Multi-mode health monitoring bracelet and data processing method thereof

Abstract

The invention discloses a multi-mode health monitoring bracelet and a data processing method thereof. The multi-mode health monitoring bracelet is mainly composed of a dial plate body, a flexible substrate and a wrist strap assembly. An annular detection module is integrated in the dial plate body, 24 piezoresistive pulse condition sensors of an inner ring of the annular detection module are arranged according to cun-guan-chi positions to collect pulse conditions, and PPG sensors and biological impedance electrodes are alternately arranged on an outer ring of the annular detection module to be used for western medicine physical sign monitoring. A multi-layer flexible circuit is embedded in the flexible substrate, and the wrist strap assembly comprises a nickel-titanium alloy framework and a pneumatic micro-capsule. An edge calculation module is adopted for data processing, a dynamic baseline calibration algorithm is carried, and a threshold value is adjusted according to data and individual characteristics of a user in the first 72 hours. According to the system, traditional Chinese medicine pulse condition digitization and western medicine physical sign continuous monitoring can be synchronously achieved, the problems that a traditional device is single in function, delayed in data processing, fixed in health baseline and the like are solved, the system has the advantages of multi-mode monitoring, real-time response, personalized adaptation, comfortable wearing, accurate signals and the like, and comprehensive and accurate health monitoring and service are provided for a user.

Description

Technical Field

[0001] The present invention relates to the technical field of wearable medical devices, and specifically refers to a multi-modal health monitoring bracelet and its data processing method. Background Art

[0002] With the continuous improvement of people's health awareness, the demand for wearable health monitoring devices is increasing day by day. Traditional health monitoring bracelets have relatively single functions and often can only focus on monitoring some physical signs in Western medicine, such as heart rate and steps, and cannot achieve synchronous and continuous monitoring of traditional Chinese medicine pulse digitization and Western medicine physical signs. This makes it impossible for users to comprehensively and comprehensively understand their own health conditions and cannot meet the needs of modern people for diversified health monitoring.

[0003] In the prior art, the data processing of devices usually relies on the cloud. Although the cloud has powerful computing capabilities, there will inevitably be delays in the data transmission process, which poses a greater risk for some health monitoring scenarios that require real-time response, such as early warning of sudden diseases. For example, when a patient has acute heart rate abnormalities, due to data processing delays, it may lead to the failure to detect and take corresponding measures in a timely manner, delaying the best treatment opportunity.

[0004] In addition, when setting the health baseline for existing health monitoring devices, most of them use fixed thresholds. However, the physiological rhythms of the human body are dynamically changing, and there are differences in the normal ranges of physiological indicators for different individuals at different time periods. For example, the normal heart rate ranges of young people and the elderly are different, and there are also obvious changes in various physiological indicators of the human body before and after exercise, during sleep and wakefulness. Fixed health baseline thresholds cannot adapt to the dynamic changes of individual physiological rhythms, easily cause misjudgments, resulting in inaccurate monitoring results and unable to provide users with accurate health assessments and early warnings. Summary of the Invention

[0005] The present invention aims to solve the above technical problems and provides a multi-modal health monitoring bracelet and its data processing method.

[0006] To solve the above technical problems, the technical solution provided by the present invention is: A multi-modal health monitoring bracelet, comprising:

[0007] A watch face body, which integrates a ring-shaped detection module inside. The ring-shaped detection module includes: an inner ring array, which is composed of 24 piezoresistive pulse sensors arranged at equal intervals according to the anatomical positions of cun, guan, and chi, with the diameter of a single sensor being 2 ± 0.1 mm and the spacing being 1.5 ± 0.2 mm; an outer ring array, which is composed of 8 PPG sensors and 4 pairs of bio-impedance electrodes arranged alternately;

[0008] A flexible substrate, which is integrally formed by medical-grade liquid silicone, with a thickness of 1.2 ± 0.05 mm and embedded with multiple layers of flexible circuits;

[0009] The wristband assembly includes a nickel-titanium alloy memory skeleton and a plurality of pneumatic microcapsules distributed on the inner side. The diameter of the pneumatic microcapsules is 3 ± 0.2 mm, and the circumferential spacing is 5 mm.

[0010] Furthermore, the piezoresistive pulse sensor includes: a multi-layer composite structure, which successively includes a polyimide substrate, a nano-silver conductive layer, a PVDF piezoelectric thin film, and a hemispherical contact protrusion from bottom to top.

[0011] Furthermore, the PPG sensor includes:

[0012] A three-wavelength LED array, which includes three light-emitting units of 660 nm, 805 nm, and 940 nm;

[0013] A photodiode, made of InGaAs material, with a photosensitive area of 2 × 2 mm;

[0014] An optical isolation ring, made of black silicone, with a height of 0.8 mm and an isolation angle ≥ 60°.

[0015] Furthermore, the bio-impedance electrode is: a four-electrode structure, which includes two pairs of current injection electrodes and voltage detection electrodes.

[0016] Furthermore, the multi-layer flexible circuit includes a signal acquisition layer, which adopts a serpentine wiring structure, with a line width of 50 ± 5 μm, a line pitch of 80 ± 10 μm, and a stretching rate ≥ 200%;

[0017] An electromagnetic shielding layer, composed of a nano-silver coating, with a thickness of 50 - 80 nm;

[0018] A power supply layer, integrated with a flexible lithium battery unit, with a thickness ≤ 0.3 mm.

[0019] Furthermore, it also includes a piezoelectric drive module, which specifically includes: a micro air pump, a pressure sensor, and an airbag control unit.

[0020] Furthermore, it also includes a data fusion module and a wireless transmission module, which are integrated in the watch body and include:

[0021] A signal preprocessing unit, which performs wavelet noise reduction processing on the pulse signal;

[0022] A feature extraction unit, which synchronously extracts the time-frequency features of the pulse waveform and the impedance spectrum features;

[0023] A fusion analysis unit, which uses a deep neural network to realize the classification of traditional Chinese medicine pulses and the joint calibration of physiological parameters.

[0024] A data processing method for a multi-modal health monitoring bracelet further includes an edge computing module equipped with a dynamic baseline calibration algorithm and a multi-modal fusion decision-making model. The dynamic baseline calibration algorithm satisfies: Threshold = Baseline value × (1 + 0.05 × Age coefficient + 0.03 × BMI), where the baseline value is calculated based on the user's data in the previous 72 hours.

[0025] The specific steps are as follows:

[0026] S1: Synchronously collect traditional Chinese medicine pulse condition and Western medicine physical sign data;

[0027] S2: Perform wavelet denoising and feature extraction at the edge;

[0028] S3: Trigger a three-level early warning response based on the dynamic baseline.

[0029] The three-level early warning response includes:

[0030] Level 1 early warning: When the detected value exceeds 10% of the baseline value, issue a mild warning;

[0031] Level 2 early warning: When the detected value exceeds 20% of the baseline value, issue a moderate warning;

[0032] Level 3 early warning: When the detected value exceeds 30% of the baseline value, issue a severe warning and recommend seeking medical treatment.

[0033] The advantages of the present invention compared with the prior art are as follows:

[0034] 1. Multi-modal monitoring

[0035] The bracelet of the present invention can synchronously digitize traditional Chinese medicine pulse conditions and continuously monitor Western medicine physical signs, combining the traditional pulse diagnosis of traditional Chinese medicine with the physical sign monitoring of modern Western medicine, providing users with more comprehensive and comprehensive health information. Users can comprehensively understand their physical conditions through the bracelet. Whether it is the qi and blood and zang-fu organ functions concerned by traditional Chinese medicine or the indicators such as heart rate and body fat of Western medicine, they can be clearly seen at a glance.

[0036] 2. Real-time response

[0037] By adopting an edge computing module, data processing does not need to rely on the cloud, greatly reducing the delay of data transmission and processing. When abnormal user health indicators are detected, it can timely trigger an early warning response, winning precious treatment time for users. For example, when a user suddenly suffers from a heart disease resulting in abnormal heart rate, the bracelet can issue a warning within a very short time, reminding the user and surrounding people to take corresponding measures and effectively reducing risks.

[0038] 3. Personalized adaptation

[0039] The dynamic baseline calibration algorithm can dynamically adjust the health baseline threshold according to individual characteristics such as the user's age and BMI, as well as physiological data in the previous 72 hours, adapting to the changes in individual physiological rhythms. Users of different ages and physical conditions can obtain accurate health monitoring and evaluation, avoiding misjudgments caused by fixed thresholds and improving the accuracy and reliability of monitoring results.

[0040] 4. Comfortable to wear

[0041] The flexible substrate is integrally formed by medical-grade liquid silicone, which is soft and comfortable and non-irritating to the skin. The nickel-titanium alloy memory skeleton of the wristband component can adapt to different wrist thicknesses, and the pneumatic microcapsules can adjust the fit, ensuring that the bracelet is both stable and comfortable during wearing, without bringing additional burden to the user. Users can wear it for a long time to ensure the continuity of health monitoring.

[0042] 5. Accurate signal

[0043] Structures such as the signal acquisition layer design of the multi-layer flexible circuit, the electromagnetic shielding layer, and the optical isolation ring in the detection module effectively reduce external interference and ensure the accuracy of signal acquisition. Whether it is the pulse signal or the western medicine physical sign signal, they can be accurately collected and transmitted, providing a reliable data basis for subsequent data processing and analysis, making the health assessment results more credible. Brief description of the drawings

[0044] Figure 1 is the structural schematic diagram of the multi-modal health monitoring bracelet of the present invention Figure 1 .

[0045] Figure 2 is the structural schematic diagram of the multi-modal health monitoring bracelet of the present invention Figure 2 .

[0046] Figure 3 is the structural schematic diagram of the annular detection module of the multi-modal health monitoring bracelet of the present invention.

[0047] Figure 4 is the structural schematic diagram of the piezoresistive pulse sensor group of the multi-modal health monitoring bracelet of the present invention.

[0048] As shown in the figure: 100, dial body; 101, annular detection module; 1011, piezoresistive pulse sensor; 1011a, polyimide substrate; 1011b, silver nanowire conductive layer; 1011c, PVDF piezoelectric film; 1011d, hemispherical contact protrusion; 1012, PPG sensor; 1013, bioimpedance electrode; 102, flexible substrate; 1021, multi-layer flexible circuit; 200, wristband component; 201, pneumatic microcapsule. Detailed implementation manners

[0049] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0050] I. Working principle of the present invention:

[0051] The multi-modal health monitoring bracelet of the present invention mainly includes a watch dial body 100, a flexible substrate 102 and a wristband assembly 200.

[0052] Ring detection module

[0053] Inner ring array: It is composed of 24 piezoresistive pulse sensors 1011 arranged at equal intervals according to the anatomical positions of cun, guan and chi on the human wrist. The diameter of a single pulse sensor is 2±0.1 mm, and the spacing is 1.5±0.2 mm. This design can accurately collect the pulse information of the cun, guan and chi parts on the human wrist, simulating the traditional Chinese medicine pulse-taking method. The piezoresistive pulse sensor adopts a multi-layer composite structure, which is successively composed of a polyimide substrate 1011a, a nano-silver conductive layer 1011b, a PVDF piezoelectric film 1011c, and a hemispherical contact protrusion 1011d from bottom to top. When the pressure of the human pulse acts on the hemispherical contact protrusion, the PVDF piezoelectric film will deform, which will cause a change in its resistance. The resistance change signal is transmitted to the subsequent processing circuit through the nano-silver conductive layer to realize the acquisition of the pulse signal.

[0054] Outer ring array: It is composed of 8 PPG sensors 1012 and 4 pairs of bio-impedance electrodes 1013 arranged alternately. The PPG sensor is used to detect the pulse wave signal of the human body. It includes a three-wavelength LED array, which contains three light-emitting units of 660 nm, 805 nm, and 940 nm. Lights of different wavelengths can penetrate different depths of human tissues to obtain more comprehensive pulse wave information. The photodiode is made of InGaAs material, with a photosensitive area of 2×2 mm, and can efficiently receive the reflected light signal. The optical isolation ring is made of black silicone, with a height of 0.8 mm and an isolation angle ≥60°, which can effectively reduce the interference of external light and ensure the accuracy of the detection signal. The bio-impedance electrode 1013 is a four-electrode structure, which includes two pairs of current injection electrodes and voltage detection electrodes. By injecting a weak current into the human body, the change in the impedance of human tissues is measured, so as to obtain the physiological information of the human body, such as body fat percentage, water content, etc.

[0055] Data fusion module and wireless transmission module: Integrated within the watch body. The signal preprocessing unit performs wavelet denoising on the pulse signal to remove noise interference during acquisition and improve signal quality. The feature extraction unit synchronously extracts the time-frequency features and impedance spectrum features of the pulse waveform, providing a data basis for subsequent analysis. The fusion analysis unit uses a deep neural network to achieve the joint calibration of traditional Chinese medicine pulse classification and physiological parameters, fusing traditional Chinese medicine pulse information with Western medicine physical sign data for a more comprehensive and accurate health assessment result. The wireless transmission module is responsible for transmitting the processed data to external devices such as mobile phones and computers for easy viewing and further analysis by the user.

[0056] Piezoelectric drive module: Specifically includes a micro air pump, a pressure sensor, and an airbag control unit. Its function is to adjust the fit between the bracelet and the wrist by controlling the inflation and deflation of the pneumatic microcapsule 201 inside the wristband assembly, ensuring that the detection module can stably and accurately collect signals. When the pressure sensor detects that the fit between the bracelet and the wrist is insufficient, the micro air pump operates to inflate the pneumatic microcapsule, making the bracelet fit more closely to the wrist; conversely, when the fit is too high, the airbag control unit controls the pneumatic microcapsule to deflate to adjust the fit.

[0057] The flexible substrate 102 is integrally formed of medical-grade liquid silicone with a thickness of 1.2 ± 0.05 mm and embedded with multiple flexible circuits 1021. The multiple flexible circuits include a signal acquisition layer with a serpentine wiring structure, a line width of 50 ± 5 μm, a line pitch of 80 ± 10 μm, and a stretch rate ≥ 200%. This structural design can ensure signal transmission stability while adapting to the bending and stretching of the bracelet during wear, ensuring that the detection module always maintains good contact with the human skin. The electromagnetic shielding layer is composed of a nano-silver coating with a thickness of 50 - 80 nm, which can effectively shield external electromagnetic interference and improve the accuracy of signal acquisition. The power supply layer integrates a flexible lithium battery unit with a thickness ≤ 0.3 mm, providing stable power support for the entire bracelet, and the flexible design can better adapt to the shape of the bracelet.

[0058] The wristband assembly 200 includes a nickel-titanium alloy memory skeleton and multiple pneumatic microcapsules 201 distributed on the inner side, with a pneumatic microcapsule diameter of 3 ± 0.2 mm and a circumferential pitch of 5 mm. The nickel-titanium alloy memory skeleton has good elasticity and shape memory characteristics, which can adapt to the thickness of different users' wrists and provide a comfortable wearing experience. The pneumatic microcapsules are inflated and deflated under the control of the piezoelectric drive module to adjust the fit between the bracelet and the wrist.

[0059] Data processing method

[0060] Edge computing module: Equipped with a dynamic baseline calibration algorithm and a multi-modal fusion decision-making model. The dynamic baseline calibration algorithm satisfies: Threshold = Baseline value × (1 + 0.05 × Age coefficient + 0.03 × BMI), where the baseline value is calculated based on the user's data in the previous 72 hours. In this way, the health baseline threshold can be dynamically adjusted according to the user's individual characteristics and recent physiological data, improving the accuracy and adaptability of monitoring.

[0061] Data processing steps

[0062] S1: Synchronously collect traditional Chinese medicine pulse and Western medicine physical sign data: Through the annular detection module of the watch body, the piezoresistive pulse sensors in the inner ring array collect traditional Chinese medicine pulse data, and the PPG sensors and bio-impedance electrodes in the outer ring array collect Western medicine physical sign data, realizing the synchronous collection of multi-modal data.

[0063] S2: Perform wavelet denoising and feature extraction at the edge: The collected data is first subjected to wavelet denoising processing by the signal preprocessing unit in the edge computing module to remove noise interference, and then the time-frequency features of the pulse waveform and impedance spectrum features are synchronously extracted by the feature extraction unit.

[0064] S3: Trigger a three-level warning response based on the dynamic baseline: When the detected value exceeds different proportions of the dynamic baseline value, different levels of warning responses are triggered. Level 1 warning: When the detected value exceeds 10% of the baseline value, a mild warning is issued; Level 2 warning: When the detected value exceeds 20% of the baseline value, a moderate warning is issued; Level 3 warning: When the detected value exceeds 30% of the baseline value, a severe warning is issued and medical treatment is recommended. This hierarchical warning mechanism can timely remind users to pay attention to their own health conditions and take corresponding measures according to different degrees of abnormalities.

[0065] II. Implementation method:

[0066] 2.1 Manufacturing

[0067] First, manufacture the annular detection module. For the piezoresistive pulse sensors in the inner ring array, a nano-silver conductive layer and a PVDF piezoelectric film are successively fabricated on a polyimide substrate through processes such as photolithography and coating, and hemispherical contact protrusions are processed on the PVDF piezoelectric film. Then, 24 fabricated piezoresistive pulse sensors are arranged and fixed at equal intervals according to the cun-guan-chi anatomical positions. For the outer ring array, 8 PPG sensors and 4 pairs of bio-impedance electrodes are alternately arranged and fixed according to the design requirements. Electronic components such as the data fusion module, wireless transmission module, and piezoelectric drive module are integrated into the corresponding positions inside the watch body and circuit connections are made.

[0068] Through the die-casting process of injection molding, medical-grade liquid silicone is integrally formed into a flexible substrate with a thickness of 1.2 ± 0.05 mm. During the forming process, the pre-made multi-layer flexible circuits are embedded therein to ensure the accurate positions of each layer of circuits. The multi-layer flexible circuits are tested to ensure the normal functions of the signal acquisition layer, electromagnetic shielding layer, and power supply layer.

[0069] A nickel-titanium alloy memory skeleton is fabricated and through processes such as heat treatment, it is made to have good elasticity and shape memory characteristics. A plurality of pneumatic microcapsules are installed inside the nickel-titanium alloy memory skeleton to ensure that the diameters and circumferential spacings of the pneumatic microcapsules meet the design requirements and are connected to the piezoelectric drive module.

[0070] The watch dial body is installed at one end of the flexible substrate to ensure a firm connection between the watch dial body and the flexible substrate and normal circuit connectivity. The wristband assembly is connected to the other end of the flexible substrate to form a complete bracelet structure. The bracelet is subjected to an overall test, including detecting the signal acquisition function, data processing function, wireless transmission function of the detection module, and the function of the piezoelectric drive module to adjust the fit of the bracelet, etc., to ensure that all performance indicators of the bracelet meet the design requirements.

[0071] 2.2 Usage method

[0072] When the user wears the bracelet, according to the thickness of their own wrist, by adjusting the nickel-titanium alloy memory skeleton of the wristband assembly, the bracelet is initially adapted to the wrist.

[0073] After the bracelet is turned on, the piezoelectric drive module is automatically activated, and the pressure sensor detects the fit between the bracelet and the wrist. The micro air pump inflates or deflates the pneumatic microcapsules according to the feedback of the pressure sensor to adjust the fit between the bracelet and the wrist to ensure that the detection module can stably collect signals.

[0074] The annular detection module starts to synchronously collect traditional Chinese medicine pulse conditions and western medicine physical sign data. The piezoresistive pulse sensor collects the pulse signals at the cun, guan, and chi positions, the PPG sensor collects the pulse wave signals, and the bio-impedance electrodes collect the human body impedance signals.

[0075] The collected data is transmitted to the edge computing module, where wavelet noise reduction processing is performed at the edge to remove noise interference, and then the feature extraction unit synchronously extracts the time-frequency features of the pulse waveform and the impedance spectrum features.

[0076] The fusion analysis unit uses a deep neural network to analyze the extracted features, realizes the classification of traditional Chinese medicine pulse conditions and the joint calibration of physiological parameters, and judges whether the detected value is abnormal according to the threshold calculated by the dynamic baseline calibration algorithm.

[0077] If the detected value is abnormal, corresponding-level warning responses are triggered according to the proportion exceeding the baseline value. During a level-1 warning, the bracelet reminds the user by vibrating or emitting a slight prompt sound; during a level-2 warning, the prompt sound is enhanced, and at the same time, a moderate warning message is displayed on the bracelet screen; during a level-3 warning, the bracelet emits a strong prompt sound, and the warning information is sent to the mobile phone of the emergency contact preset by the user through the wireless transmission module. At the same time, it is recommended that the user seek medical attention in a timely manner.

[0078] Users can view detailed health monitoring data and analysis reports, including historical data, trend analysis, etc., through the mobile phone APP or computer software connected to the bracelet, so as to better understand their own health status.

[0079] The above describes the present invention and its implementation manners. Such a description is not restrictive. What is shown in the drawings is only one of the implementation manners of the present invention, and the actual structure is not limited thereto. All in all, if those of ordinary skill in the art are inspired by it and, without departing from the purpose of the present invention, design similar structural manners and embodiments to this technical solution without creative efforts, they shall fall within the protection scope of the present invention.

Claims

1. Multimodal health monitoring bracelet, characterized in that: Including: A watch dial body (100) with an annular detection module (101) integrated inside. The annular detection module (101) includes: an inner ring array composed of 24 piezoresistive pulse sensors (1011) arranged at equal intervals according to the anatomical positions of cun, guan, and chi. The diameter of a single pulse sensor is 2 ± 0.1 mm, and the spacing is 1.5 ± 0.2 mm; an outer ring array composed of 8 PPG sensors (1012) and 4 pairs of bioimpedance electrodes (1013) arranged alternately. A flexible substrate (102) formed by one-piece molding of medical-grade liquid silicone, with a thickness of 1.2 ± 0.05 mm and multiple layers of flexible circuits (1021) embedded. A wristband assembly (200) including a nickel-titanium alloy memory skeleton and multiple pneumatic microcapsules (201) distributed on the inner side. The diameter of the pneumatic microcapsules (201) is 3 ± 0.2 mm, and the circumferential spacing is 5 mm.

2. The multimodal health monitoring bracelet according to claim 1, wherein: The piezoresistive pulse sensor (1011) is a multi-layer composite structure, which is successively composed of a polyimide substrate (1011a), a nano-silver conductive layer (1011b), a PVDF piezoelectric film (1011c), and a hemispherical contact protrusion (1011d) from bottom to top.

3. The multimodal health monitoring bracelet according to claim 1, wherein: The PPG sensor (1012) includes: A three-wavelength LED array including three light-emitting units of 660 nm, 805 nm, and 940 nm; A photodiode made of InGaAs material with a photosensitive area of 2 × 2 mm; An optical isolation ring made of black silicone, with a height of 0.8 mm and an isolation angle ≥ 60°.

4. The multi-modal health monitoring bracelet according to claim 1, characterized in that: The bioimpedance electrode (1013) is a four-electrode structure, including two pairs of current injection electrodes and voltage detection electrodes.

5. The multi-modal health monitoring bracelet according to claim 1, wherein: The multi-layer flexible circuit (1021) includes a signal acquisition layer with a serpentine trace structure, a line width of 50 ± 5 μm, a line pitch of 80 ± 10 μm, and a stretching rate ≥ 200%; An electromagnetic shielding layer composed of a nano-silver coating, with a thickness of 50 - 80 nm; A power supply layer integrating a flexible lithium battery unit, with a thickness ≤ 0.3 mm.

6. The multimodal health monitoring bracelet according to claim 1, wherein: It also includes a piezoelectric drive module, specifically including: a micro air pump, a pressure sensor, and an airbag control unit.

7. The multimodal health monitoring bracelet according to claim 1, wherein: It also includes a data fusion module and a wireless transmission module, which are integrated in the watch dial body (100) and include: A signal preprocessing unit for performing wavelet noise reduction on the pulse signal; A feature extraction unit for synchronously extracting the time-frequency features of the pulse waveform and the impedance spectrum features; A fusion analysis unit for realizing the classification of traditional Chinese medicine pulses and the joint calibration of physiological parameters by using a deep neural network.

8. The data processing method of a multi-modal health monitoring bracelet according to any one of claims 1-7, characterized in that: It also includes an edge computing module equipped with a dynamic baseline calibration algorithm and a multi-modal fusion decision model. The dynamic baseline calibration algorithm satisfies: threshold = baseline value × (1 + 0.05 × age coefficient + 0.03 × BMI), where the baseline value is calculated based on the user's data in the previous 72 hours.

9. The data processing method of a multimodal health monitoring bracelet according to claim 8, characterized in that: The specific steps are as follows: S1: Synchronously collect traditional Chinese medicine pulse and western medicine physical sign data; S2: Perform wavelet noise reduction and feature extraction at the edge; S3: Trigger a three-level early warning response based on the dynamic baseline.

10. The data processing method of a multi-modal health monitoring bracelet according to claim 9, wherein: The three-level early warning response includes: Level 1 early warning: When the detected value exceeds 10% of the baseline value, a mild warning is issued; Level 2 early warning: When the detected value exceeds 20% of the baseline value, a moderate warning is issued; Level 3 warning: When the detected value exceeds 30% of the baseline value, a severe warning is issued and medical treatment is recommended.

Images