CN111973164A - Intelligent wearable monitoring device and system and training method of blood pressure estimation model - Google Patents

Abstract

The invention relates to an intelligent wearable monitoring device, a system and a training method for a blood pressure estimation model. The intelligent wearable device includes a clothing body and a first physiological sensor, a second physiological sensor, a third physiological sensor, a first physiological sensor, a second physiological sensor, a third physiological sensor, A signal preprocessing module, a digital processing module, a data storage module, a communication module and a power supply module, wherein the first physiological sensor, the second physiological sensor and the third physiological sensor are used to collect physiological signals of the monitored object, and the signal preprocessing module is used for Preprocessing the collected physiological signals; the digital processing module is used to digitize the preprocessed physiological signals, extract feature points based on the digitized physiological signals, and calculate the physiological parameters of the monitoring object based on the feature points . The smart wearable monitoring device provides an effective diagnostic basis for active health management of the human body and disease prevention, diagnosis and treatment without interfering with daily life.

Description

Intelligent wearable monitoring device, system and training method of blood pressure estimation model

technical field

The invention relates to the fields of intelligent wearable equipment and medical measurement, in particular to an intelligent wearable monitoring equipment, a system and a training method for a blood pressure estimation model.

Background technique

With the aging of the population and the prevalence of chronic diseases, the modern society bears huge medical costs. The core of improving this situation is to achieve early diagnosis and treatment of diseases and to improve patients' awareness of actively participating in health management, that is, 4P medical care ( Preventive, Predictive, Personalized, Participatory Medicine).

A core technology for the development of 4P medical care and active health management is wearable medical devices. With the development of technologies such as sensing, electronic integration and wireless communication, medical sensing devices are becoming more and more miniaturized, and they have begun to enter people's daily lives from hospitals, actively and continuously for a long time without interfering with people's daily activities. Collect health information to realize active and continuous long-term monitoring of human health status.

Wearable medical devices fill the gap of out-of-hospital long-term vital parameter monitoring, and provide key technologies for long-term monitoring and early warning of major diseases, especially chronic diseases.

Blood pressure is an important risk indicator of cardiovascular disease. At present, the clinical measurement methods are mainly based on the cuff type, which is characterized in that only one blood pressure value can be measured once inflated and deflated. A single blood pressure measurement in the hospital is difficult to reflect the real blood pressure of the patient. Although the ambulatory sphygmomanometer based on this principle can provide 24-hour blood pressure measurement, it is still intermittent rather than continuous. It affects users, especially sleep at night, and to a certain extent, the real blood pressure, and also limits its scope of use, especially in application scenarios such as disease prevention. Therefore, the current blood pressure technology is very useful in clinical diagnosis and treatment and health monitoring. big limitations.

Long-term continuous blood pressure monitoring has important clinical diagnosis and treatment and health monitoring value. Compared with traditional office blood pressure, continuous blood pressure can more accurately predict cardiovascular risk events, guide medication more rationally, diagnose nocturnal hypertension, and distinguish white coat hypertension from masked hypertension. In the field of health monitoring, long-term continuous blood pressure monitoring can more actively monitor the user's cardiovascular health status, thereby providing early warning of diseases.

Most wearable products currently on the market, such as wristbands, watches, and clothing, have high wearing comfort, but are usually only used to measure physiological parameters such as heart rate and blood oxygen saturation based on ECG and pulse waves. Continuous blood pressure monitoring is not possible. In the field of wearable continuous blood pressure monitoring, some products such as SOMNOtouch ^TM NIBP and Shenzhen Changsang Technology Co., Ltd.'s cuffless blood pressure monitor can realize continuous blood pressure monitoring and have passed the medical device certification, but the design of both products involves a The finger sleeves clipped on the fingers and the wires connected to the ECG patch are not suitable for wearable scenarios, which limits their use.

A literature search on the prior art found that patents in similar technical fields mainly include two categories: smart devices for cuffless blood pressure monitoring and smart clothing for monitoring multiple physiological parameters. For cuffless blood pressure monitoring smart devices, the wearable carrier design is not suitable for long-term comfortable use. For example, the technical solution disclosed in the Chinese invention patent with the authorization announcement number CN101708121B requires at least two sensing devices, namely an ear clip and a watch, and a wire connection is required between the two devices, which is not suitable for daily long-term monitoring. For smart clothing monitoring multiple physiological parameters, most patents are utility models, emphasizing that by integrating multiple physiological sensors with textile clothing, the function of real-time monitoring of health status can be achieved and a certain degree of comfort can be guaranteed, but these documents usually do not consider blood pressure. long-term monitoring and related technical implementations. For example, the authorization announcement number is CN206949578U Chinese utility model patent provides a smart clothing, belonging to the field of clothing technology, which can monitor the human body's muscle fatigue, heart rate, temperature and posture and other physiological and behavioral information, but does not mention the specific technical implementation method . Another example is the technical solution disclosed in the Chinese utility model patent with the authorization announcement number CN209074581U, which integrates a blood oxygen sensor body temperature sensor, an electrocardiogram sensor, a myocardial sensor, and an EEG sensor on the clothing for comprehensive health monitoring. The authorization announcement number is CN206151438U. The technical solution disclosed in the Chinese utility model patent can collect breath, blood oxygen and body temperature, and the sensor and conductive fiber are covered with a layer of flexible material. The monitoring clothing proposed in the Chinese invention patent application with publication number CN109222926A can monitor blood pressure signals, but its blood pressure calculation method is completely based on the waveform of the pulse wave, and the calculation method is relatively simple, which is easily affected by motion artifacts, and usually adopts flexible Electrodes or sensors made of materials often have lower signal quality than traditional electrodes or sensors. Therefore, in the existing technical solutions, there is a contradiction between the comfort of the wearable device and the accuracy of the measured signal quality and physiological parameters.

SUMMARY OF THE INVENTION

(1) Technical problems solved

Aiming at the contradiction between the existing patents and products in long-term wearing comfort and the measurement accuracy of physiological parameters, the patent of the present invention provides a training method for an intelligent wearable monitoring device, a system and a blood pressure estimation model, which can monitor blood pressure for a long time without disturbance. Physiological parameters. In terms of comfort and convenience in use, the present invention adopts conductive fabric or conductive film as electrodes, and conductive fabric embedded in clothing as connecting wire to ensure wearing comfort. The convenience of the system is guaranteed. At the same time, two power supply methods are used, namely, flexible battery and near-magnetic field wireless energy power supply, while ensuring the battery life and wearability of the system. In terms of the accuracy of physiological parameter monitoring, this patent designs the position of the pulse wave sensor near the outer side of the jacket, and the area of the ECG electrode is large, thereby reducing the influence of motion artifacts on the physiological signal. In the continuous blood pressure estimation model, without manual feature extraction, ECG signal, pulse wave signal and motion signal are directly used as model input to estimate blood pressure. In the training of the continuous blood pressure estimation model, a combination of crowd training and personalized training is adopted, and the data set and model can be updated in a timely manner to reduce the impact of changes in the physiological state of the human body on the accuracy of the model.

(2) Technical solutions

To achieve the above object, the present invention provides the following technical solutions:

An embodiment of the present invention provides an intelligent wearable monitoring device, including a clothing body and a first physiological sensor, a second physiological sensor, a third physiological sensor, a signal preprocessing module, and a digital processing module assembled on the clothing body. , data storage modules, communication modules, and power modules, where:

The first physiological sensor, the second physiological sensor and the third physiological sensor are used to collect the physiological signals of the monitored object, wherein the first physiological sensor is made of conductive fabric or conductive film, and is used to collect cardiac signals. The electrical signal includes at least two independent conductive electrodes, wherein at least one of the conductive electrodes is assembled on the front chest or back of the garment body or covers the front chest and back at the same time, and at least one of the conductive electrodes is assembled on the front chest and back. the arm of the garment body; the second physiological sensor is used to collect pulse wave signals, and is assembled at the outer sides of the two upper arms of the garment body; the third physiological sensor is used to collect body motion signals, and is assembled at A plurality of parts of the garment body including the upper arm, the forearm, the front chest and the back, and at least one of the third physiological sensors is respectively assembled in the vicinity of the first physiological sensor and the second physiological sensor;

The signal preprocessing module is used for preprocessing the collected physiological signal;

The digital processing module is used to digitize the preprocessed physiological signal, extract feature points based on the digitized physiological signal, and calculate the physiological parameters of the monitoring object based on the feature points, wherein the physiological parameters include blood pressure, at least one of blood pressure change rate, heart rate, heart rate change rate, and blood oxygen saturation;

the data storage module is used for storing the physiological signal and the physiological parameter;

The communication module is used to communicate with an intelligent health management module, wherein the intelligent health management module is used to display the physiological parameters, and the intelligent monitoring management module is an independent device located outside the intelligent wearable monitoring device. The module is either a part of the smart wearable monitoring device disposed on the garment body;

The power supply module is used for powering the first physiological sensor, the second physiological sensor, the third physiological sensor, the signal preprocessing module, the digital processing module, the storage module and the communication module;

The digital processing module is used to control the acquisition state of the first physiological sensor, the second physiological sensor, and the third physiological sensor, wherein the acquisition state includes stop acquisition, intermittent acquisition and continuous acquisition.

Obviously, the wearable monitoring device can realize long-term undisturbed monitoring of multiple physiological parameters.

Optionally, the first physiological sensor, the second physiological sensor, and the third physiological sensor are all detachable relative to the garment body.

Optionally, the wearable monitoring device further includes a near field communication coil and an energy storage module assembled on the garment body, wherein the near field communication coil is used for receiving wireless energy, and the energy storage module is used for The wireless energy received by the near field communication coil is stored, and the wireless energy received by the near field communication coil comes from the intelligent health management module or other wireless charging devices.

Optionally, the near field communication coil is made of conductive fabric.

Optionally, the power module uses a flexible battery, which is detachable.

Optionally, the first physiological sensor, the second physiological sensor, the third physiological sensor, the signal preprocessing module, the digital processing module, the data storage module, the communication module and Conductive fabrics are used as connecting wires between the power modules.

Optionally, the signal preprocessing module, the digital processing module, the data storage module and the communication module are all detachable relative to the garment body.

Optionally, the communication between the communication module and the intelligent health management module includes: the communication module is configured to automatically upload the updated data in the data storage module to the intelligent health management module, and receive The control instruction of the intelligent health management module.

Optionally, the intelligent health management module is further configured to communicate with a cloud server, and the intelligent health management module and the cloud server jointly run a blood pressure estimation model to estimate the blood pressure of the object to be monitored.

Optionally, the training of the blood pressure estimation model is implemented on the intelligent health management module and the cloud server, wherein the cloud server is used to store the population training data and the population blood pressure estimation of the blood pressure estimation model. Model parameters, the intelligent health management module is used to store the individual training data of the blood pressure estimation model and the parameters of the individual blood pressure estimation model.

Optionally, the training method of the blood pressure estimation model includes:

Pre-training the model by using the crowd training data on the cloud server to obtain a crowd blood pressure estimation model, and transmitting the model parameters of the crowd blood pressure estimation model to the intelligent health management module;

Using the individual training data on the intelligent health management module to train the population blood pressure estimation model to obtain an individual blood pressure estimation model, wherein the individual blood pressure estimation model is used to estimate the to-be-monitored model on the intelligent health management module Subject's blood pressure.

Optionally, the training method of the blood pressure estimation model includes:

Using the crowd training data to train the model on the cloud server to obtain the crowd blood pressure estimation model;

transmitting the individual training data to the cloud server through the intelligent health management module;

The cloud server uses the individual training data to train the population blood pressure estimation model to obtain an individual blood pressure estimation model, and transmits the model parameters of the individual blood pressure estimation model to the intelligent health management module, wherein the The individual blood pressure estimation model is used to estimate the blood pressure of the subject to be monitored on the intelligent health management module.

Optionally, the training method of the blood pressure estimation model includes:

The individual training data is transmitted to the cloud server through the intelligent health management module, the individual training data is added to the crowd training data on the cloud server to form new crowd training data, and the new The population training data is used for the subsequent training of the population blood pressure estimation model.

Wherein, the blood pressure estimation model can be implemented as a long-term continuous cuffless blood pressure estimation model.

Another embodiment of the present invention provides an intelligent wearable monitoring system, which includes the above-mentioned intelligent health management module and the above-mentioned intelligent wearable monitoring device.

Yet another embodiment of the present invention provides a method for training a blood pressure estimation model, which is suitable for smart wearable monitoring devices, wherein the training of the blood pressure estimation model is implemented on an intelligent health management module and a cloud server, wherein the The cloud server is used for storing the population training data of the blood pressure estimation model and the parameters of the population blood pressure estimation model, and the intelligent health management module is used for storing the individual training data of the blood pressure estimation model and the parameters of the individual blood pressure estimation model.

Optionally, the training method includes:

Pre-training the model by using the crowd training data on the cloud server to obtain a crowd blood pressure estimation model, and transmitting the model parameters of the crowd blood pressure estimation model to the intelligent health management module;

Using the individual training data on the intelligent health management module to train the population blood pressure estimation model to obtain an individual blood pressure estimation model, wherein the individual blood pressure estimation model is used to estimate the to-be-monitored model on the intelligent health management module Subject's blood pressure.

Optionally, the training method includes:

Using the crowd training data to train the model on the cloud server to obtain the crowd blood pressure estimation model;

transmitting the individual training data to the cloud server through the intelligent health management module;

The cloud server uses the individual training data to train the population blood pressure estimation model to obtain an individual blood pressure estimation model, and transmits the model parameters of the individual blood pressure estimation model to the intelligent health management module, wherein the The individual blood pressure estimation model is used to estimate the blood pressure of the subject to be monitored on the intelligent health management module.

Optionally, the training method includes: transmitting the individual training data to the cloud server through the intelligent health management module, and adding the individual training data to the crowd training data on the cloud server to form a new The new population training data is used for the subsequent training of the population blood pressure estimation model.

Among them, using this training method can realize the training of long-term continuous cuffless blood pressure estimation model.

(3) Beneficial effects

Compared with the prior art, the present invention provides an intelligent wearable monitoring device, a system and a training method for a blood pressure estimation model, which can monitor multiple physiological parameters without disturbance for a long time, and has the following beneficial effects: under the condition of not interfering with daily life Through the selection and design of sensor materials and positions, high-quality acquisition of three signals of ECG, pulse wave and motion signal can be achieved, providing an effective diagnostic basis for active health management of the human body and disease prevention, diagnosis and treatment.

Description of drawings

Fig. 1 is the principle block diagram of a kind of intelligent wearable monitoring system provided by the present invention;

2 is a schematic front view of an embodiment of the smart wearable system provided by the present invention;

3 is a schematic diagram of the reverse side of an embodiment of the smart wearable monitoring system provided by the present invention;

4 is a flowchart of a blood pressure estimation model training method;

FIG. 5 is a flowchart of a second method of training a blood pressure estimation model.

In the figure: 1 clothing body, 2 first physiological sensor, 3 second physiological sensor, 4 integrated chip, 5 third physiological sensor, 6 near field communication coil, 7 power module.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

1-5, an intelligent wearable monitoring system provided by the present invention includes a first physiological sensor 2, a second physiological sensor 3, a third physiological sensor 5, a signal preprocessing circuit module, a digital processing module, and a storage module , communication module and energy storage module, power module 7, intelligent health management module and cloud server.

The first physiological sensor 2 is used to collect ECG signals, adopts conductive fabric or conductive film, and includes more than two independent conductive electrodes, one of which is assembled at the arm, and the other electrodes are assembled on the front and left of the garment in a belt-like manner. The side or the left and right sides of the back or the front chest and back can be washed and disassembled at the same time; the second physiological sensor 3 is used to collect the pulse wave signal, which is assembled on the left arm or the right arm, and can be quickly disassembled when cleaning the clothing; The third physiological sensor 5 is used to collect body motion signals, and is distributed in many parts of the clothing, including the upper arm, front chest, back, etc. At least one is installed near the first physiological sensor 2 and the second physiological sensor 3. When cleaning the clothing It can be quickly disassembled; the first physiological sensor, the second physiological sensor and the third physiological sensor are installed on the clothing in detachable forms such as snap-on buckles or Velcro, and the corresponding parts on the clothing are already connected to them; the first physiological sensor The sensor, the second physiological sensor, the third physiological sensor, the signal preprocessing module, the digital processing module, the data storage module, the communication module, and the power module use conductive fabrics as connecting wires to connect the components, and the wires can be connected to the components. Plugging and unplugging; the signal preprocessing module amplifies and filters the collected signal, which can be realized through the integrated chip 4 or through a discrete circuit system. The processed signal is sent to the digital processing module, which is the main control center and micro-processing center of the system. It can control the acquisition state of the sensor according to the instructions of the intelligent health management module, that is, stop acquisition, intermittent acquisition and continuous acquisition, and can also According to the quality of the collected signal, the acquisition state of the sensor is determined by itself, and the physiological signal processed by the signal preprocessing module can be digitized, and the feature points can be extracted to calculate the multiple physiological parameters, including heart rate, heart rate change rate and blood oxygen saturation. The obtained physiological parameters will be stored in the storage module and transmitted to the intelligent health management module through the communication module when necessary, or not stored, and the physiological parameters will be directly transmitted to the intelligent health management module through the communication module. With the cooperation of the cloud server, the final physiological parameters can be obtained.

In an implementation manner, the intelligent health management module and the cloud server are independent of the clothing body 1, other modules are integrated on the clothing, and the intelligent health management module is a mobile phone or a computer terminal. In another feasible implementation manner, the intelligent health management module can also be directly installed on the garment body 1 . An application program can be run on the intelligent health management module to display the processed physiological parameters, and the system will automatically give corresponding medical advice according to the obtained parameters. At the same time, users can also set the acquisition status through the application program, check the remaining power of the system and control the wireless energy transmission status.

In an implementation manner, the power supply mode of the system includes two modes: flexible battery power supply and near-magnetic field wireless power supply. The flexible battery can be quickly disassembled during cleaning, and the near field communication coil 6 is made of conductive fabric, which can be directly washed with water, and is assembled on the left and right sides of the clothes near the trouser pocket. The coils of the clothing can automatically receive mobile phones or other devices with power supply capability, and store them in the energy storage module, which can supply power to the system through the power module. In some other implementation manners, the power supply manner of the system may also include only one of flexible battery power supply and near-magnetic field wireless power supply.

The first physiological sensor 2 and all the wires of the system are made of conductive fabric, which can be washed directly. The second physiological sensor 3 , the third physiological sensor 5 , the integrated chip 4 and the power module 7 can all be quickly disassembled when cleaning the clothing.

The blood pressure estimation model training is implemented in the intelligent health management module and cloud server. By collecting population blood pressure data and uploading it to the cloud server, the population blood pressure estimation model is obtained through training, and the blood pressure data and model are updated regularly. In the cloud server or the intelligent health management module, the individual data is combined with the population data for training to obtain an accurate individual blood pressure model. The individual data will be continuously updated, and the individual blood pressure estimation model will also be continuously updated. The intelligent health management module can estimate individual blood pressure according to the latest individual blood pressure model. The addition of population samples reduces the influence of individuals on the accuracy of blood pressure estimation due to their own vascular status, health status and other factors.

Through the selection and design of sensor materials and positions, this intelligent wearable monitoring system can achieve high-quality acquisition of three signals of ECG, pulse wave and motion signal, providing active health management of the human body and disease prevention, diagnosis and treatment. Effective diagnostic basis to achieve long-term undisturbed multi-physiological parameter intelligent monitoring. In addition, the smart wearable monitoring system can combine local and cloud storage and computing resources to achieve accurate calculation of physiological parameters such as blood pressure, blood pressure change rate, heart rate, heart rate change rate, and blood oxygen saturation through the blood pressure estimation model. Health management and disease prevention, diagnosis and treatment provide a more effective basis for diagnosis.

Another embodiment of the present invention further provides an intelligent wearable monitoring device, which together with the above-mentioned intelligent health management module and cloud service constitutes the above-mentioned intelligent wearable monitoring system. The specific composition and functions of the smart wearable monitoring device can be found in the above description, which will not be repeated here. To make the description clearer, only the specific composition of the smart wearable monitoring device will be described below.

In a specific implementation, the smart wearable monitoring device includes a clothing body and a first physiological sensor, a second physiological sensor, a third physiological sensor, a signal preprocessing module, a digital processing module, a data Storage modules, communication modules, and power modules, including:

The first physiological sensor, the second physiological sensor and the third physiological sensor are used to collect the physiological signals of the monitored object, wherein the first physiological sensor is made of conductive fabric or conductive film, and is used to collect cardiac signals. The electrical signal includes at least two independent conductive electrodes, wherein at least one of the conductive electrodes is assembled on the front chest or back of the garment body or covers the front chest and back at the same time, and at least one of the conductive electrodes is assembled on the front chest and back. the arm of the garment body; the second physiological sensor is used to collect pulse wave signals, and is assembled at the outer sides of the two upper arms of the garment body; the third physiological sensor is used to collect body motion signals, and is assembled at A plurality of parts of the garment body including the upper arm, the forearm, the front chest and the back, and at least one of the third physiological sensors is respectively assembled in the vicinity of the first physiological sensor and the second physiological sensor;

The signal preprocessing module is used for preprocessing the collected physiological signal;

The digital processing module is used to digitize the preprocessed physiological signal, extract feature points based on the digitized physiological signal, and calculate the physiological parameters of the monitoring object based on the feature points, wherein the physiological parameters include blood pressure, at least one of blood pressure change rate, heart rate, heart rate change rate, and blood oxygen saturation;

the data storage module is used for storing the physiological signal and the physiological parameter;

The communication module is used to communicate with an intelligent health management module, wherein the intelligent health management module is used to display the physiological parameters, and the intelligent monitoring management module is an independent device located outside the intelligent wearable monitoring device. The module is either a part of the smart wearable monitoring device disposed on the garment body;

The power supply module is used for powering the first physiological sensor, the second physiological sensor, the third physiological sensor, the signal preprocessing module, the digital processing module, the storage module and the communication module;

The digital processing module is used to control the acquisition state of the first physiological sensor, the second physiological sensor, and the third physiological sensor, wherein the acquisition state includes stop acquisition, intermittent acquisition and continuous acquisition.

Obviously, the wearable monitoring device can realize long-term undisturbed monitoring of multiple physiological parameters.

Optionally, the first physiological sensor, the second physiological sensor, and the third physiological sensor are all detachable relative to the garment body.

Optionally, the wearable monitoring device further includes a near field communication coil and an energy storage module assembled on the garment body, wherein the near field communication coil is used for receiving wireless energy, and the energy storage module is used for The wireless energy received by the near field communication coil is stored, and the wireless energy received by the near field communication coil comes from the intelligent health management module or other wireless charging devices.

Optionally, the near field communication coil is made of conductive fabric.

Optionally, the power module uses a flexible battery, which is detachable.

Optionally, the first physiological sensor, the second physiological sensor, the third physiological sensor, the signal preprocessing module, the digital processing module, the data storage module, the communication module and Conductive fabrics are used as connecting wires between the power modules.

Optionally, the signal preprocessing module, the digital processing module, the data storage module and the communication module are all detachable relative to the garment body.

Optionally, the communication between the communication module and the intelligent health management module includes: the communication module is configured to automatically upload the updated data in the data storage module to the intelligent health management module, and receive The control instruction of the intelligent health management module.

Optionally, the intelligent health management module is further configured to communicate with a cloud server, and the intelligent health management module and the cloud server jointly run a blood pressure estimation model to estimate the blood pressure of the object to be monitored.

Optionally, the training of the blood pressure estimation model is implemented on the intelligent health management module and the cloud server, wherein the cloud server is used to store the population training data and the population blood pressure estimation of the blood pressure estimation model. Model parameters, the intelligent health management module is used to store the individual training data of the blood pressure estimation model and the parameters of the individual blood pressure estimation model.

Optionally, the training method of the blood pressure estimation model includes:

Pre-training the model by using the crowd training data on the cloud server to obtain a crowd blood pressure estimation model, and transmitting the model parameters of the crowd blood pressure estimation model to the intelligent health management module;

Using the individual training data on the intelligent health management module to train the population blood pressure estimation model to obtain an individual blood pressure estimation model, wherein the individual blood pressure estimation model is used to estimate the to-be-monitored model on the intelligent health management module Subject's blood pressure.

Optionally, the training method of the blood pressure estimation model includes:

Using the crowd training data to train the model on the cloud server to obtain the crowd blood pressure estimation model;

transmitting the individual training data to the cloud server through the intelligent health management module;

The cloud server uses the individual training data to train the population blood pressure estimation model to obtain an individual blood pressure estimation model, and transmits the model parameters of the individual blood pressure estimation model to the intelligent health management module, wherein the The individual blood pressure estimation model is used to estimate the blood pressure of the subject to be monitored on the intelligent health management module.

Optionally, the training method of the blood pressure estimation model includes:

The individual training data is transmitted to the cloud server through the intelligent health management module, the individual training data is added to the crowd training data on the cloud server to form new crowd training data, and the new The population training data is used for the subsequent training of the population blood pressure estimation model.

Wherein, the blood pressure estimation model can be implemented as a long-term continuous cuffless blood pressure estimation model.

In the smart wearable monitoring device of the embodiment of the present invention, by setting the materials and positions of the first physiological sensor, the second physiological sensor and the third physiological sensor, the three signals of electrocardiogram, pulse wave and body motion signal can be realized. High-quality collection provides an effective diagnostic basis for the active health management of the human body and the prevention, diagnosis and treatment of diseases.

An embodiment of the present invention also provides a blood pressure estimation model training method. Wherein, the first implementation method of the blood pressure estimation model training method is shown in Figure 4, and the specific steps include:

S100: Pre-training the model is performed on the cloud server by using the population pre-training data to obtain a population blood pressure estimation model, and transmit the model parameters to the intelligent health management module.

S200: If the conditions for updating the individual blood pressure estimation module are met, the communication module on the garment body 1 transmits the individual training data on the garment body 1 to the intelligent health management module through wireless communication, otherwise, skip to step S400.

S300: Using the individual training data on the intelligent health management module to train the population blood pressure estimation model to obtain the individual blood pressure estimation model.

S400: The individual blood pressure estimation model estimates blood pressure on the intelligent health management module.

S500: Display the result on the intelligent health management module.

The second implementation method of the blood pressure estimation model training method is shown in Figure 5, and the specific steps include:

S10: Pre-training the model is performed on the cloud server using the population pre-training data to obtain a population blood pressure estimation model.

S20: If the conditions for updating the individual blood pressure estimation module are met, the intelligent health management module transmits the individual training data to the cloud server (wherein the individual training data comes from the storage module of the garment body 1, and the communication module on the garment transfers the individual training data to the cloud server). The training data is transmitted to the intelligent health management module through wireless communication, otherwise, skip to step S40.

S30: Use the individual training data to train the population blood pressure estimation model on the cloud server, obtain the individual blood pressure estimation model, and transmit the model parameters to the intelligent health management module.

S40: Estimate blood pressure by using an individual blood pressure estimation model on the intelligent health management module.

S50: Display the result on the intelligent health management module.

Although the embodiments of the present invention have been shown and described, it will be apparent to those skilled in the art that various changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles of the present invention. The scope is defined by the appended claims and their equivalents.

Claims (18)

1. An intelligent wearable monitoring device, characterized in that it comprises a clothing body (1) and a first physiological sensor (2), a second physiological sensor (3), a third physiological sensor (3), a first physiological sensor (2), a second physiological sensor (3), a third A physiological sensor (5), a signal preprocessing module, a digital processing module, a data storage module, a communication module and a power supply module (7), wherein: The first physiological sensor (2), the second physiological sensor (3) and the third physiological sensor (5) are used to collect physiological signals of the monitoring object, wherein the first physiological sensor (2) adopts It is made of conductive fabric or conductive film and is used for collecting electrocardiographic signals, which includes at least two independent conductive electrodes, wherein at least one of the conductive electrodes is assembled on the front chest or back of the clothing body (1) or covers at the same time In the front chest and back position, at least one of the conductive electrodes is assembled at the arm of the garment body (1); the second physiological sensor (3) is used to collect pulse wave signals, and is assembled on the garment body (1). ) at the outer sides of the two upper arms; the third physiological sensor (5) is used to collect body motion signals, and it is assembled at multiple parts of the garment body (1) including the upper arm, the forearm, the chest and the back, and at least one of the third physiological sensor (5) is respectively assembled in the vicinity of the first physiological sensor (2) and the second physiological sensor (3); The signal preprocessing module is used for preprocessing the collected physiological signal; The digital processing module is used to digitize the preprocessed physiological signal, extract feature points based on the digitized physiological signal, and calculate the physiological parameters of the monitoring object based on the feature points, wherein the physiological parameters include blood pressure, at least one of blood pressure change rate, heart rate, heart rate change rate, and blood oxygen saturation; the data storage module is used for storing the physiological signal and the physiological parameter; The communication module is used to communicate with an intelligent health management module, wherein the intelligent health management module is used to display the physiological parameters, and the intelligent monitoring management module is an independent device located outside the intelligent wearable monitoring device. The module is either a part of the smart wearable monitoring device arranged on the garment body (1); The power supply module (7) is used for the first physiological sensor (2), the second physiological sensor (3), the third physiological sensor (5), the signal preprocessing module, and the digital processing module , the storage module and the communication module are powered; The digital processing module is used to control the acquisition states of the first physiological sensor (2), the second physiological sensor (3), and the third physiological sensor (5). 2 . The intelligent wearable monitoring device according to claim 1 , wherein the first physiological sensor ( 2 ), the second physiological sensor ( 3 ), and the third physiological sensor ( 5 ) are all opposite to each other. 3 . It is detachable from the clothing body (1). 3. The intelligent wearable monitoring device according to claim 1, characterized in that, the wearable monitoring device further comprises a near field communication coil (6) and an energy storage module assembled on the clothing body (1), Wherein, the near field communication coil (6) is used for receiving wireless energy, the energy storage module is used for storing the wireless energy received by the near field communication coil (6), and the near field communication coil (6) receives The wireless energy comes from the smart health management module or other wireless charging devices. 4. The intelligent wearable monitoring device according to claim 3, characterized in that, the near field communication coil (6) is made of conductive fabric. 5 . The smart wearable monitoring device according to claim 1 , wherein the power module ( 7 ) uses a flexible battery and is detachable. 6 . 6. The intelligent wearable monitoring device according to claim 1, wherein the first physiological sensor (2), the second physiological sensor (3), the third physiological sensor (5), Conductive fabrics are used as connecting wires between the signal preprocessing module, the digital processing module, the data storage module, the communication module and the power module (7). 7. The intelligent wearable monitoring device according to claim 1, wherein the signal preprocessing module, the digital processing module, the data storage module and the communication module are all relative to the clothing body ( 1) Detachable. 8 . The smart wearable monitoring device according to claim 1 , wherein the communication between the communication module and the smart health management module comprises: the communication module is used to store the data in the data storage module. The updated data is automatically uploaded to the intelligent health management module, and a control instruction from the intelligent health management module is received. 9. The smart wearable monitoring device according to claim 1, wherein the smart health management module is further configured to communicate with a cloud server, and the smart health management module and the cloud server jointly run a blood pressure estimation model to estimate the blood pressure of the subject to be monitored. 10. The smart wearable monitoring device according to claim 9, wherein the training of the blood pressure estimation model is implemented on the smart health management module and the cloud server, wherein the cloud server uses The intelligent health management module is used for storing the population training data of the blood pressure estimation model and the parameters of the population blood pressure estimation model, and the intelligent health management module is used for storing the individual training data of the blood pressure estimation model and the parameters of the individual blood pressure estimation model. 11. The smart wearable monitoring device according to claim 10, wherein the training method of the blood pressure estimation model comprises: Pre-training the model by using the crowd training data on the cloud server to obtain a crowd blood pressure estimation model, and transmitting the model parameters of the crowd blood pressure estimation model to the intelligent health management module; Using the individual training data on the intelligent health management module to train the population blood pressure estimation model to obtain an individual blood pressure estimation model, wherein the individual blood pressure estimation model is used to estimate the to-be-monitored model on the intelligent health management module Subject's blood pressure. 12. The smart wearable monitoring device according to claim 10, wherein the training method of the blood pressure estimation model comprises: Using the crowd training data to train the model on the cloud server to obtain the crowd blood pressure estimation model; transmitting the individual training data to the cloud server through the intelligent health management module; The cloud server uses the individual training data to train the population blood pressure estimation model to obtain an individual blood pressure estimation model, and transmits the model parameters of the individual blood pressure estimation model to the intelligent health management module, wherein the The individual blood pressure estimation model is used to estimate the blood pressure of the subject to be monitored on the intelligent health management module. 13. The smart wearable monitoring device according to claim 10, wherein the training method of the blood pressure estimation model comprises: The individual training data is transmitted to the cloud server through the intelligent health management module, the individual training data is added to the crowd training data on the cloud server to form new crowd training data, and the new The population training data is used for the subsequent training of the population blood pressure estimation model. 14. An intelligent wearable monitoring system, comprising the intelligent health management module and the intelligent wearable monitoring device according to any one of claims 1-13. 15. A training method for a blood pressure estimation model, which is suitable for intelligent wearable monitoring equipment, wherein the training of the blood pressure estimation model is implemented on an intelligent health management module and a cloud server, wherein the cloud server uses The intelligent health management module is used for storing the population training data of the blood pressure estimation model and the parameters of the population blood pressure estimation model, and the intelligent health management module is used for storing the individual training data of the blood pressure estimation model and the parameters of the individual blood pressure estimation model. 16. The training method of the blood pressure estimation model according to claim 15, characterized in that, comprising: Pre-training the model by using the crowd training data on the cloud server to obtain a crowd blood pressure estimation model, and transmitting the model parameters of the crowd blood pressure estimation model to the intelligent health management module; Using the individual training data on the intelligent health management module to train the population blood pressure estimation model to obtain an individual blood pressure estimation model, wherein the individual blood pressure estimation model is used to estimate the to-be-monitored model on the intelligent health management module Subject's blood pressure. 17. The training method of the blood pressure estimation model according to claim 15, characterized in that, comprising: Using the crowd training data to train the model on the cloud server to obtain the crowd blood pressure estimation model; transmitting the individual training data to the cloud server through the intelligent health management module; The cloud server uses the individual training data to train the population blood pressure estimation model to obtain an individual blood pressure estimation model, and transmits the model parameters of the individual blood pressure estimation model to the intelligent health management module, wherein the The individual blood pressure estimation model is used to estimate the blood pressure of the subject to be monitored on the intelligent health management module. 18. The training method of the blood pressure estimation model according to claim 15, characterized in that, comprising: The individual training data is transmitted to the cloud server through the intelligent health management module, the individual training data is added to the crowd training data on the cloud server to form new crowd training data, and the new The population training data is used for the subsequent training of the population blood pressure estimation model.

Images