Wristband device and system integrating gesture recognition function and continuous blood pressure monitoring function

Abstract

A wristband device and system integrating a gesture recognition function and a continuous blood pressure monitoring function. The wristband device uses a reusable electrode assembly (120) in a wearable wristband (100) to acquire wrist impedance distribution data and radial artery pulse impedance waveform data in two functional modes, and converts the biological impedance data into human body gestures and blood pressure information by means of an algorithm deployed in an upper computer (400) of the system, so as to monitor the human body physiological state. The wristband device and system use the reusable electrode array to integrate the two functions in one system, thereby reducing sensor volume, complexity and integration difficulty.

Description

Bracelet device and system integrating gesture recognition function and continuous blood pressure monitoring function Technical Field

[0001] The present invention relates to bioimpedance measurement technology, as well as gesture recognition and pulse blood pressure measurement technology for monitoring human physiological status. In particular, it discloses a wearable wristband device and system based on the electrical impedance principle that integrates gesture recognition function and continuous blood pressure monitoring function, belonging to the technical field of measurement and testing. Background Art

[0002] In recent years, user-friendly design of electronic devices has become increasingly important, and wearable electronic devices such as smart bracelets and AR / VR glasses have become very popular. Users can carry these devices with them and use them for various purposes such as human-computer interaction and daily health monitoring.

[0003] As these electronic devices are used, user convenience and comfort are becoming increasingly important optimization priorities. Gesture recognition is a key way to facilitate human-computer interaction. Traditional gesture recognition technologies often rely on computer vision, flexible angular displacement sensors, and electromyographic sensors. However, computer vision-based gesture recognition technology is limited by lighting conditions and computing power requirements, making it difficult to achieve wearable requirements. Flexible angular displacement sensors are typically integrated into gloves for use, but the enveloping nature of the gloves hinders natural hand movement. Electromyographic sensor electrodes require a large contact area with the skin, making them difficult to integrate into small wearable devices. This has necessitated the need for a gesture recognition technology that requires low computing power, is user-friendly, and facilitates integration. This requirement is met by gesture recognition technology based on the principle of electrical impedance.

[0004] For portable wearable devices, daily health monitoring is one of their important functions, and blood pressure is a key indicator of health monitoring. Traditional blood pressure monitoring uses an inflatable cuff-type blood pressure monitor. This type of blood pressure monitor is large and not portable, and a single measurement requires a long process of pressurizing and depressurizing the arm, making it impossible to achieve time-series continuous blood pressure monitoring. Currently, popular smart bracelets on the market generally use blood pressure monitoring solutions based on pulse waves. Some products integrate photoplethysmographs (PPG) sensors into the bracelet to extract pulses and calculate blood pressure. However, this method is fundamentally affected by ambient light and human skin color, resulting in limited accuracy, and is more expensive than impedance methods. Therefore, pulse and blood pressure monitoring solutions based on the principle of electrical impedance are a lower-cost and more popular alternative.

[0005] Wearable devices are becoming increasingly important in our lives, making breakthroughs in improving sensing methods to enhance monitoring accuracy and range, while also reducing the cost and size of wearable devices, particularly important. Electrical impedance sensing technology uses electrodes as sensors, applying an excitation current and collecting the response voltage for sensing. This technology offers low-cost, fast, accurate, and convenient detection. This breakthrough, combining bioelectrical impedance sensing technology with wearable devices, is expected to lead to new breakthroughs in future wearable devices, including reducing cost and size, or improving monitoring accuracy and range.

[0006] Electrodes are a key factor influencing the accuracy of bioelectrical impedance measurement data. Existing bioelectrical impedance measurement devices use electrodes distributed over a fixed area to be measured, resulting in a relatively small number of sensing electrodes. Some devices even place the electrodes in a circle around the limb, effectively monitoring only the cross-sectional area enclosed by the electrodes. This fails to detect impedance changes outside the sensing electrode area, making it unsuitable for complex human conditions. Consequently, existing electrode designs are inadequate for measuring bioelectrical impedance for gesture recognition, hindering the integration of gesture recognition technology with wearable devices.

[0007] When monitoring the weak impedance change signal from the radial artery, the radial artery's position cannot be quickly located due to variations in wrist shape. This deviation significantly impacts the actual measurement results. For sensor arrays composed of a small number of electrodes, manual adjustment of the array's position to align it with the radial artery is often necessary to achieve optimal monitoring. This deviation significantly impacts the actual measurement results, which is both inconvenient for users and reduces the accuracy of continuous pulse blood pressure monitoring. Therefore, the electrode design used in existing bioelectrical impedance measurement technology cannot meet the requirements of continuous pulse blood pressure measurement for wearable devices.

[0008] In summary, electrical impedance-based gesture recognition and pulse blood pressure monitoring technologies have not yet been well integrated with wearable devices. This invention aims to propose a wearable wristband device and system that integrates gesture recognition and continuous blood pressure monitoring functions to overcome these shortcomings.

[0009] Summary of the Invention

[0010] The purpose of the present invention is to address the shortcomings of the above-mentioned background technology and provide a wearable bracelet device and system based on the principle of electrical impedance sensing with integrated gesture recognition function and continuous blood pressure monitoring function, combining electrical impedance sensing technology with wearable devices to realize human-computer interaction and health monitoring functions, and provide a solution for user-friendliness, simplification of sensing front end, and miniaturization of integration of wearable daily health monitoring equipment, broaden the application of electrical impedance sensing technology in the field of health monitoring, and solve the technical problem that the existing gesture recognition technology based on the principle of electrical impedance and the pulse blood pressure monitoring technology based on the principle of electrical impedance cannot be well integrated with wearable devices.

[0011] The present invention adopts the following technical solutions to achieve the above-mentioned purpose:

[0012] A wristband device integrating gesture recognition and continuous blood pressure monitoring functions comprises: a wearable wristband, a PCB board, and a display screen; the wearable wristband comprises at least two wristband components, each comprising: a component body, at least two contact sensing electrodes embedded in the inner wall of the component body, the at least two contact sensing electrodes embedded in the same component body being arranged vertically, and each contact sensing electrode being configured to operate in an excitation mode or an acquisition mode; the PCB board is used to switch between a gesture recognition mode and a continuous blood pressure monitoring mode. In the gesture recognition mode, an excitation current is applied to an excitation current transmitter composed of one electrode selected from each adjacent wristband component, and a response voltage signal fed back by a response voltage receiver composed of one electrode selected from each other adjacent wristband component is collected. The response voltage signal collected in the gesture recognition function mode is modulated into wrist impedance distribution data. In the continuous blood pressure monitoring function mode, an excitation current is applied to an electrode pair on a wristband component near the radial artery and the response voltage signals fed back by the electrode pairs on other wristband components are collected. The optimal response voltage signal collected in the continuous blood pressure monitoring function mode is modulated into radial artery pulse impedance waveform data for each cardiac cycle. The wrist impedance distribution data or the radial artery pulse impedance waveform data for each cardiac cycle is transmitted to the host computer, and the gesture recognition classification results or blood pressure prediction values ​​fed back by the host computer are received. The display screen is used to visualize the user's heart rate and the radial artery pulse impedance waveform data for each cardiac cycle, and is used to visualize the gesture recognition classification results or blood pressure prediction values ​​received by the PCB board.

[0013] As a further optimization solution for a wristband device that integrates gesture recognition and continuous blood pressure monitoring functions, the optimal response voltage signal collected in the continuous blood pressure monitoring mode is obtained by extracting the complex amplitude of the response voltage signal fed back by the electrode pairs on other wristband components as a bioimpedance representation of the artery and its surrounding tissues, continuously collecting the complex amplitude of the response voltage signal fed back by the electrode pairs on other wristband components, and using the complex amplitude of the response voltage signal that continuously changes in time series as a pulse blood flow representation waveform, comparing the amplitudes of the pulse blood flow representation waveforms obtained based on the response voltage signals collected by the electrode pairs on other wristband components, and selecting the pulse blood flow representation waveform with the largest amplitude and the largest peak-to-peak value of the pulse wave signal as the optimal response voltage signal.

[0014] As a further optimization solution for a wristband device that integrates gesture recognition function and continuous blood pressure monitoring function, at least two wristband components are connected by drilling or snapping to form a wearable wristband that wraps around the wrist. The component body is made of materials including but not limited to nylon and silicone; the contact sensing electrode is one of a hemispherical electrode, a square electrode, a patch electrode, and a button electrode.

[0015] As a further optimization solution for a wristband device integrating gesture recognition function and continuous blood pressure monitoring function, the PCB board includes: an excitation source module, a signal demodulation module, a multiplexing module, a control module, a communication module and a power supply module; the excitation source module is used to apply an excitation current to the contact sensing electrode in the excitation mode; the signal demodulation module is used to receive the response voltage feedback from the contact sensing electrode in the acquisition mode, modulate the response voltage signal collected in the gesture recognition function mode into wrist impedance distribution data, and modulate the best response voltage signal collected in the continuous blood pressure monitoring function mode into radial artery pulse impedance waveform data of each cardiac cycle; the multiplexing module cyclically connects any excitation current transmitter to the excitation source module and connects other response voltage receivers to the signal receiver in the gesture recognition function mode. The operation of connecting the signal demodulation module until all excitation current transmitters are applied with excitation current, and in the continuous blood pressure monitoring function mode, the electrode pair on a bracelet component near the radial artery is connected to the excitation source module and the electrode pairs on other bracelet components are connected to the signal demodulation module; the control module is used to control the selection of each channel in the multiplexing module, control the communication between the communication module and the display screen, and control the start and stop of the excitation source module and the signal demodulation module; the communication module is used to transmit wrist impedance distribution data or radial artery pulse impedance waveform data of each cardiac cycle to the host computer, receive the gesture recognition classification results or blood pressure prediction values ​​fed back by the host computer, and transmit the gesture recognition classification results or blood pressure prediction values ​​to the display screen; the power supply module is used to provide the PCB board with the operating voltage and power consumption of the device under full load operation.

[0016] As a further optimization solution for wristband devices that integrate gesture recognition and continuous blood pressure monitoring functions, the excitation source module includes: a waveform lookup table, a digital-to-analog converter, and a voltage-controlled current source; the waveform lookup table is used to generate a unipolar sinusoidal voltage signal; the digital-to-analog converter is used to convert the unipolar sinusoidal voltage signal into an analog signal for output; and the voltage-controlled current source is used to convert the analog signal output by the digital-to-analog converter into a differential current signal for output.

[0017] As a further optimization solution for a wristband device with integrated gesture recognition and continuous blood pressure monitoring functions, the signal demodulation module includes: a differential amplifier, an analog-to-digital converter, and a data demodulator; the differential amplifier is used to differentially amplify the received response voltage signal and then output it; the analog-to-digital converter is used to convert the differential signal output by the differential amplifier into a single-ended signal, and then convert the single-ended signal into a digital signal and output it; the data demodulator is used to extract wrist impedance distribution data or radial artery pulse impedance waveform data for each cardiac cycle from the digital signal output by the analog-to-digital converter.

[0018] As a further optimization solution for the wristband device that integrates gesture recognition and continuous blood pressure monitoring functions, the multiplexing module is implemented through four multiplexer chips. The common terminals of the four multiplexer chips are respectively connected to the two output terminals of the voltage-controlled current source and the two input terminals of the differential amplifier. Each contact sensor electrode is electrically connected to one optional channel of the multiplexer chip, and the address lines that control the channel selection of the four multiplexer chips are connected to the control module.

[0019] As a further optimization solution for the wristband device with integrated gesture recognition function and continuous blood pressure monitoring function, the display is fixed in the mechanical groove, which is seamlessly adhered to the PCB board directly above the mechanical groove through the laminated adhesive for overlapping assembly, and the wearable wristband is electrically connected to the PCB board.

[0020] A system integrating gesture recognition function and continuous blood pressure monitoring function comprises: the above-mentioned wristband device and a PC terminal which wirelessly communicates with the wristband device, wherein the PC terminal comprises: a communication control module, a pulse feature extraction module, a blood pressure prediction module, a wrist imaging operation module, a gesture classification module and a result display module; the communication control module is used to control the start and stop of the communication function between the PC terminal and the wristband device, receive wrist impedance distribution data or radial artery pulse impedance waveform data of each cardiac cycle transmitted by the wristband device, and transmit the gesture recognition classification result or blood pressure prediction value calculated by the PC terminal back to the wristband device; the pulse feature extraction module is used to convert the radial artery pulse impedance waveform data of each cardiac cycle into pulse feature data and construct a pulse feature set; the blood pressure prediction module is used to calculate the blood pressure according to the pulse The feature set is used to train a pre-deployed neural network predictor, and the trained neural network predictor calculates the blood pressure prediction value based on the real-time radial artery pulse impedance waveform data transmitted by the bracelet device; the wrist imaging operation module is used to convert the wrist impedance distribution data transmitted by the bracelet device into wrist cross-sectional impedance change distribution imaging; the gesture classification module is used to train a pre-deployed neural network classifier based on the wrist cross-sectional impedance change distribution imaging, and the trained neural network classifier predicts the gesture recognition classification result based on the real-time wrist impedance distribution data transmitted by the bracelet device; the result display module is used to visualize the wrist impedance distribution data or radial artery pulse impedance waveform data transmitted by the bracelet device, and is used to visualize the gesture classification results predicted by the gesture classification module or the blood pressure prediction value calculated by the blood pressure prediction module.

[0021] As a further optimization solution for systems integrating gesture recognition and continuous blood pressure monitoring, the pulse feature extraction module constructs a pulse feature set that includes pulse feature data and the user's blood pressure value calibrated at the same acquisition time point as the radial artery pulse impedance waveform data. The pulse feature data includes, but is not limited to, the maximum slope, impedance amplitude, time interval, and area enclosed by the impedance amplitude axis and time axis for the four waveform segments: from the start point of the pulse blood flow characteristic wave to the peak of the main wave, from the peak of the main wave to the trough of the dicrotic wave, from the trough of the dicrotic wave to the peak of the dicrotic wave, and from the peak of the replay wave to the end point.

[0022] Compared with the prior art, the technical solution of the present invention has the following advantages and beneficial effects:

[0023] (1) The wristband device proposed in the present invention arranges electrodes into multiple rows and then embeds them into the components of the wearable wristband, thereby increasing the distribution density of electrodes on the human body surface and the contact area between the electrodes and the body part to be measured, so that the effective sensing area is expanded from a single cross-section to the area between two cross-sections, which is suitable for monitoring complex human body conditions; by detecting the electrical impedance of the wrist in real time, gesture recognition and pulse and blood pressure monitoring can be realized, and the clever reuse of electrodes can make one device have these two functions, thereby reducing the size, complexity and integration difficulty of the sensor.

[0024] (2) The design of the electrode array proposed in the present invention makes it unnecessary to adjust the position of the wristband when performing pulse blood pressure monitoring. The high-density distribution of electrodes at the radial artery completely covers the area near the radial artery. There are several electrode pairs in the area near the radial artery for sensing. The peak-to-peak values ​​of the pulse wave characterization waveforms at multiple pairs of response electrodes can be compared to determine the electrode pair with the highest measurement sensitivity to achieve the best sensing effect. Only two pairs of electrodes with the best sensing effect need to be selected, which greatly simplifies the use process, ensures measurement accuracy, and does not require additional adjustment of the wristband position, thereby improving user experience. In addition, compared with the solution of integrating electrodes in gloves, integrating electrodes in watches or wristbands is more practical and comfortable, and is particularly suitable for long-term wear and daily use.

[0025] (3) The arc-shaped raised electrode design proposed in the present invention ensures close contact between the electrode and the skin when the wristband is tightened, thereby improving the stability and accuracy of the measurement and enhancing the wearing experience. This design is more effective than traditional planar electrodes, especially when measuring weak impedance change signals such as pulse or blood pressure monitoring. Weak bioimpedance signals can be extracted through a low-cost, simple electrode array, and the sensitivity of wrist impedance signal monitoring can be improved through optimized electrode configuration.

[0026] (4) Unlike traditional disposable ECG electrodes, the electrodes in the electrode array proposed in the present invention can be reused multiple times and can be integrated into fabrics, making them wearable. This not only reduces costs but also increases user convenience, eliminating the need for frequent electrode replacement or the use of conductive glue.

[0027] (5) The system proposed in the present invention does not need to integrate multiple types of sensors to achieve multi-functions. It only uses a reused electrode array to take into account both gesture recognition and blood pressure monitoring functions, thereby reducing the size, complexity and cost of the sensor.

[0028] (6) The system proposed in the present invention selects the pulse impedance waveform characteristics, including the slope, amplitude, time interval and other information of each stage of a cardiac cycle, and obtains a blood pressure prediction model exclusive to the user through neural network training.

[0029] (7) The system proposed in the present invention provides a wearable smart bracelet solution that can quickly complete human-computer interaction through gestures and perform health monitoring on human blood pressure indicators. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] FIG1 is a schematic diagram of a wristband terminal device provided in an embodiment.

[0031] FIG2 is a schematic diagram of a wristband terminal device in an embodiment of correct wearing.

[0032] FIG3 is a schematic diagram of a wearable wristband band composed of multiple wristband components provided in an embodiment.

[0033] FIG4 is a schematic diagram of the structural design of a single wristband component provided in an embodiment.

[0034] FIG5 is a logic block diagram of a gesture recognition and pulse blood pressure monitoring multiplexing system provided in an embodiment.

[0035] FIG6 is a schematic diagram of an excitation measurement mode in each functional mode provided in the embodiment.

[0036] FIG7 is a schematic diagram of the PC software layout provided in the embodiment.

[0037] FIG8 is a schematic diagram of a neural network classifier structure for gesture recognition on a PC provided in an embodiment.

[0038] FIG9 is a schematic diagram of a neural network predictor structure for pulse to blood pressure conversion on a PC provided in an embodiment.

[0039] Explanation of the accompanying reference numerals: 100, wearable bracelet, 200, PCB board, 300, display screen, 400, PC terminal; 110, hemispherical contact sensor electrode, 120, bracelet component; 101, first contact sensor electrode, 102, second contact sensor electrode, 103, third contact sensor electrode, 104, fourth contact sensor electrode. DETAILED DESCRIPTION

[0040] The technical solutions of the present invention are described in detail below using specific examples and accompanying drawings. Those skilled in the art will readily appreciate the other advantages and benefits of the present invention from the disclosure herein. The present invention may also be implemented or applied through various other specific embodiments, and the details in this specification may be modified or altered based on different viewpoints and applications without departing from the principles of the present invention.

[0041] It should be noted that the diagrams provided in this embodiment are merely illustrative of the basic principles, component structures, working processes, and functions of the present invention. Therefore, the diagrams only show components related to the present invention and are not drawn according to the number, formation, and size of components in actual implementation. In actual implementation, the type, quantity, and proportion of each component may vary, and the component layout may also be more complex.

[0042] 1 to 9 , this embodiment provides a wristband device and system that integrates gesture recognition function and continuous pulse blood pressure monitoring function.

[0043] As shown in FIG1 , the wristband terminal device includes: a wearable wristband 100 , a PCB board 200 , and a display screen 300 .

[0044] As shown in Figure 2, the display screen 300 is fixed in a mechanical groove, which is seamlessly adhered to the PCB board 200 via a laminated adhesive for overlapping assembly. The display screen 300 is bound to the PCB board 200 and placed in the watch case. The wearable wristband 100 passes through the watch case and is electrically connected to the PCB board 200 via wires. When using the above-mentioned wristband terminal device, the user puts the wristband terminal device on the wrist through the wearable wristband 100, with the visible surface of the display screen 300 facing the user.

[0045] In this embodiment, as shown in FIG3 , the wearable wristband 100 includes at least two wristband components 120. As shown in FIG4 , the wristband component 120 includes: a component body, at least two vertically arranged contact sensing electrodes 110, each contact sensing electrode is embedded in the inner wall of the component body, and each wristband component 120 is connected by drilling or snapping to form a wearable wristband 100 that surrounds the wrist. The component realizes the snap connection of adjacent wristband components through the protruding structure and the groove structure on the side of the body. The groove structure is not shown in FIG4 . The groove structure is located on the side of the component body on the opposite side of the protruding structure. When the wearable wristband 100 is fixed on the wrist to measure the electrical impedance, the position of the wearable wristband relative to the wrist remains unchanged, and all electrodes are ensured to maintain reliable contact with the user's skin during the measurement to reduce interference.

[0046] The wearable wristband 100 uses reasonable materials such as nylon and silicone to prepare the component body structure of the embedded electrode, providing good fixing ability.

[0047] In this embodiment, a tag is provided on the wearable wristband 100, which is used to identify the electrode directly above the radial artery to obtain an accurate radial artery impedance change signal.

[0048] The contact sensing electrode at the front end of the bioimpedance sensor has an exposed surface facing the wrist and is approximately 0.5 cm in size. 2 , which can be configured in excitation mode or acquisition mode to inject excitation current into the human wrist or measure the response voltage; the gesture recognition function mode requires that multiple electrodes on the wearable bracelet are arranged into a ring array, which forms a circle to surround the user's wrist during measurement; the pulse blood pressure monitoring function mode requires that the wearable bracelet contain at least two pairs of vertically arranged electrodes, and during measurement, at least two pairs of vertically arranged electrodes are close to the skin above the radial artery about 2 cm below the radial styloid process.

[0049] Specifically, in this embodiment, the wearable wristband 100 includes 16 wristband components 120, each wristband component 120 includes two contact sensing electrodes 110 embedded in the inner wall of the component body, and the contact sensing electrodes on the wearable wristband 100 are arranged in two parallel ring electrodes, each ring electrode including 16 electrodes. Designers can select an appropriate number of electrodes as sensor devices according to different needs, and can also specify different electrode excitation and acquisition modes.

[0050] In this embodiment, the contact sensing electrode 110 adopts a hemispherical electrode to ensure good contact between the electrode and the wrist while improving the user's wearing experience and achieving the effect of massaging the wrist; according to different application scenarios, square electrodes, patch electrodes, button electrodes, etc. can also be used, without limitation.

[0051] In this embodiment, the contact sensing electrode 110 is made of copper plated with silver chloride, which is low-cost, has good conductivity, is corrosion-resistant, and is harmless to the human body. Depending on the designer's needs, the contact sensing electrode 110 can also be made of other conductive materials that are harmless to the human body.

[0052] In this embodiment, the PCB board 200 has a structure as shown in FIG5 , which is an impedance acquisition hardware circuit board, and includes an excitation source module, a signal demodulation module, a multiplexing module, a control module, a communication module, and a power supply module.

[0053] The excitation source module is used to inject an excitation current into the electrodes on the contact wrist. This excitation current signal is a sinusoidal current signal within the safe current range of a selected frequency. As shown in Figure 5, the excitation source module obtains a unipolar sinusoidal voltage signal with adjustable amplitude and frequency from a waveform lookup table. This unipolar sinusoidal voltage signal is converted to digital-to-analog via a digital-to-analog converter. The analog signal output by the digital-to-analog converter is input to a voltage-controlled current source, which performs a voltage-to-current conversion on the input analog signal to generate a differential current signal. This differential current signal is then injected into the contact sensing electrodes via a multiplexing module.

[0054] The signal demodulation module is used to extract the response voltage signal fed back by the wrist contact sensor electrode. As shown in Figure 5, the signal demodulation module collects the response voltage signal from the contact sensor electrode. The response voltage signal is a weak sinusoidal voltage signal loaded at the same frequency as the excitation source module. The response voltage signal is converted from differential to single-ended and analog to digital through a differential amplifier and an analog-to-digital converter, and the noise is filtered out. Finally, the digital phase-sensitive demodulation module extracts the accurate bioelectrical impedance signal. In the gesture recognition function mode, the signal demodulation module modulates the response voltage signal collected by the electrode pairs composed of electrodes in all adjacent bracelet components into wrist impedance distribution data; in the pulse blood pressure monitoring function mode, the signal demodulation module modulates the time-series continuously changing response voltage signal collected by the electrode pair closest to the radial artery into radial artery pulse impedance waveform data for each cardiac cycle.

[0055] The multiplexing module is used to enable the selection of multiple contact sensing electrodes and switch between excitation and acquisition modes. More specifically, under the coordination of the control module, this module switches the selected contact sensing electrode pairs. In excitation mode, it receives the differential current output by the excitation source module. In acquisition mode, it transmits the acquired response voltage signal to the signal demodulation module, achieving multiplexing of multi-function and multi-excitation modes. This module can use four 32-to-1 multiplexer chips, which can be analog switch chips or relays. The common terminals of the four multiplexer chips are connected to the two output terminals of the voltage-controlled current source and the two input terminals of the differential amplifier, respectively. The 32 selectable channels are connected to the 32 electrodes. The address lines for the four multiplexer chips to control channel selection are connected to the control module.

[0056] The control module is used to coordinate the work of other modules. More specifically, the control module is mainly used to control the electrode selection of the multiplexing module, control the communication module to communicate with the display screen and PC, and control the start and stop of other modules.

[0057] The communication module is used to transmit the collected bioelectrical impedance signal to the PC, receive the calculation results from the PC, and transmit the calculation results from the PC to the display screen.

[0058] The power supply module is used to provide the entire PCB board with the operating voltage and power consumption of the devices under full load operation, and is not shown in Figure 5.

[0059] More specifically, the module of the impedance acquisition hardware circuit board also adopts the following specific implementation method:

[0060] The control module is mainly implemented in the MCU. The selected MCU is the STM32 series. Other series of qualified MCUs or FPGAs can also be used as the main control chip to complete the control and calculation of the device;

[0061] The communication module is implemented by Bluetooth to ensure the portability of the device. The communication module can also adopt other feasible communication solutions, such as Ethernet communication.

[0062] The impedance acquisition hardware circuit board also features a user control knob that connects to the MCU and controls the multiplexing module. The knob allows users to select the device's functional mode, which can be configured as gesture recognition mode or pulse and blood pressure monitoring mode.

[0063] In this embodiment, the multiplexing module implements the gating of the contact sensing electrodes 110 and the switching of the excitation and measurement modes, as shown in FIG6 . The excitation and measurement modes in each functional mode include:

[0064] In gesture recognition mode, when one electrode is selected from any two adjacent wristband components among multiple wristband components arranged in a circle around the wrist to form an electrode pair and the electrode pair is configured as an excitation current transmitter, one electrode is selected from each of the remaining adjacent wristband components to be configured as a response voltage receiver responsible for measuring the response voltage. The electrode pairs consisting of one electrode each on all adjacent wristband components are circulated and configured as excitation current transmitters until all excitation current transmitters are applied with an overexcitation current signal;

[0065] In the pulse blood pressure monitoring function mode, a pair of electrodes on a wristband component near the radial artery - the first contact sensing electrode 101 and the second contact sensing electrode 102 are selected as the differential excitation electrode pair, and a differential excitation current is input to the differential excitation electrode pair. At the same time, the electrode pair on the same wristband component near the differential excitation electrode pair is configured to the acquisition mode to measure the response voltage. The complex amplitudes V1, V2, and V3 of the demodulated response voltage signals are used as the bioimpedance representation of the artery and its surrounding tissues. The complex amplitudes are continuously collected and the complex amplitudes of the response voltage signals that continuously change in time sequence are used as the waveform representing the pulse blood flow. The waveform is compared with the root cause. According to the amplitude of the pulse blood flow characterization waveform obtained from the response voltage signals collected by different response electrode pairs, the response electrode pair with the largest amplitude and the largest peak-to-peak value of the pulse wave signal is selected as the optimal collection electrode pair. For example, according to the pulse blood flow characterization waveform obtained from the response voltage signal collected by the response electrode pair composed of the third contact sensing electrode 103 and the fourth contact sensing electrode 104, the pulse blood flow characterization waveform with the largest amplitude and the largest peak-to-peak value of the pulse blood flow characterization wave signal is continuously collected from the response electrode pair composed of the third contact sensing electrode 103 and the fourth contact sensing electrode 104, and the alternating voltage that continuously changes in time sequence is used as the optimal response voltage signal.

[0066] In this embodiment, the display screen 300 is implemented using a small-sized LCD screen, which is used to display the human pulse waveform and heart rate in real time, and display the gesture recognition results or blood pressure monitoring values ​​calculated by the PC end; specifically, the calculation results are obtained through Bluetooth communication with the communication module of the PCB board.

[0067] This embodiment provides a system integrating gesture recognition function and continuous pulse blood pressure monitoring function, as shown in FIG2 , which includes: a wristband device and a cloud or APP end or PC end 400 that wirelessly communicates with the wristband device.

[0068] The cloud, APP or PC 400 receives the bioelectrical impedance measurement signal detected by the wristband terminal device and returns the calculation result of the neural network operator to the wristband terminal device.

[0069] In this embodiment, the PC terminal 400 can be an ordinary desktop computer, a laptop computer or a tablet computer, and the type is not limited; when the system adopts the PC terminal 400, the PC terminal 400 is configured with the software module shown in Figure 7.

[0070] Specifically, the software modules configured in the PC terminal 400 include: a communication control module, a pulse wave feature extraction module, a blood pressure prediction module, a wrist imaging calculation module, a gesture classification module and a result display module.

[0071] The communication control module is used to control the start and stop of the communication function between the PC and the wristband terminal device, receive the bioelectrical impedance data transmitted by the wristband terminal device, and transmit the gesture classification and blood pressure calculation results of the PC 400 back to the wristband terminal device for display on the display screen 300.

[0072] The pulse feature extraction module is used to convert radial artery impedance waveform data into a pulse feature set, which is used to input into the neural network predictor for blood pressure prediction calculation.

[0073] The blood pressure prediction module is used to calibrate and test blood pressure prediction. Any user needs to perform a blood pressure calibration operation when using the bracelet terminal device for the first time. The steps include: collecting synchronous calibration blood pressure information, calibrating the neural network predictor, and learning the characteristics of the pulse feature set data. The characteristics of the radial artery pulse impedance waveform data of each cardiac cycle include: the maximum slope, impedance amplitude, time interval, and area enclosed by the impedance amplitude axis and the time axis of the four waveforms from the starting point of the pulse blood flow characteristic wave to the peak of the main wave, the peak of the main wave to the trough of the dicrotic wave, the trough of the dicrotic wave to the peak of the dicrotic wave, and the peak of the replay wave to the end point. These characteristics are the same as the radial artery pulse impedance signal. The user's blood pressure value calibrated by the instrument at the time of acquisition is input into the machine learning model for training; after the calibration is completed, the neural network predictor converts the pulse impedance data transmitted from the bracelet terminal into a blood pressure value and displays it on the result display module on the PC and the display screen 300 of the bracelet terminal device.

[0074] The wrist imaging calculation module is used to convert the wrist impedance distribution data from the bracelet terminal device into wrist cross-sectional impedance change distribution imaging through an algorithm. The wrist cross-sectional impedance change distribution imaging is used to input the neural network classifier for gesture recognition calculation.

[0075] The gesture classification module is used for calibration and testing of gesture recognition. Any user needs to perform gesture calibration when using the wristband terminal device for the first time. The steps include: collecting calibration gesture information, calibrating the neural network gesture classifier, and learning the characteristics of wrist cross-sectional electric potential distribution imaging data. After calibration, the neural network classifier classifies the wrist impedance distribution data transmitted by the terminal as gestures. The gesture classification results are displayed on the result display module on the PC and the display screen 300 of the wristband terminal device.

[0076] The result display module is used to visually display the bioelectrical impedance data from the bracelet terminal device and the gesture classification results and blood pressure prediction results after neural network calculation.

[0077] In this embodiment, the neural network operator includes a neural network classifier for gesture recognition and a neural network predictor for blood pressure calculation; the neural network operator is configured offline or online. When configured offline, it is configured on a PC, a smartphone or a tablet computer; when configured online, it is configured on the cloud.

[0078] The neural network classifier consists of a three-layer structure, as shown in Figure 8, including: a cross-sectional imaging input layer, a hidden layer for feature extraction, and a classification result output layer. Specifically, the processed wrist cross-sectional impedance change distribution image is input into the neural network, the hidden layer extracts imaging features, and the trained neural network can output gesture recognition results in the classification result output layer.

[0079] The neural network predictor consists of a three-layer structure, as shown in Figure 9, including: a pulse feature input layer, a hidden layer, and a prediction result output layer. Specifically, the selected pulse feature set and the calibrated blood pressure value at the same time are input into the neural network. After the hidden layer dimensionality reduction processing, the blood pressure prediction result can be output in the prediction result output layer.

[0080] In this embodiment, the neural network structure used is simple, consuming only a short computing time and fewer hardware resources, making it easy to migrate it to other electronic devices with computing capabilities that integrate ASIC chips, FPGA chips, or AI chips.

[0081] The above embodiments are merely illustrative of the basic principles, component structures, working processes, and effects of the present invention, and are not intended to limit the application of the present invention. Any person skilled in the art may modify or alter the above embodiments without violating the principles of the present invention. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the principles and technical ideas disclosed by the present invention shall still be considered as the scope of protection of the present invention and covered by the claims of the present invention.

Claims

1. A bracelet device integrating a gesture recognition function and a continuous blood pressure monitoring function, characterized in that, Comprising: A wearable bracelet, including at least two bracelet components, each bracelet component including: a component body, at least two contact sensing electrodes embedded in the inner wall of the component body, at least two contact sensing electrodes embedded in the same component body being arranged vertically, and each contact sensing electrode being configured in an excitation mode or a collection mode; A PCB board, used to switch between a gesture recognition function mode and a continuous blood pressure monitoring function mode. In the gesture recognition function mode, an excitation current is applied to an excitation current emitter composed of one selected electrode on each adjacent bracelet component and the response voltage signal fed back by a response voltage receiver composed of one selected electrode on each other adjacent bracelet component is collected. The response voltage signal collected in the gesture recognition function mode is modulated into wrist impedance distribution data. In the continuous blood pressure monitoring function mode, an excitation current is applied to an electrode pair on a bracelet component near the radial artery and the response voltage signal fed back by an electrode pair on other bracelet components is collected. The best response voltage signal collected in the continuous blood pressure monitoring function mode is modulated into the radial artery pulse impedance waveform data of each cardiac cycle, and the wrist impedance distribution data or the radial artery pulse impedance waveform data of each cardiac cycle is transmitted to a host computer, and the gesture recognition classification result or blood pressure prediction value fed back by the host computer is received; and, A display screen, used to visualize the user's heart rate and the radial artery pulse impedance waveform data of each cardiac cycle, and used to visualize the gesture recognition classification result or blood pressure prediction value received by the PCB board.

2. The bracelet device integrating the gesture recognition function and the continuous blood pressure monitoring function according to claim 1, characterized in that, The best response voltage signal collected in the continuous blood pressure monitoring function mode is obtained by the following method: extracting the complex amplitude of the response voltage signal fed back by the electrode pair on other bracelet components as the biological impedance characterization of the artery and its surrounding tissues, continuously collecting the complex amplitude of the response voltage signal fed back by the electrode pair on other bracelet components, and using the complex amplitude of the response voltage signal with continuous temporal variation as the characterization waveform of the pulsatile blood flow. Comparing the amplitude sizes of the pulsatile blood flow characterization waveforms obtained from the response voltage signals collected by the electrode pair on other bracelet components, and selecting the response voltage signal of the pulsatile blood flow characterization waveform with the largest amplitude and the largest peak-to-peak value of the pulse wave signal as the best response voltage signal.

3. The bracelet device integrating a gesture recognition function and a continuous blood pressure monitoring function according to claim 1, characterized in that, The at least two bracelet components are connected by drilling or buckling to form a wearable bracelet that surrounds the wrist once, and the component body is prepared from materials including but not limited to nylon and silica gel; the contact sensing electrode is one of a hemispherical electrode, a square electrode, a patch electrode, and a button electrode.

4. The bracelet device integrating a gesture recognition function and a continuous blood pressure monitoring function according to claim 1, wherein, The PCB board includes: An excitation source module, used to apply an excitation current to the contact sensing electrode in the excitation mode; A signal demodulation module, used to receive the response voltage fed back by the contact sensing electrode in the collection mode, modulate the response voltage signal collected in the gesture recognition function mode into wrist impedance distribution data, and modulate the best response voltage signal collected in the continuous blood pressure monitoring function mode into the radial artery pulse impedance waveform data of each cardiac cycle; The multiplexing module, in the gesture recognition function mode, cyclically connects any one of the excitation current emitters to the excitation source module and connects the other response voltage receivers to the signal demodulation module until all the excitation current emitters are applied with excitation current. In the continuous blood pressure monitoring function mode, it connects the electrode pair on one bracelet component near the radial artery to the excitation source module and connects the electrode pairs on other bracelet components to the signal demodulation module; The control module is used to control the gating of each channel in the multiplexing module, control the communication between the communication module and the display screen, and control the start and stop of the operation of the excitation source module and the signal demodulation module; The communication module is used to transmit the wrist impedance distribution data or the radial artery pulse impedance waveform data of each cardiac cycle to the host computer, receive the gesture recognition classification result or blood pressure prediction value fed back by the host computer, and transmit the gesture recognition classification result or blood pressure prediction value to the display screen; and, The power supply module is used to provide the working voltage and power consumption for the PCB board under full load operation of the components.

5. The bracelet device integrating a gesture recognition function and a continuous blood pressure monitoring function according to claim 4, wherein, The excitation source module includes: A waveform look-up table for generating a unipolar sinusoidal voltage signal; A digital-to-analog converter for converting the unipolar sinusoidal voltage signal into an analog signal and then outputting it; and, A voltage-controlled current source for converting the analog signal output by the digital-to-analog converter into a differential current signal and then outputting it.

6. The bracelet device integrating a gesture recognition function and a continuous blood pressure monitoring function according to claim 5, characterized in that, The signal demodulation module includes: A differential amplifier for differentially amplifying the received response voltage signal and then outputting it; An analog-to-digital converter for converting the differential signal output by the differential amplifier into a single-ended signal, converting the single-ended signal into a digital signal and then outputting it; and, A data demodulator for extracting the wrist impedance distribution data or the radial artery pulse impedance waveform data of each cardiac cycle from the digital signal output by the analog-to-digital converter.

7. The bracelet device integrating a gesture recognition function and a continuous blood pressure monitoring function according to claim 6, characterized in that, The multiplexing module is implemented by 4 multiplexer chips. The common ends of the 4 multiplexer chips are respectively connected to the two output terminals of the voltage-controlled current source and the two input terminals of the differential amplifier. Each contact sensing electrode is electrically connected to one selectable channel of the multiplexer chip. The address lines for controlling the channel gating of the 4 multiplexer chips are connected to the control module.

8. The bracelet device integrating a gesture recognition function and a continuous blood pressure monitoring function according to claim 1, wherein The display screen is fixed in a mechanical groove, and the mechanical groove is adhesively and seamlessly attached to the PCB board directly above by a laminating adhesive, and the wearable bracelet is electrically connected to the PCB board.

9. A system integrating a gesture recognition function and a continuous blood pressure monitoring function, characterized in that, It includes: The bracelet device according to any one of claims 1 to 8 and a PC terminal wirelessly communicating with the bracelet device. The PC terminal includes: A communication control module for controlling the start and stop of the communication function between the PC terminal and the bracelet device, receiving the wrist impedance distribution data or the radial artery pulse impedance waveform data of each cardiac cycle transmitted by the bracelet device, and transmitting the gesture recognition classification result or blood pressure prediction value calculated by the PC terminal back to the bracelet device; A pulse feature extraction module for converting the radial artery pulse impedance waveform data of each cardiac cycle into pulse feature data and constructing a pulse feature set; A blood pressure prediction module for training a pre-deployed neural network predictor according to the pulse feature set, and the trained neural network predictor calculates the blood pressure prediction value according to the real-time radial artery pulse impedance waveform data transmitted by the bracelet device; A wrist imaging operation module, configured to convert the wrist impedance distribution data transmitted by the bracelet device into an image of the impedance change distribution of the wrist cross-section; A gesture classification module, configured to train a pre-deployed neural network classifier according to the impedance change distribution image of the wrist cross-section. The trained neural network classifier predicts the gesture recognition classification result according to the real-time wrist impedance distribution data transmitted by the bracelet device; and A result display module, configured to visualize the wrist impedance distribution data or the radial artery pulse impedance waveform data transmitted by the bracelet device, and configured to visualize the gesture classification result predicted by the gesture classification module or the blood pressure prediction value calculated by the blood pressure prediction module.

10. The system integrating a gesture recognition function and a continuous blood pressure monitoring function according to claim 9, wherein, The pulse feature set constructed by the pulse feature extraction module includes: pulse feature data and the user's blood pressure value calibrated at the same acquisition time point as the radial artery pulse impedance waveform data. The pulse feature data includes, but is not limited to: the maximum slope, impedance amplitude, time interval, and the area enclosed by the impedance amplitude axis and the time axis of the four segments of waveforms from the starting point of the pulse blood flow representation wave to the main wave peak, from the main wave peak to the dicrotic wave valley, from the dicrotic wave valley to the dicrotic wave peak, and from the dicrotic wave peak to the end point.

Images