through a recurrent neural network, wherein a normalization formula of the basic comprehensive health index is: , wherein CHI is the basic comprehensive health index, is a physiological parameter, is a maximum value of the physiological parameter, is a minimum value of the physiological parameter, is a weight of each physiological parameter; and The monitoring and early warning module fuses multi-channel sensor data through a deep learning model to generate a high-level comprehensive health index, wherein a generation formula of the high-level comprehensive health index is: , wherein, is the high-level comprehensive health index, is a multi-parameter vector at time t, is a deep learning model function.

[0010] In some embodiments, the non-invasive vital sign monitoring system further comprises a model calibration module, which calibrates the deep learning model of the high-level comprehensive health index through transfer learning.

[0011] In some embodiments, the non-invasive vital sign monitoring system further comprises a data synchronization module configured to synchronize data streams obtained by each sensor in time through a cross-correlation method.

[0012] In some embodiments, the non-invasive vital sign monitoring system is a low-profile patch or a wearable integrated sensing device.

[0013] The embodiments of the present disclosure also provide a non-invasive vital sign monitoring method applied to a non-invasive vital sign monitoring system, wherein the non-invasive vital sign monitoring system comprises a basic vital sign monitoring module, an extended vital sign monitoring module, a hemodynamic monitoring module, and a monitoring and early warning module, and the method comprises: The basic vital sign monitoring module acquires a respiration-related signal through at least one of an electrocardiogram sensor, a three-axis accelerometer, and an acoustic sensor, and monitors a respiration-related index and a stress index according to the respiration-related signal; The blood oxygen saturation monitoring unit of the extended vital sign monitoring module acquires a PPG signal through a photoelectric volume sensor, and monitors blood oxygen saturation according to the PPG signal; The cuffless blood pressure monitoring unit of the hemodynamic monitoring module performs cuffless blood pressure monitoring based on a pulse transit time, wherein the pulse transit time is determined according to an electrocardiogram signal acquired by the electrocardiograph sensor and a PPG signal acquired by the photoplethysmograph sensor; and the hemodynamic parameter monitoring unit of the hemodynamic monitoring module monitors stroke volume and cardiac output according to the electrocardiogram signal, the PPG signal and the pulse transit time. The monitoring and early warning module determines a comprehensive health index of the patient according to the monitoring data of the basic vital sign monitoring module, the extended vital sign monitoring module and the hemodynamic monitoring module, and predicts and early warns the physiological state of the patient.

[0014] The present disclosure also provides an electronic device comprising at least a memory and a processor, wherein the memory stores a computer program, and the processor implements the non-invasive vital sign monitoring method described above when executing the computer program stored in the memory.

[0015] The present disclosure also provides a computer-readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement the non-invasive vital sign monitoring method described above.

[0016] The non-invasive vital sign monitoring system and method provided by the present disclosure can integrate the non-invasive continuous monitoring of various physiological parameters of the patient by setting the non-invasive vital sign monitoring system to comprise a basic vital sign monitoring module, an extended vital sign monitoring module and a hemodynamic monitoring module, monitoring the respiratory-related indicators and stress indicators by using the basic vital sign monitoring module, monitoring the blood oxygen saturation by using the extended vital sign monitoring module, and performing cuffless blood pressure monitoring and hemodynamic parameter monitoring by using the hemodynamic monitoring module, thereby providing comprehensive and continuous monitoring of the physiological state of the patient. Meanwhile, the monitoring and early warning module can fuse various physiological parameters to timely and accurately warn the health status of the patient, thereby facilitating active and accurate clinical intervention in the hospital and home care environment and early warning and personalized treatment plan for the patient. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or the prior art, brief descriptions will be given below to the drawings needed to be used in the embodiments or prior art descriptions. Obviously, the drawings described below are only some embodiments described in the present disclosure, and other drawings can be obtained by those skilled in the art without creative labor.

[0018] Figure 1 FIG. 1 is a structural schematic diagram of the non-invasive vital sign monitoring system according to an embodiment of the present disclosure; Figure 2 Flowchart of a non-invasive vital sign monitoring method according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0019] Various aspects of the disclosure are now described with reference to the drawings. While the disclosure is amenable to various modifications and alternative forms, specifics thereof have been shown by way of example in the drawings and will be described in detail. It should be understood, however, that the intention is not to limit the disclosure to the particular embodiments described. On the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the scope and spirit of the disclosure as defined by the appended claims.

[0020] It should be understood that various alterations in detail can be made to the embodiments described herein. Therefore, the above description is not to be considered exhaustive, but rather is given as a singular example of the embodiments. Those skilled in the art will recognize that other modifications can be made to the embodiments within the scope and spirit of the disclosure.

[0021] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and, together with the general description of the disclosure given above, and the detailed description of the embodiments given below, serve to explain the principles of the present disclosure.

[0022] These and other characteristics of the present disclosure will become apparent from the following description and appended claims, taken in conjunction with the accompanying drawings.

[0023] It should also be understood that, although the present disclosure has been described in relation to certain specific examples, many other equivalents falling within the scope of the present disclosure will be apparent to those skilled in the art in view of this specification.

[0024] The above and other aspects, features, and advantages of the present disclosure will become apparent from the following detailed description, taken in conjunction with the accompanying drawings, when considered in conjunction with the following detailed description.

[0025] Specific embodiments of the present disclosure are described hereinafter; however, it should be understood that the disclosed embodiments are merely examples of the present disclosure, which can be embodied in various forms. Well-known and / or redundant functions and structures are not described in detail to avoid obscuring the present disclosure unnecessarily. Therefore, specific structural and functional details disclosed herein are not intended to limit the present disclosure but are merely as examples to be used as a basis for the claims and a representative basis for teaching one skilled in the art to variously employ the present disclosure in virtually any appropriate detailed structure.

[0026] The specification can use phrases such as "in one embodiment," "in another embodiment," "in yet another embodiment," or "in other embodiments," which can refer to one or more embodiments of the same or different embodiments of the present disclosure.

[0027] Embodiment One Figure 1 Flowchart of a non-invasive vital sign monitoring system according to an embodiment of the present disclosure is shown. As Figure 1As shown, the embodiment of the present disclosure provides a non-invasive vital sign monitoring system, comprising: a basic vital sign monitoring module 10 configured to acquire a respiration-related signal through at least one of an electrocardiogram sensor, a three-axis accelerometer and an acoustic sensor, and monitor a respiration-related index and a stress index according to the respiration-related signal; an extended vital sign monitoring module 20 comprising a blood oxygen saturation monitoring unit configured to acquire a PPG signal through a photoelectric volume sensor and monitor blood oxygen saturation according to the PPG signal; a hemodynamic monitoring module 30 comprising a cuffless blood pressure monitoring unit and a hemodynamic parameter monitoring unit, the cuffless blood pressure monitoring unit is configured to perform cuffless blood pressure monitoring based on a pulse transit time, wherein the pulse transit time is determined according to an electrocardiogram (ECG) signal acquired by the electrocardiogram sensor and a PPG signal acquired by the photoelectric volume sensor; the hemodynamic parameter monitoring unit is configured to monitor stroke volume and cardiac output according to the electrocardiogram signal, the PPG signal and the pulse transit time; a monitoring and early warning module 40 configured to determine a comprehensive health index of a patient according to monitoring data of the basic vital sign monitoring module 10, the extended vital sign monitoring module 20 and the hemodynamic monitoring module 30, and to predict and warn the physiological state of the patient.

[0028] Specifically, the basic vital sign monitoring module 10 comprises an electrocardiogram sensor, a three-axis accelerometer, an acoustic sensor and a basic vital sign data processing unit. The electrode sheet of the electrocardiogram sensor is in contact with the body surface and can capture the electrocardiogram changes of the patient. The basic vital sign data processing unit can realize R wave detection and QRS complex analysis by using Pan-Tompkins algorithm according to the electrode signal acquired by the electrocardiogram sensor, and monitor the electrocardiogram changes of the patient. The three-axis accelerometer is used to capture the body movement and respiratory movement of the patient, and the basic vital sign data processing unit calculates the respiratory rate and its variability of the patient according to the body movement and respiratory movement. The acoustic sensor can record the chest or abdominal sound of the patient, and the basic vital sign data processing unit extracts the respiration feature assisted by the chest or abdominal sound, and calculates the respiration index (SRI) and the stress index (SI).

[0029] The formula for calculating the respiration index is:

[0030] wherein, , and are the maximum, minimum and average respiratory rates, and AM is the respiratory amplitude.​ The formula for calculating the stress index is:

[0031] in, for The amplitude of the interval histogram, It is its plural.

[0032] The blood oxygen saturation monitoring unit includes a photoplethysmography (PPG) sensor. The PPG sensor consists of a light-emitting diode (LED) and a photodetector. The PPG sensor uses optical sensing technology to detect changes in blood volume, achieving non-invasive physiological parameter monitoring. Specifically, the PPG sensor utilizes the absorption and scattering properties of light to monitor blood dynamics. By illuminating the skin with a light source, the photodetector captures the light reflected or transmitted back by the blood. Under the influence of blood flow, the light attenuates. The PPG sensor captures these light rays with varying levels of reflection and converts them into a digital signal (PPG signal, photoplethysmography signal). The changes in light intensity are then analyzed to determine blood flow characteristics. The PPG signal represents the waveform of blood volume changes over time.

[0033] The extended vital signs monitoring module 20 also includes an extended vital signs data processing unit. This unit calculates blood oxygen saturation (SpO2) using a dual-wavelength photoplethysmography method. Specifically, the calculation ratio R is first calculated.

[0034] in, For the red light exchange component, This represents the DC component of red light. This refers to the AC component of infrared light. This is the DC component of infrared light; Next, the extended vital signs data processing unit calculates blood oxygen saturation according to the empirical formula for blood oxygen saturation. The empirical formula for blood oxygen saturation is:

[0035] Where R is the ratio mentioned above, and A and B are calibration constants.

[0036] In practice, the extended vital signs data processing unit can calculate blood oxygen saturation using multi-wavelength photoplethysmography. Dual-wavelength photoplethysmography is convenient and fast, and can effectively improve data processing efficiency.

[0037] In this embodiment, the PPG sensor acquires peripheral blood volume signals in real time to achieve dual-wavelength signal acquisition, in order to calculate blood oxygen saturation and serve as a distal signal for pulse transmission time delay (PTT).

[0038] The cuffless blood pressure monitoring unit performs cuffless blood pressure monitoring based on Pulse Transit Time (PTT), where PTT is the time interval from the R wave peak on an electrocardiogram (ECG) to the characteristic point of the pulse pulse (PPG). Elevated blood pressure causes arterial wall stiffening and increases pulse wave velocity; therefore, blood pressure can be derived from the PTT. This cuffless blood pressure monitoring method is more comfortable and enables non-invasive, continuous blood pressure monitoring.

[0039] The hemodynamic parameter monitoring unit calculates stroke volume (SV) and cardiac output (CO) using a machine learning model based on the electrocardiogram (ECG) signal, the PPG signal, and the pulse transmission delay, achieving non-invasive and accurate monitoring of these parameters. Stroke volume refers to the amount of blood ejected from one ventricle during a single heartbeat; it is a core indicator of cardiac pumping function. Cardiac output is an important indicator of cardiac pumping function, reflecting the heart's ability to deliver blood throughout the body per minute.

[0040] In addition to monitoring blood pressure, stroke volume, and cardiac output, the extended vital signs monitoring module 20 can also monitor parameters such as heart rate variability (HRV).

[0041] The data processing unit of the monitoring and early warning module 40 comprehensively assesses the patient's physiological state based on different monitoring data from the basic vital signs monitoring module 10, the extended vital signs monitoring module 20, and the hemodynamic monitoring module 30, calculates the patient's comprehensive health index, and then predicts and issues early warnings based on the comprehensive health index. For example, the monitoring and early warning module 40 can determine whether the patient has hypertension and the severity of hypertension based on the blood pressure monitored by the cuffless blood pressure monitoring unit. Since hypertension may affect the patient's heart, the monitoring and early warning module 40 can determine whether the patient's cardiac output capacity is normal based on the stroke volume and cardiac output monitored by the hemodynamic parameter monitoring unit, thereby providing a comprehensive and accurate assessment of the patient's health status for timely intervention.

[0042] It is understandable that the comprehensive health index generated by the monitoring and early warning module 40 may include corresponding health indices generated separately based on monitoring data from the vital signs monitoring module 10, the extended vital signs monitoring module 20, and the hemodynamics monitoring module 30, or it may include a total health index generated by fusing different monitoring data from the vital signs monitoring module 10, the extended vital signs monitoring module 20, and the hemodynamics monitoring module 30. The comprehensive health index can include both quantitative and non-quantitative indicators, thereby enabling a reasonable and accurate assessment of the patient's health status.

[0043] It is understandable that the data processing units of the different monitoring modules mentioned above can be set up separately and integrated with the corresponding sensors to form corresponding monitoring modules; each data processing unit can also be integrated with the monitoring and early warning module 40 to form an independent data processing module, which processes the monitoring data acquired from the sensors. The data processing module can be, for example, an MCU microcontroller, which can process the sensor data collected by each sensor and then calculate the patient's respiratory index, stress index, blood oxygen saturation, blood pressure, stroke volume, cardiac output, and other physiological parameters. The integration of the data processing units of each monitoring module with the corresponding sensors, closer to the sensing nodes, can reduce signal source noise, lower power consumption, and reduce transmission delay.

[0044] In practice, the sensors of different monitoring modules can be integrated together to form a multimodal sensor array. This sensor array includes an electrocardiogram sensor, a triaxial accelerometer, an acoustic sensor, and a photoplethysmography sensor. The corresponding sensor data is acquired in real time, and the data processing module performs calculations on the sensor data to improve data processing efficiency.

[0045] The data processing module (including the monitoring and early warning module 40 and the corresponding data processing units for each monitoring module) continuously collects the raw signals from each sensor. Then, the digital signal processing unit performs preliminary filtering and signal segmentation. The preprocessed signals are then subjected to feature extraction to extract key features for calculating physiological parameters (such as R-wave, PPG amplitude, etc.). Subsequently, the key features are fused and optimized as needed to eliminate motion and noise interference. The extracted parameter features are then input into the preset AI engine, which uses the corresponding physiological parameter monitoring model to calculate physiological parameters. The monitoring and early warning model is then used to fuse the various physiological parameters for predictive analysis, generate a comprehensive health index, and generate corresponding physiological state prediction alarms.

[0046] The monitoring and early warning module 40 also includes a data transmission unit. The data transmission unit can receive monitoring data from each monitoring module in real time through communication methods such as Bluetooth and Wi-Fi. After analyzing and processing the monitoring data, it transmits the generated comprehensive health index and physiological state prediction alerts to mobile devices and clinical dashboards in real time, so that doctors or family members can understand the patient's health status in a timely manner and intervene in treatment in a timely manner. It is particularly suitable for the early prediction and treatment of diseases.

[0047] The monitoring and early warning module 40 also includes a report generation module, which can generate comprehensive health status reports to provide clinicians and patients with detailed health trends and early warning information.

[0048] The non-invasive vital sign monitoring system provided in this embodiment of the invention, by setting up a basic vital sign monitoring module 10, an extended vital sign monitoring module 20, and a hemodynamic monitoring module 30, integrates multiple physiological parameters of the patient for non-invasive continuous monitoring, providing comprehensive and continuous monitoring of the patient's physiological state. Simultaneously, the monitoring and early warning module 40 can integrate multiple physiological parameters to provide timely and accurate early warnings of the patient's health status, facilitating proactive and precise clinical intervention in hospital and home care environments, enabling early warning and personalized treatment plans for patients.

[0049] Preferably, the non-invasive vital signs monitoring system is a low-profile patch or a wearable integrated sensing device. The low-profile patch can be applied to the skin to monitor physiological parameters comfortably and conveniently, and can provide continuous monitoring, thus providing timely, comprehensive, and accurate information on the patient's physiological status. Wearable integrated sensing devices can be finger cots, wristbands, watches, etc.

[0050] In some embodiments, the cuffless blood pressure monitoring unit calculates blood pressure according to a preset blood pressure estimation model, wherein the blood pressure estimation model includes: Basic linear model: Where BP is the patient's blood pressure and PTT is the pulse transit time. and Calibrate constants for patients; Advanced nonlinear models: Where BP is the patient's blood pressure, PTT is the pulse transmission delay, and A and B are patient calibration constants. A, B and n are determined through patient-specific calibration.

[0051] When monitoring blood pressure using a cuffless blood pressure monitoring unit, different blood pressure estimation models are provided to facilitate timely and accurate output of blood pressure parameters. For example, when PTT data is limited (e.g., monitoring time is short) or PTT data noise is low (e.g., few outliers), a basic linear model can be used for blood pressure estimation, improving computational efficiency. Conversely, when PTT data is abundant, an advanced nonlinear model is used for blood pressure estimation, resulting in more accurate calculation of blood pressure parameters. The advanced nonlinear model can be a deep learning regression model.

[0052] When estimating blood pressure, blood pressure models can be established separately for systolic blood pressure (SBP) and diastolic blood pressure (DBP):

[0053]

[0054] in, and Calibrate constants for patients.

[0055] In some embodiments, the hemodynamic parameter monitoring unit calculates stroke volume and cardiac output according to a preset hemodynamic parameter estimation model, wherein the hemodynamic parameter estimation model includes: Basic model of hemodynamic parameters: , , Where CO is cardiac output, HR (Heart Rate) is heart rate, and SV is stroke volume. and This is the calibration constant; Advanced models of hemodynamic parameters: , , in, .

[0056] Similar to blood pressure estimation, this embodiment uses a basic hemodynamic parameter model to input PTT data into the model to calculate stroke volume and cardiac output, suitable for situations with limited or low-noise PTT data. The advanced hemodynamic parameter model utilizes a feature vector X composed of PTT, PPG amplitude, and ECG features to calculate stroke volume and cardiac output through deep learning. This model can combine PPG amplitude and ECG features to more accurately calculate and evaluate PTT data, resulting in more accurate stroke volume and cardiac output.

[0057] In some embodiments, the extended vital signs monitoring module further includes a temperature measurement unit, which is configured to monitor skin temperature via a temperature sensor and calibrate the skin temperature using a preset temperature calibration formula, wherein the temperature calibration formula is: , Where V is the output voltage of the temperature sensor, and C and D are temperature calibration constants.

[0058] In addition to monitoring blood pressure saturation, the extended vital signs monitoring module can also continuously monitor skin temperature through a temperature sensor and calculate the patient's core body temperature through a temperature calibration algorithm.

[0059] In practice, the extended vital signs monitoring module can also monitor other metabolic parameters such as activity level, and the specific types are not specifically limited in this disclosure.

[0060] In some embodiments, the monitoring and early warning module 40 generates a basic comprehensive health index through a recurrent neural network, wherein the normalization formula for the basic comprehensive health index is: , CHI stands for Basic Comprehensive Health Index. Physiological parameters This represents the maximum value of the physiological parameter. This represents the minimum value of the physiological parameter. The weights of each physiological parameter.

[0061] In some embodiments, the monitoring and early warning module 40 utilizes a deep learning model to fuse multi-channel sensor data to generate an advanced comprehensive health index, wherein the formula for generating the advanced comprehensive health index is: , in, As an advanced comprehensive health index, Let be a multi-parameter vector at time t. This is a function for deep learning models.

[0062] The data processing unit of the monitoring and early warning module 40 can combine traditional signal processing methods and deep learning technology to fuse data from various sensors and generate a corresponding comprehensive health index, thus predicting and issuing early warnings about the patient's health status. For example, when the physiological parameters monitored by each monitoring module are normal (within a preset threshold range), a basic comprehensive health index can be generated using the normalization formula of the basic comprehensive health index. When some physiological parameters are abnormal, an advanced comprehensive health index generation model can be used to comprehensively analyze the physiological parameters and provide a more accurate comprehensive health index.

[0063] Furthermore, the non-invasive vital signs monitoring system also includes a model calibration module. The model calibration module calibrates the deep learning model of the advanced comprehensive health index through transfer learning, and realizes the transfer learning and personalized calibration of the model according to multiple physiological parameters of different patients, thereby improving the predictive accuracy of the model.

[0064] In some embodiments, the non-invasive vital signs monitoring system further includes a data synchronization module, which is configured to synchronize the data streams acquired by each sensor in time using a cross-correlation method in order to obtain more accurate health monitoring and early warning results.

[0065] Example 2 Figure 2 A schematic diagram of the structure of a non-invasive vital sign monitoring method according to an embodiment of this disclosure is shown, as follows: Figure 1 and Figure 2 As shown, this disclosure provides a non-invasive vital sign monitoring method applied to a non-invasive vital sign monitoring system. The non-invasive vital sign monitoring system includes a basic vital sign monitoring module 10, an extended vital sign monitoring module 20, a hemodynamic monitoring module 30, and a monitoring and early warning module 40. The method includes: S101: The basic vital signs monitoring module 10 acquires respiratory-related signals through at least one of an electrocardiogram sensor, a triaxial accelerometer, and an acoustic sensor, and monitors respiratory-related indicators and stress indicators based on the respiratory-related signals; S102: The blood oxygen saturation monitoring unit of the extended vital signs monitoring module 20 acquires the PPG signal through a photoplethysmography sensor and monitors blood oxygen saturation based on the PPG signal; S103: The cuffless blood pressure monitoring unit of the hemodynamic monitoring module 30 performs cuffless blood pressure monitoring based on the pulse transmission delay, wherein the pulse transmission delay is determined according to the electrocardiogram signal acquired by the electrocardiogram sensor and the PPG signal acquired by the photoplethysmography sensor; the hemodynamic parameter monitoring unit of the hemodynamic monitoring module 30 monitors the stroke volume and cardiac output based on the electrocardiogram signal, the PPG signal, and the pulse transmission delay; S104: The monitoring and early warning module 40 determines the patient's comprehensive health index based on the monitoring data of the basic vital signs monitoring module 10, the extended vital signs monitoring module 20 and the hemodynamic monitoring module 30, and predicts and warns the patient's physiological state.

[0066] The above steps S101 to S103 can be performed simultaneously to generate comprehensive and accurate physiological state monitoring data, thereby obtaining a more accurate comprehensive health index.

[0067] In some embodiments, in step S103, the cuffless blood pressure monitoring unit of the hemodynamic monitoring module 30 performs cuffless blood pressure monitoring based on pulse transmission delay, including: the cuffless blood pressure monitoring unit calculates blood pressure according to a preset blood pressure estimation model, wherein the blood pressure estimation model includes: Basic linear model: Where BP is the patient's blood pressure and PTT is the pulse transit time. and Calibrate constants for patients; Advanced nonlinear models: Where BP is the patient's blood pressure, PTT is the pulse transmission delay, and A and B are patient calibration constants. A, B and n are determined through patient-specific calibration.

[0068] In some embodiments, in step S103, the hemodynamic parameter monitoring unit of the hemodynamic monitoring module 30 monitors the stroke volume and cardiac output based on the electrocardiogram signal, the PPG signal, and the pulse transmission delay, including: The hemodynamic parameter monitoring unit calculates stroke volume and cardiac output according to a preset hemodynamic parameter estimation model, wherein the hemodynamic parameter estimation model includes: Basic model of hemodynamic parameters: , , Where CO is cardiac output, HR is heart rate, and SV is stroke volume. and This is the calibration constant; Advanced models of hemodynamic parameters: , , in, .

[0069] In some embodiments, in step S102, the extended vital signs monitoring module 20 monitors blood oxygen saturation based on the PPG signal, including: The hemodynamic parameter monitoring unit calculates blood oxygen saturation using a dual-wavelength photoplethysmography method.

[0070] In some embodiments, the extended vital signs monitoring module further includes a temperature measurement unit, and the method further includes: The temperature measurement unit is configured to monitor skin temperature via a temperature sensor and calibrate the skin temperature using a preset temperature calibration formula, wherein the temperature calibration formula is: , Where V is the output voltage of the temperature sensor, and C and D are temperature calibration constants.

[0071] In some embodiments, in step S104, the monitoring and early warning module 40 determines the patient's comprehensive health index based on the monitoring data from the basic vital signs monitoring module 10, the extended vital signs monitoring module 20, and the hemodynamics monitoring module 30, including: The monitoring and early warning module generates a basic comprehensive health index through a recurrent neural network, wherein the normalization formula for the basic comprehensive health index is: , CHI stands for Basic Comprehensive Health Index. Physiological parameters This represents the maximum value of the physiological parameter. This represents the minimum value of the physiological parameter. The weights of each physiological parameter; and The monitoring and early warning module utilizes a deep learning model to fuse multi-channel sensor data to generate an advanced comprehensive health index. The formula for generating this advanced comprehensive health index is as follows: , in, As an advanced comprehensive health index, Let be a multi-parameter vector at time t. This is a function for deep learning models.

[0072] In some embodiments, the non-invasive vital signs monitoring system further includes a model calibration module, and the method further includes: The model calibration module calibrates the deep learning model of the advanced comprehensive health index through transfer learning.

[0073] In some embodiments, the non-invasive vital signs monitoring system further includes a data synchronization module, and the method further includes: The data synchronization module is configured to synchronize the data streams acquired by each sensor in time using a cross-correlation method.

[0074] The non-invasive vital sign monitoring method provided in this disclosure corresponds to the non-invasive vital sign monitoring system of the above embodiments. Any option in the embodiments of the non-invasive vital sign monitoring system is also applicable to the embodiments of the non-invasive vital sign monitoring method, and will not be repeated here.

[0075] Example 3 This disclosure also provides an electronic device, including at least a memory and a processor, wherein the memory stores a computer program, and the processor implements the above-described non-invasive vital sign monitoring method when executing the computer program in the memory.

[0076] In some embodiments, the processor executing a computer program may be a processing device that includes one or more general-purpose processing devices, such as a microprocessor, a central processing unit (CPU), a graphics processing unit (GPU), etc. More specifically, the processor may be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a processor that runs other instruction sets, or a processor that runs a combination of instruction sets. The processor may also be one or more special-purpose processing devices, such as application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), system-on-a-chip (SoCs), etc.

[0077] The memory may be a read-only memory (ROM), random access memory (RAM), phase-change random access memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), electrically erasable programmable read-only memory (EEPROM), other types of random access memory (RAM), flash drives or other forms of flash memory, cache, registers, static memory, optical disc read-only memory (CD-ROM), digital versatile optical disc (DVD) or other optical storage, magnetic tape cassette or other magnetic storage devices, or any other possible non-transitory medium used to store information or instructions that can be accessed by computer equipment.

[0078] The electronic devices disclosed herein may include, but are not limited to, wearable devices such as low-profile patches, smartwatches, and smart bracelets.

[0079] Example 4 This disclosure also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described non-invasive vital sign monitoring method.

[0080] The computer-readable storage medium of this disclosure can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. In this disclosure, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device; for example, it can be the memory described above.

[0081] The computer programs of embodiments of this disclosure can be organized into one or more computer-executable components or modules. Various aspects of this disclosure can be implemented with any number and combination of such components or modules. For example, aspects of this disclosure are not limited to the specific computer-executable instructions or particular components or modules shown in the drawings and described herein. Other embodiments may include different computer-executable instructions or components having more or fewer functions than those shown and described herein.

[0082] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

Claims

1. A non-invasive vital sign monitoring system, characterized in that, include: The basic vital signs monitoring module is configured to acquire respiratory-related signals through at least one of an electrocardiogram (ECG) sensor, a triaxial accelerometer, and an acoustic sensor, and monitor respiratory-related indicators and stress indicators based on the respiratory-related signals. The extended vital signs monitoring module includes a blood oxygen saturation monitoring unit configured to acquire PPG signals through a photoplethysmography (PPG) sensor and monitor blood oxygen saturation based on the PPG signals. The hemodynamic monitoring module includes a cuffless blood pressure monitoring unit and a hemodynamic parameter monitoring unit. The cuffless blood pressure monitoring unit is configured to perform cuffless blood pressure monitoring based on pulse transit delay, wherein the pulse transit delay is determined based on the ECG signal acquired by the ECG sensor and the PPG signal acquired by the PPG sensor. The hemodynamic parameter monitoring unit is configured to monitor stroke volume and cardiac output based on the ECG signal, the PPG signal, and the pulse transit delay. The monitoring and early warning module is configured to determine the patient's comprehensive health index based on the monitoring data from the basic vital signs monitoring module, the extended vital signs monitoring module, and the hemodynamic monitoring module, and to predict and warn about the patient's physiological state.

2. The non-invasive vital sign monitoring system according to claim 1, characterized in that, The cuffless blood pressure monitoring unit calculates blood pressure according to a preset blood pressure estimation model, wherein the blood pressure estimation model includes: a basic linear model: Where BP is the patient's blood pressure and PTT is the pulse transit time. and Patient calibration constants; advanced nonlinear models: Where BP is the patient's blood pressure, PTT is the pulse transmission delay, and A and B are patient calibration constants. A, B and n are determined through patient-specific calibration.

3. The non-invasive vital sign monitoring system according to claim 1, characterized in that, The hemodynamic parameter monitoring unit calculates stroke volume and cardiac output based on a preset hemodynamic parameter estimation model, wherein the hemodynamic parameter estimation model includes: a basic hemodynamic parameter model: , Where CO is cardiac output, HR is heart rate, and SV is stroke volume. and For calibration constants; advanced model of hemodynamic parameters: , ,in, .

4. The non-invasive vital sign monitoring system according to claim 1, characterized in that, The hemodynamic parameter monitoring unit calculates blood oxygen saturation using a dual-wavelength photoplethysmography method.

5. The non-invasive vital sign monitoring system according to claim 1, characterized in that, The extended vital signs monitoring module also includes a temperature measurement unit, which is configured to monitor skin temperature via a temperature sensor and calibrate the skin temperature using a preset temperature calibration formula. The temperature calibration formula is as follows: Where V is the output voltage of the temperature sensor, and C and D are temperature calibration constants.

6. The non-invasive vital sign monitoring system according to claim 1, characterized in that, The monitoring and early warning module generates a basic comprehensive health index through a recurrent neural network, wherein the normalization formula for the basic comprehensive health index is: Among them, CHI stands for Basic Comprehensive Health Index. Physiological parameters This represents the maximum value of the physiological parameter. This represents the minimum value of the physiological parameter. The weights of each physiological parameter are assigned; and the monitoring and early warning module uses a deep learning model to fuse multi-channel sensor data to generate an advanced comprehensive health index, wherein the formula for generating the advanced comprehensive health index is: ,in, As an advanced comprehensive health index, Let be a multi-parameter vector at time t. This is a function for deep learning models.

7. The non-invasive vital sign monitoring system according to claim 6, characterized in that, The non-invasive vital signs monitoring system also includes a model calibration module, which calibrates the deep learning model of the advanced comprehensive health index through transfer learning.

8. The non-invasive vital signs monitoring system according to claim 1, characterized in that, The non-invasive vital signs monitoring system also includes a data synchronization module, which is configured to synchronize the data streams acquired by each sensor in time using a cross-correlation method.

9. The non-invasive vital sign monitoring system according to claim 1, characterized in that, The non-invasive vital signs monitoring system is a low-profile patch or wearable integrated sensing device.

10. A non-invasive method for monitoring vital signs, characterized in that, An application is made in a non-invasive vital sign monitoring system, comprising a basic vital sign monitoring module, an extended vital sign monitoring module, a hemodynamic monitoring module, and a monitoring and early warning module. The method includes: the basic vital sign monitoring module acquiring respiratory-related signals via at least one of an electrocardiogram sensor, a triaxial accelerometer, and an acoustic sensor, and monitoring respiratory-related indicators and stress indicators based on the respiratory-related signals; the extended vital sign monitoring module's blood oxygen saturation monitoring unit acquiring PPG signals via a photoplethysmography (PPG) sensor, and monitoring blood oxygen saturation based on the PPG signals; the hemodynamic monitoring module's non-invasive vital sign monitoring module... The cuff-based blood pressure monitoring unit performs cuffless blood pressure monitoring based on pulse transmission delay, wherein the pulse transmission delay is determined based on the electrocardiogram signal acquired by the electrocardiogram sensor and the PPG signal acquired by the photoplethysmography (PPG) sensor; the hemodynamic parameter monitoring unit of the hemodynamic monitoring module monitors stroke volume and cardiac output based on the electrocardiogram signal, the PPG signal, and the pulse transmission delay; the monitoring and early warning module determines the patient's comprehensive health index based on the monitoring data from the basic vital signs monitoring module, the extended vital signs monitoring module, and the hemodynamic monitoring module, and predicts and warns of the patient's physiological state.

Images