Self-adaptive ear health monitoring equipment based on multi-modal sensor

Abstract

The invention discloses a self-adaptive ear health monitoring device based on a multi-modal sensor, and relates to the technical field of data processing, and the device comprises a data obtaining module which is used for referring to a monitoring period and obtaining x-axis, y-axis and z-axis average acceleration and average massage intensity; the first correction module is used for acquiring the exercise intensity of the to-be-processed period based on the exercise model and acquiring a first correction parameter according to the exercise intensity of the to-be-processed period; the photoelectric volume pulse wave monitoring module is used for acquiring standard monitoring parameters, acquiring current monitoring parameters and acquiring multiple groups of current blood pressure data, current blood oxygen data and current heart rate data; the second correction module is used for acquiring a second correction parameter based on the parameter model; and the ear massage module is used for acquiring the current massage intensity according to the second correction parameter and the average massage intensity, and performing ear massage according to the current massage intensity. The device has the advantages of double progressive correction, dynamic self-adaption and accurate massage control.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically to an adaptive ear health monitoring device based on a multimodal sensor. Background Technology

[0002] With the accelerating aging of the global population and the continuous rise in the prevalence of chronic diseases, real-time and accurate health monitoring technology has become a core requirement of modern healthcare systems. The ear, as an important window into the human body's bio-information, with its rich capillary network, vagus nerve branches, and the mapping relationship between auricular acupoints in Traditional Chinese Medicine and the internal organs of the body (e.g., the central concha corresponds to the heart, and the area around the crus of the helix corresponds to the digestive system), has become an ideal medium for non-invasive health monitoring.

[0003] While existing ear monitoring devices can acquire basic physiological parameters, they face numerous technical bottlenecks in practical applications. Specifically, interference from motion noise on physiological signal acquisition is difficult to suppress effectively. When users are walking, running, or engaged in daily activities, it affects photoplethysmography (PPG) signals, leading to inaccuracies in key parameters such as blood pressure, blood oxygen, and heart rate. Furthermore, existing devices cannot adaptively distinguish between static, low-intensity activity, and high-intensity exercise scenarios, resulting in a lack of adaptability between static monitoring and intervention modes. Moreover, existing ear monitoring devices generally use preset fixed light source intensity, sampling frequency, and mechanical massage parameters. This means they cannot dynamically adjust monitoring strategies based on the user's real-time physiological state, nor can they optimize biostimulation intensity based on changes in exercise load. During the post-exercise heart rate rise phase, fixed strong light exposure can cause PPG signal saturation due to vasoconstriction, while rigid massage intensity may cause user discomfort or even increase cardiovascular burden. Furthermore, this model severely limits the theoretical advantages of TCM auricular therapy in diagnosis and treatment, and cannot achieve the therapeutic goal of dynamically regulating the function of internal organs through differentiated stimulation of specific ear areas. Moreover, existing ear massage devices are prone to displacement when the user moves, leading to problems such as optical path deviation and uneven contact pressure. The stability of the device fit affects the accuracy of various data acquisitions, which also creates a vicious cycle for improving the quality of massage. Summary of the Invention

[0004] To address the shortcomings of existing ear monitoring devices in multimodal data collaborative processing under dynamic environments, which prevents them from achieving adaptive optimization of the "sensing-analysis-intervention" closed loop, and especially the difficulty in resolving the dual interference of motion state and individual physiological differences on monitoring accuracy and treatment efficacy, this invention provides an adaptive ear health monitoring device based on multimodal sensors.

[0005] An adaptive ear health monitoring device based on multimodal sensors includes: a data acquisition module for acquiring the previous monitoring cycle as a reference monitoring cycle, and acquiring the average acceleration and average massage intensity along the x, y, and z axes within the reference monitoring cycle; a first correction module for acquiring the motion intensity of the current monitoring cycle based on a motion model and the average acceleration along the x, y, and z axes within the reference monitoring cycle, and acquiring a first correction parameter based on the motion intensity of the current monitoring cycle; a photoplethysmography (PPG) monitoring module for acquiring standard monitoring parameters, acquiring current monitoring parameters based on the first correction parameter and the standard monitoring parameters, and acquiring multiple sets of current blood pressure data, current blood oxygen data, and current heart rate data according to the current monitoring parameters within the current monitoring cycle; a second correction module for acquiring a second correction parameter based on a parameter model, the motion intensity of the current monitoring cycle, multiple sets of current blood pressure data, current blood oxygen data, and current heart rate data; and an ear massage module for acquiring the current massage intensity based on the second correction parameter and the average massage intensity, and performing ear massage according to the current massage intensity.

[0006] Optionally, the first correction module is further configured to: set multiple different and continuous numerical ranges, each numerical range corresponding to a different first correction parameter; obtain the numerical range into which the motion intensity of the cycle to be processed falls and obtain the first correction parameter corresponding to that numerical range.

[0007] Optionally, the photoplethysmography (PPG) monitoring module is also used to: obtain the standard light source intensity and standard sampling frequency according to the standard monitoring parameters; configure the current light source intensity according to the first correction parameter and the standard light source intensity, and configure the current sampling frequency according to the first correction parameter and the standard sampling frequency; and form the current monitoring parameters according to the current light source intensity and the current sampling frequency.

[0008] Optionally, the motion model in the motion intensity of the period to be processed, obtained based on the motion model and the average acceleration of the x, y, and z axes within the reference monitoring period, is represented as follows: ;in, The motion intensity of the cycle to be processed. For reference, the data sampling points within the monitoring period, For the x-axis acceleration of the t-th data sampling point within the reference monitoring period, For the reference monitoring period, the y-axis acceleration of the t-th data sampling point, The z-axis acceleration is the data sampling point t within the reference monitoring period.

[0009] Optionally, the second correction module is further configured to: obtain the exercise influence ratio based on the exercise intensity and intensity threshold of the period to be processed, obtain the blood pressure influence ratio based on multiple sets of current blood pressure data and upper limit blood pressure threshold, obtain the blood oxygen influence ratio based on multiple sets of current blood oxygen data and blood oxygen threshold, and obtain the heart rate influence ratio based on multiple sets of current heart rate data and upper limit heart rate threshold; and obtain the second correction parameter based on the parameter model, exercise influence ratio, blood pressure influence ratio, blood oxygen influence ratio, and heart rate influence ratio.

[0010] Optionally, the parameter model in the second correction parameter obtained based on the parameter model, the exercise intensity of the period to be processed, multiple sets of current blood pressure data, current blood oxygen data, and current heart rate data is represented as follows: ;in, This is the second correction parameter. This is the scaling factor. The motion intensity of the cycle to be processed. The intensity threshold, This represents the number of sampling time points within the current monitoring period. This refers to the blood pressure data at the i-th sampling time point within the current monitoring period. This is the upper limit threshold for blood pressure. This refers to the blood oxygen data at the i-th sampling time point within the current monitoring period. The lowest threshold for blood oxygenation. This refers to the heart rate data at the i-th sampling time point within the current monitoring period. This is the upper limit threshold for heart rate.

[0011] Optionally, it also includes an inflatable adhesive layer disposed on the surface of the ear massage module and in contact with the skin, and a tightness adjustment module, the tightness adjustment module being used to: obtain the average acceleration of the x, y, and z axes at the current moment, and obtain the motion intensity at the current moment based on the average acceleration of the x, y, and z axes at the current moment; obtain a motion trend value based on the motion intensity of the period to be processed and the motion intensity at the current moment; and adjust the internal pressure of the inflatable adhesive layer based on the magnitude of the motion trend value.

[0012] Optionally, the pressure of the inflatable bonding layer can be adjusted according to the magnitude of the movement trend, as expressed as: , ;in, For the volume increase of the inflatable bonding layer, For a single volume standard increment, This is a trend in sports.

[0013] Optionally, the ear massage module includes a control layer, in which multiple distributed vibration units are provided, the vibration intensity of the multiple vibration units being equal to the current massage intensity.

[0014] Optionally, it also includes a communication module connected to an interactive terminal, which is used to set the basic massage intensity and view historical monitoring data.

[0015] The beneficial effects of this invention are reflected in: In the entire adaptive ear health monitoring device based on multimodal sensors, firstly, the first correction module predicts motion intensity based on acceleration data, intelligently divides activity intensity ranges, and generates hardware parameter adjustment coefficients, revolutionizing the anti-interference capability of physiological signal acquisition. By predicting the user's motion state (such as transitioning from sitting to running), it dynamically increases the light source intensity to penetrate constricted blood vessels and avoid PPG saturation, while simultaneously increasing the sampling frequency to capture redundant data and offset motion artifacts, significantly reducing the measurement error of key parameters such as blood pressure and blood oxygen in dynamic scenarios such as walking and running. Furthermore, the second correction module, relying on a comprehensive assessment of real-time physiological data and exercise load, has made a breakthrough in constructing a "physiological risk-ear acupoint stimulation" regulation model. By quantifying and generating massage intensity adjustment parameters through multi-dimensional threshold analysis (such as a sudden increase in blood pressure exceeding the safety threshold or a heart rate reaching the cardiovascular load limit), when a user's heart rate and blood pressure are both high after hiking, the massage intensity of the concha cavity area (corresponding to the heart acupoint) is reduced proportionally to avoid overstimulation of the vagus nerve. Meanwhile, when sitting for a long time, the massage intensity of the helix crus (corresponding to the digestive system) is specifically guaranteed, realizing dynamic adaptation of acupoint stimulation parameters based on real-time physiological data. In summary, the two-stage correction forms a monitoring and intervention closed loop. The first correction ensures high-quality data sources, and the second correction dynamically optimizes biostimulation based on these sources. This not only avoids the cardiovascular risks caused by strong massage after exercise, but also enables acupoint zonal treatment through distributed vibration units. Attached Figure Description

[0016] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0017] Figure 1 This is a schematic diagram of the composition of the adaptive ear health monitoring device based on multimodal sensors of the present invention; Figure 2 This is a schematic diagram of the dynamic monitoring section in the adaptive ear health monitoring device based on multimodal sensors of the present invention; Figure 3 This is a schematic diagram of the stability control section in the adaptive ear health monitoring device based on multimodal sensors of the present invention; Figure 4 This is a schematic diagram of the human-computer interaction component in the adaptive ear health monitoring device based on multimodal sensors of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0019] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0020] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0021] like Figure 1 and Figure 2 As shown, an adaptive ear health monitoring device based on a multimodal sensor is provided, comprising: The data acquisition module is used to acquire the previous monitoring cycle of the current monitoring cycle as a reference monitoring cycle, and to acquire the average acceleration and average massage intensity of the x, y, and z axes within the reference monitoring cycle. The first correction module is used to obtain the motion intensity of the period to be processed based on the motion model and the average acceleration of the x, y, and z axes within the reference monitoring period, and to obtain the first correction parameter based on the motion intensity of the period to be processed. The photoplethysmography (PPG) monitoring module is used to acquire standard monitoring parameters, and to acquire current monitoring parameters based on the first correction parameter and the standard monitoring parameters. Within the current monitoring cycle, it acquires multiple sets of current blood pressure data, current blood oxygen data, and current heart rate data according to the current monitoring parameters. The second correction module is used to obtain the second correction parameters based on the parameter model, the exercise intensity of the period to be processed, multiple sets of current blood pressure data, current blood oxygen data and current heart rate data; The ear massage module is used to obtain the current massage intensity based on the second correction parameter and the average massage intensity, and to perform ear massage according to the current massage intensity.

[0022] In this embodiment, it should be noted that the data acquisition module is the foundation of the entire device's operation, primarily responsible for establishing the reference baseline for the current operation. Its core task is to accurately capture and process information from the previous complete monitoring cycle (defined as the "reference monitoring cycle") at the beginning of each new "current monitoring cycle," which is generally 3 to 5 seconds. Specifically, the module calculates and extracts the dynamic average values ​​of two key parameters from the raw or multi-frequency sampled data of this reference monitoring cycle: the average motion acceleration along the three axes (x, y, z directions) and the average massage intensity applied to the ear. For example, the module summarizes all acceleration readings from the previous minute to calculate the overall average motion activity level for that cycle, and simultaneously combines the changes in massage intensity within the same time period to derive a representative average massage intensity value. These average values ​​are not simple static values, but rather processed and compressed core characteristic values ​​representing the overall state of the past time period.

[0023] Furthermore, the triaxial average acceleration quantitatively characterizes the user's overall physical activity level (e.g., sitting, slow walking, jogging) during the recently concluded reference monitoring period, serving as the primary input for assessing the degree of external motion interference. The extracted average massage intensity value represents the level of stimulation applied to the user's ear by the device during the reference monitoring period, serving as a historical baseline for massage intervention intensity. For example, if the user was resting during the reference period, the average acceleration would be low; while if the user was undergoing rehabilitation training, the average massage intensity might remain at a moderate level. These two sets of averages provide indispensable, real-time historical contextual information for the subsequent first correction module. The first correction module relies on this past period's motion activity level to predict the intensity of potential current motion interference (e.g., assuming the user was jogging one minute, they are likely still moving in the next period), and also relies on the past average massage intensity as a baseline for adjusting future massage intensity (e.g., if the massage intensity was moderate and the user experienced no discomfort the previous minute, this could be used as a safe starting point for fine-tuning). Therefore, the data acquisition module essentially provides the ability to perceive historical states, ensuring that subsequent dynamic adjustments to parameters (such as light source, sampling frequency, and new massage intensity) are based on a recently occurred, continuous context, rather than fixed values ​​that are set out of thin air or are unrelated.

[0024] In the first correction module, an adjustment coefficient (first correction parameter) for the light source intensity and signal acquisition frequency is dynamically generated based on the user's actual movement state. Its starting point is the triaxial average acceleration data within a reference monitoring period provided by the data acquisition module. Analyzing this data using a motion model, the module first calculates the predicted motion intensity value for the user in the upcoming processing period (reflecting the intensity of physical activity, such as stillness, walking, or running). Subsequently, the module uses preset dynamic mapping rules to divide the predicted motion intensity value into multiple continuous activity intensity intervals (e.g., low / medium / high intensity activity zones), each interval corresponding to a specific correction coefficient. For example, when it is predicted that the user is about to enter a high-intensity running state, a higher correction coefficient (e.g., 1.2) is automatically matched, indicating a need to significantly enhance signal acquisition capabilities to overcome motion interference; if the predicted state is a sitting state, a coefficient close to 1.0 is used to maintain the basic monitoring strategy. This design essentially establishes a real-time translator of "motion intensity - hardware parameter adjustment coefficient," ensuring that the device has the ability to predict and adapt to changes in the motion scenario.

[0025] Furthermore, the first correction module addresses the signal distortion problem caused by motion noise to some extent. For example, when a significant increase in user activity intensity is predicted (e.g., changing from sitting to brisk walking), the module outputs a correction coefficient greater than 1, triggering the subsequent monitoring module to increase the light source intensity and sampling frequency. The enhanced light source can penetrate the ear's blood vessels, which constrict due to accelerated blood flow, preventing PPG signal saturation. A higher sampling frequency captures more raw signal points per unit time, providing redundancy for subsequent signal processing and combating motion artifacts. Conversely, when a decrease in user activity is predicted (e.g., stopping to rest after running), the module outputs a correction coefficient less than 1, guiding the monitoring module to appropriately reduce the light source intensity to avoid signal distortion due to excessive irradiation during vasodilation, while simultaneously reducing the sampling frequency to save energy. This dynamic adjustment mechanism overcomes the rigidity of fixed parameters in the past, enabling ear PPG monitoring to maintain a high signal-to-noise ratio even when the user's activity level changes. This lays the foundation for accurate calculation of key parameters such as blood pressure and blood oxygen, and provides a reliable data source for subsequent differentiated stimulation based on auricular acupoint theory (e.g., reducing the intensity of auricular foot massage after exercise to protect digestive function).

[0026] In the photoplethysmography (PPG) monitoring module, optimal operating parameters for PPG monitoring are configured in real time to ensure the capture of high-quality physiological signals even under complex motion conditions. Its operation relies on two key inputs: the first correction parameter generated by the preceding module (first correction module), and the historical light source intensity and sampling frequency provided by the data acquisition module within the historical monitoring period. The module dynamically adjusts historical hardware parameters based on the first correction parameter—when the predicted user's exercise intensity increases, the light source intensity is proportionally increased to enhance the light's penetration into ear tissues (e.g., to counteract changes in light absorption caused by vasoconstriction), and the sampling frequency is simultaneously increased to acquire more raw data points per unit time; conversely, when the predicted exercise intensity decreases, the light source intensity and sampling frequency are appropriately reduced to save energy and avoid signal overexposure. For example, when a user is jogging in the morning, the module adaptively increases the infrared light source output power to penetrate the accelerated blood flow in the concha capillaries, while simultaneously increasing the sampling frequency by more than three times to eliminate signal spikes caused by foot vibrations; and when the user sits down to rest, the module automatically reverts to the basic monitoring mode. This flexible adjustment of hardware parameters based on motion state prediction ensures the signal-to-noise ratio of PPG raw data from the source.

[0027] Furthermore, after parameter configuration, the module performs signal acquisition tasks based on the optimized current monitoring parameters. Within a single acquisition cycle, it simultaneously acquires multiple sets of blood pressure, blood oxygen, and heart rate data, forming a continuous observation dataset over time. For example, in a 5-second monitoring cycle, the module emits modulated light sources multiple times at millisecond intervals and receives reflected light signals, simultaneously resolving systolic / diastolic blood pressure, blood oxygen saturation, and instantaneous heart rate values ​​at multiple time points. This not only provides information on transient parameter fluctuations (such as the recovery curve of blood oxygen after exercise), but more importantly, it provides a basis for judgment for the subsequent second correction module—by analyzing the distribution characteristics of multiple sets of data within this cycle (such as whether the blood pressure peak exceeds a safe threshold), the second correction module can accurately assess the real-time cardiovascular load status of the user. This design is the first to achieve a closed-loop system of "motion interference suppression - high-quality data acquisition - multi-dimensional physiological assessment" on an ear monitoring device, providing a precise data foundation for the dynamic and dialectical regulation of subsequent ear acupressure massage (such as targeted stimulation of acupoints in the concha based on the shape of the blood pressure curve).

[0028] In the second correction module, a dynamic adjustment coefficient (second correction parameter) for ear massage intensity is generated based on real-time physiological status and exercise load. Its workflow begins with receiving multiple sets of blood pressure, blood oxygen, and heart rate data from the photoplethysmography (PPG) monitoring module, as well as the current exercise intensity prediction value provided by the first correction module. The module first independently calculates the excess impact ratio of four key indicators through a threshold comparison mechanism: whether the exercise intensity exceeds the safe activity threshold (e.g., continuous high-intensity running), whether blood pressure exceeds the upper warning value, whether blood oxygen is below the physiological baseline, and whether heart rate reaches the cardiovascular load limit. For example, when a user experiences a sudden rise in blood pressure and a persistently high heart rate after hiking, a significant hemodynamic overload ratio is calculated. Subsequently, these independent assessment results are input into a comprehensive parameter model, which generates the final correction coefficient through nonlinear fusion. This coefficient is essentially a quantitative feedback on the user's current cardiovascular tolerance; its value typically fluctuates around 1, with values ​​above 1 indicating a need for increased massage intervention and values ​​below 1 indicating a need to reduce stimulation to avoid risks.

[0029] Furthermore, the system adapts to physiological states and biostimulation intensity. When the second correction parameter is applied to the ear massage module, the average massage intensity of the reference period is elastically scaled according to the coefficient: if the parameter is greater than 1 (e.g., when the user's blood pressure is low while sitting), the massage intensity is moderately increased to stimulate acupoints and promote circulation; if the parameter is less than 1 (e.g., when the user's heart rate and / or blood pressure significantly exceed their preset safety threshold), the massage intensity is proportionally reduced to a safe level. This mechanism effectively solves the pain point of "rigid stimulation" in traditional devices. For example, after morning exercise, the module automatically reduces the massage intensity of the concha (corresponding to heart acupoints) by 40% for hypertensive users to avoid overstimulating the vagus nerve and increasing the burden on the heart; while when prolonged sitting at an office leads to digestive dysfunction, the module specifically increases the vibration intensity of the helix crus (corresponding to digestion) by 20%. This dynamic adjustment based on real-time data not only avoids cardiovascular risks but also aligns with the regulatory concept of "treating different diseases with the same method and treating the same disease with different methods" in traditional Chinese medicine auricular therapy.

[0030] In the ear massage module, it is the final execution unit, realizing ear biostimulation based on real-time physiological feedback. Its core workflow is to receive the safety factor (second correction parameter) generated by the second correction module, and combine it with the average massage intensity value of the reference cycle provided by the data acquisition module. The average massage intensity value is multiplied by the second correction parameter to obtain the calculation result, which is used as the actual massage intensity of the current cycle. For example, when the user is in the recovery phase after high-intensity exercise (the second correction parameter equals 1), the module will automatically reduce the global ear massage intensity to avoid aggravating the vagus nerve burden due to strong stimulation before the heart rate has recovered. At the same time, because the module uses a distributed vibration unit array, each unit can independently control the intensity, thus enabling differentiated adjustment for the ear acupoint mapping area—when slow blood oxygen recovery is detected, the low-frequency vibration of the concha cavity area (corresponding to the heart) is automatically enhanced to promote circulation; if digestive indicators are abnormal, the mid-frequency pulse of the helix crus area is increased to regulate gastrointestinal function.

[0031] Furthermore, the ear massage module integrates an inflatable bonding layer. This layer is linked to the air pressure and tightness adjustment modules: when an increase in user activity intensity is detected (such as starting a jog), the inflatable bonding layer automatically inflates to ensure the device adheres tightly to the auricle, eliminating light path misalignment and contact pressure fluctuations caused by shaking; when the user enters a static state (such as sitting down to rest), the inflatable bonding layer will not continue to inflate and increase pressure, and its depressurization can be manually controlled to avoid prolonged pressure discomfort. For example, during high-acceleration sports such as badminton, the bonding layer will inflate by 300% to ensure the device remains perfectly still; while when working at a desk, it maintains basic pressure to ensure comfort. This not only improves the accuracy of PPG signal acquisition (ensuring the light source is stably aligned with the capillary network), but also improves the efficiency of ear acupoint vibration transmission by more than 50% through optimized pressure distribution, avoiding the vicious cycle of existing ear monitoring devices becoming less accurate the looser they are, and vice versa.

[0032] In summary, the entire adaptive ear health monitoring device based on multimodal sensors firstly, the first correction module predicts exercise intensity based on acceleration data, intelligently divides activity intensity ranges, and generates hardware parameter adjustment coefficients, revolutionizing the anti-interference capability of physiological signal acquisition. By predicting the user's exercise state (such as transitioning from sitting to running), it dynamically increases the light source intensity to penetrate constricted blood vessels and avoid PPG saturation, while simultaneously increasing the sampling frequency to capture redundant data and offset motion artifacts, significantly reducing the measurement error of key parameters such as blood pressure and blood oxygen in dynamic scenarios such as walking and running. Furthermore, the second correction module, relying on a comprehensive assessment of real-time physiological data and exercise load, has made a breakthrough in constructing a "physiological risk-ear" system. The "acupoint stimulation" regulation model quantifies massage intensity adjustment parameters through multi-dimensional threshold analysis (such as a sudden increase in blood pressure exceeding the safety threshold or a heart rate reaching the cardiovascular load limit). When a user's heart rate and blood pressure are both high after hiking, the massage intensity of the concha cavity (corresponding to heart acupoints) is reduced proportionally to avoid overstimulation of the vagus nerve. When sitting for a long time, the massage intensity of the helix crus (corresponding to the digestive system) is specifically guaranteed to achieve precise intervention massage. In summary, the two-stage correction forms a monitoring and intervention closed loop. The first correction ensures high-quality data sources, and the second correction dynamically optimizes biostimulation based on these sources. This not only avoids the cardiovascular risks caused by strong massage after exercise, but also enables acupoint zonal treatment through distributed vibration units.

[0033] like Figure 1 and Figure 2 As shown, in one embodiment, the first correction module is further configured to: Multiple distinct and consecutive numerical ranges are defined, each corresponding to a different first correction parameter; Obtain the numerical range into which the motion intensity of the cycle to be processed falls, and obtain the first correction parameter corresponding to that numerical range.

[0034] In this embodiment, it should be noted that the first correction module transforms the continuously changing physical quantity of motion intensity into discrete control parameters. The module presets multiple seamlessly connected numerical intervals (e.g., 0-1 corresponds to sitting, 1-3 to walking, and 3 and above to running), each interval being associated with a specific first correction parameter (e.g., the first correction parameter for sitting is 0.5, for walking is 1, and for running is 1.5). This design essentially constructs a "dictionary-style" mapping relationship between motion scenarios and control strategies: when the motion model calculates the motion intensity value of the user's upcoming processing cycle, it quickly matches this value with a preset interval library (e.g., automatically selecting the corresponding correction coefficient when the intensity value falls within the walking interval), thereby directly transforming the abstract motion state quantity into executable hardware control instructions. For example, when it is predicted that the user will transition from sitting in the office (intensity value 0.8) to commuting while walking (intensity value 2.5), the complex real-time calculation is skipped, and the correction coefficient is instantly switched from 0.5 to 1 through interval matching, achieving seamless switching of the monitoring strategy.

[0035] like Figure 1 and Figure 2 As shown, in one embodiment, the photoplethysmography (PPG) monitoring module is also used for: Obtain the standard light source intensity and standard sampling frequency based on standard monitoring parameters; Configure the current light source intensity according to the first correction parameter and the standard light source intensity, and configure the current sampling frequency according to the first correction parameter and the standard sampling frequency; The current monitoring parameters are generated based on the current light source intensity and the current sampling frequency.

[0036] In this embodiment, it should be noted that when configuring the current monitoring parameters, the photoplethysmography (PPG) monitoring module uses preset standard light source intensity and standard sampling frequency as core base values, rather than relying entirely on historical periodic data. The standard parameters are essentially medically validated basic operating condition configurations (such as the minimum light source power required for clear PPG signal capture in a resting state, and the minimum sampling frequency required to meet heart rate monitoring accuracy), representing the optimal operating benchmark for the device under zero motion interference conditions. The module applies a first correction parameter as a dynamic scaling factor to this standard value: when the predicted exercise intensity increases (e.g., the first correction parameter is 1.5), the standard light source intensity is multiplied by this coefficient to ensure strong light penetrates the constricted blood vessels in the ear due to running; simultaneously, the sampling frequency is increased to several times the standard value, densely capturing signals per unit time to counteract motion artifacts. For example, when a user suddenly transitions from sitting to brisk walking, the standard light source intensity is used as an anchor point, and it is increased by 20% using the first correction coefficient (e.g., 1.2), avoiding potential adjustment lag in historical parameters and preventing signal saturation due to excessive illumination.

[0037] Furthermore, the correction strategy based on standard parameters eliminates the risk of historical data error propagation. Traditional iterative adjustments relying on historical parameters can lead to error accumulation during continuous high-intensity exercise (e.g., signal distortion in the previous cycle causing subsequent light source intensities to remain consistently high). The standard-value-based reset mechanism ensures that each monitoring cycle returns to a medically certified baseline for rescaling. For example, in a marathon scenario, after multiple speed changes, the module periodically recalculates the current light source intensity based on the standard light source intensity (rather than potentially excessively high historical values ​​from the previous cycle) using the latest motion prediction coefficient, fundamentally avoiding data drift caused by strong light exposure during vasoconstriction. The synchronously configured dynamic adjustment of the sampling frequency directly improves the original signal quality—when the standard sampling frequency is 100Hz, if the first correction parameter is 1.8, it immediately switches to 180Hz high-frequency sampling, capturing a raw PPG waveform with 10 times the density of the conventional sample at the moment the user lands, providing sufficient data redundancy for the motion noise filtering algorithm and significantly reducing the error rate of dynamic blood pressure monitoring.

[0038] In one implementation, the motion model in obtaining the motion intensity of the period to be processed based on the motion model and the average acceleration of the x, y, and z axes within the reference monitoring period is represented as follows: ;in, The motion intensity of the cycle to be processed. For reference, the data sampling points within the monitoring period, For the x-axis acceleration of the t-th data sampling point within the reference monitoring period, For the reference monitoring period, the y-axis acceleration of the t-th data sampling point, The z-axis acceleration is the data sampling point t within the reference monitoring period.

[0039] In this embodiment, it should be noted that the vector synthesis (square root operation) of the three-axis acceleration includes the following calculation logic: the (x, y, z) axis acceleration components for each sampling point t. Perform square root and sum of squares calculations. Since the direction of acceleration during human movement is random (e.g., the direction of arm swing varies during running), but the intensity of movement needs to reflect the overall level of physical activity, this calculation synthesizes three-dimensional accelerations into a scalar (the magnitude of a vector), eliminating directional interference and retaining only intensity information. If only single-axis acceleration (e.g., the z-axis) is relied upon, device tilt or localized vibrations can lead to misjudgments (e.g., a head hitting a pillow when stationary is misjudged as movement); three-axis synthesis can resist occasional noise along specific axes.

[0040] Furthermore, The average instantaneous combined acceleration is calculated for all sampling points (t=1 to N). Short-term actions (such as sneezing or brief head shaking) may produce high acceleration values, but they are not continuous motions. The mean calculation can smooth out transient interference and ensure output strength. It reflects the average activity level over the entire reference period. The mean result is directly related to the duration of exercise. For example, the high mean of jogging for 5 minutes is different from the low mean of short jumps, thus distinguishing between low-intensity continuous exercise and high-intensity interval exercise.

[0041] Furthermore, output motion intensity The dynamic mapping effect; output value As a continuous variable, subsequent modules pre-determine the numerical range (e.g., A value not exceeding 1 indicates a stationary state. Points between 1 and 3 indicate walking. (More than 3 are running), dynamically mapped to different sports scenarios. Example: When walking slowly ( This is classified as a low-intensity activity, triggering a mild signal anti-interference strategy; during running ( This is categorized as high-intensity exercise, triggering a strong light source to compensate for vasoconstriction. It also implements predictive regulation, using the exercise intensity from a reference monitoring cycle to predict the current monitoring cycle's state. If the previous cycle... (High intensity) predicts the continued motion in the next cycle and increases the intensity of the PPG light source in advance to avoid signal saturation from the source.

[0042] like Figure 1 and Figure 2 As shown, in one embodiment, the second correction module is further configured to: The proportion of exercise influence is obtained based on the exercise intensity and intensity threshold of the cycle to be processed, the proportion of blood pressure influence is obtained based on multiple sets of current blood pressure data and upper limit blood pressure threshold, the proportion of blood oxygen influence is obtained based on multiple sets of current blood oxygen data and blood oxygen threshold, and the proportion of heart rate influence is obtained based on multiple sets of current heart rate data and upper limit heart rate threshold. The second correction parameter is obtained based on the parameter model, the proportion of influence of exercise, the proportion of influence of blood pressure, the proportion of influence of blood oxygen, and the proportion of influence of heart rate.

[0043] In this embodiment, it should be noted that the second correction module converts exercise and physiological data into calculable risk coefficients through a risk quantification process.

[0044] First, based on the deviation between the exercise intensity of the period to be processed and a preset safety threshold (such as a moderate exercise intensity threshold), the proportion exceeding the safe range is calculated. For example, when a user suddenly sprints 100 meters, the instantaneous exercise intensity may exceed the safety threshold by 50%, and the system automatically determines the cardiovascular stress caused by the surge in exercise load.

[0045] Furthermore, the maximum value from multiple sets of current blood pressure data (systolic / diastolic pressure) is extracted and compared with the hypertension risk threshold. If a momentary breach of the warning line in systolic blood pressure is detected during a measurement (e.g., due to emotional stress), a positive proportional value is generated.

[0046] Furthermore, the minimum value in multiple sets of blood oxygen data was analyzed, and when it was lower than the physiological tolerance baseline (such as a sudden drop in blood oxygen saturation during high-altitude mountain climbing), a proportional coefficient was generated according to the degree of hypoxia.

[0047] Furthermore, by comparing the current peak heart rate data with the upper limit of exercise heart rate (preset according to age and health status), when persistent tachycardia occurs (such as delayed heart rate recovery after exercise), the proportion exceeding the safe range is calculated.

[0048] Furthermore, after obtaining four independent influencing proportions, the module achieves collaborative analysis through a parametric model: a nonlinear function is used to fuse the proportion values. When a single proportion slightly exceeds the limit (e.g., only exercise intensity exceeds the limit by 10%), the output adjustment parameter increases slightly; however, if multiple proportions exceed the limit simultaneously (e.g., exercise intensity exceeds 30% and heart rate exceeds 20%), a risk multiplication effect is triggered based on exponential characteristics, causing the output parameter to decrease sharply. Furthermore, only proportions with positive values ​​(i.e., exceeding the safety threshold) are included in the calculation to avoid diluting core crisis signals with low-risk data. For example: in a low-risk scenario, when sitting still, all proportions are within limits, resulting in a zero value and an output second correction parameter of 1 (maintaining baseline massage); in a high-risk scenario, during a marathon sprint, when exercise intensity exceeds 40% and heart rate exceeds 25%, the output parameter drops sharply to 0.6, compressing the massage intensity by 40%.

[0049] It should also be noted that when a hypertensive patient experiences elevated systolic blood pressure after exercise, the stimulation intensity of the concha (cardiac acupoint) is automatically reduced to a safe level through a blood-impact ratio linkage mechanism, avoiding the risk of "massage increasing the burden on the heart" as with traditional devices. If digestive dysfunction is detected simultaneously (such as slow blood oxygen recovery), the massage ratio of the helix crus (corresponding area of ​​the digestive system) is dynamically increased, achieving synergistic regulation of "cardiac decompression and gastrointestinal motility promotion." In extreme scenarios where users encounter both elevated blood oxygen saturation and blood pressure during mountain climbing, the blood pressure and blood oxygen impact ratios have a cumulative effect, causing the second correction parameter to drop below 0.4. The device immediately switches to ultra-low frequency massage mode, dynamically optimizing stimulation parameters based on cardiovascular status.

[0050] In one implementation, the parameter model for obtaining the second correction parameter based on the parameter model, the exercise intensity of the period to be processed, multiple sets of current blood pressure data, current blood oxygen data, and current heart rate data is represented as follows: ;in, This is the second correction parameter. This is the scaling factor. The motion intensity of the cycle to be processed. The intensity threshold, This represents the number of sampling time points within the current monitoring period. This refers to the blood pressure data at the i-th sampling time point within the current monitoring period. This is the upper limit threshold for blood pressure. This refers to the blood oxygen data at the i-th sampling time point within the current monitoring period. The lowest threshold for blood oxygenation. This refers to the heart rate data at the i-th sampling time point within the current monitoring period. This is the upper limit threshold for heart rate.

[0051] In this embodiment, it should be noted that all four calculations are in the form of relative over-limit ratios: the numerator is the deviation between the actual value and the threshold (e.g., ...). ), the denominator is the threshold itself (e.g. This is used to eliminate dimensional differences, allowing for horizontal comparisons of exercise intensity (g), blood pressure (mmHg), blood oxygen (%), and heart rate (bpm). Simultaneously, the percentage value visually reflects the level of risk: 0% exceeding the limit (actual value ≤ threshold) results in a risk contribution of 0; 20% exceeding the limit results in a risk score of 0.2; and 50% exceeding the limit results in a risk score of 0.5.

[0052] Furthermore, the calculation logic is dynamically adjusted based on the characteristics of the indicators. Higher positive risk indicators, such as the calculated percentage of exercise intensity, blood pressure, and heart rate exceeding the threshold, indicate greater danger. Conversely, lower negative risk indicators, such as the calculated percentage of blood oxygen saturation below the actual value, indicate greater danger. This addresses the shortcomings of traditional models that cannot distinguish indicator characteristics (e.g., blood oxygen saturation requires reverse calculation), and accurately quantifies special risks such as tissue hypoxia.

[0053] Furthermore, Risk activation is implemented: when the actual value does not exceed the threshold, the output is 0 (no risk); when the actual value exceeds the threshold, the output is a proportional value. This filters signals, such as brief fluctuations in heart rate during sitting (80→85bpm) that do not exceed the threshold (150bpm), which are not counted as risk, and only abnormal data that exceeds the safe range are integrated.

[0054] Furthermore, the risks are superimposed through nonlinear responses. Conversion. Low-risk scenario (single indicator slightly exceeding the limit): the sum of the exceedance rates = 0.6. =0.5, =e^{-0.3}≈0.74; thus reducing massage intensity by 26%, indicating a need for gentle intervention. High-risk scenario (multiple indicators deteriorating synergistically): Over-limit ratio = 4, =0.5, =e^{-2}≈0.135, the massage intensity dropped sharply by 86.5%, triggering the safety protection. It also achieved a risk multiplication effect; when the excess ratio increased from 0.5 to 2, The response speed decreased from 0.61 to 0.14 (a reduction of 77%), far exceeding that of the linear model. It only requires a small over-limit ratio to significantly reduce the massage intensity. In particular, the exponential model enters a steep attenuation zone when the over-limit ratio and the value exceeds 1, achieving rapid and safe protection.

[0055] Furthermore, Clinical suitability of the coefficient. Healthy individuals: The massage intensity can be gently adjusted. For cardiovascular patients: Even a slight risk can trigger a sharp drop in intensity. Simultaneously, it can incorporate a dynamic learning mechanism, allowing the device to automatically optimize based on historical data. Values ​​(such as increased blood pressure after repeated exercise) ).

[0056] like Figure 3 As shown, in one embodiment, it further includes an inflatable adhesive layer disposed on the surface of the ear massage module and in contact with the skin, and also includes a tightness adjustment module, the tightness adjustment module being used for: Obtain the average acceleration along the x, y, and z axes at the current moment, and obtain the motion intensity at the current moment based on the average acceleration along the x, y, and z axes at the current moment; The motion trend value is obtained based on the motion intensity of the period to be processed and the motion intensity at the current moment; Adjust the internal pressure of the inflatable bonding layer according to the magnitude of the motion trend value.

[0057] In this embodiment, it should be noted that the tightness adjustment module predicts sudden changes in motion state through real-time acceleration analysis and triggers adaptive adjustment of the bonding layer. Specifically, the tightness adjustment module first acquires the triaxial acceleration data at the current moment, calculates the instantaneous motion intensity (reflecting the intensity of the user's current action, such as a sudden start or stop), and then compares the current intensity with the predicted motion intensity of the processing cycle to generate a motion trend value. For example, when the user's motion trend value is positive, the tightness adjustment module will determine that the motion intensity is increasing. At the same time, the inflatable bonding layer determines whether to increase the inflation volume by one unit based on the positive or negative of the motion trend value, thereby increasing the predetermined pressure and improving the tightness of the device's attachment to the auricle under increasingly intense motion conditions.

[0058] Furthermore, the inflatable bonding layer is pressurized according to the trend value, and through the principle of elasticity, the device is evenly fitted to the curved surface of the ear, eliminating photoelectric displacement caused by sweat or vibration, ensuring that the light source is continuously aligned with the capillary network of the concha, and avoiding PPG signal interruption. It also ensures the reliability of acupoint stimulation, as the bonding pressure is directly related to the efficiency of ear acupoint massage. When the module detects an increasing trend in motion and various physiological parameters are within safe ranges, the inflatable bonding layer automatically inflates and pressurizes, ensuring the efficiency of massage vibration transmission—for example, delivering effective biostimulation for mid-frequency vibration therapy to the crus of the helix (the digestive area).

[0059] In one embodiment, adjusting the pressure of the inflatable bonding layer according to the magnitude of the movement tendency is expressed as follows: , ;in, For the volume increase of the inflatable bonding layer, For a single volume standard increment, This is a trend in sports.

[0060] In this embodiment, it should be noted that, To determine the direction of movement trends, and at the same time... Quantify the direction of deviation between the current action and the prediction. Among them, This indicates that the user's exercise intensity increased more than expected; This indicates that the user's exercise intensity was lower than expected; This indicates that the actual motion matches the prediction. It distinguishes between pressurized and sustained scenarios. When, boost is triggered; when At that time, the max function is used to filter out, so that Without adjustment, the pressure is maintained.

[0061] Furthermore, Used for constraint control, the output value is limited to... The value is larger than 0. This prevents negative volume adjustments from being performed, avoiding ineffective pressure reduction and preventing mechanical pressure reduction when the activity intensity decreases (e.g., from running to walking), as the bonding layer needs to maintain a base pressure to ensure signal stability. In conjunction with the active pressure relief control transfer, pressure reduction operations are only allowed to be manually triggered by the user (e.g., by pressing a button), preventing accidental loosening caused by automatic pressure relief from the algorithm.

[0062] Furthermore, the single volume standard increment Achieving a stepped boost design and achieving an engineering balance in boost accuracy: setting For a fixed increment (e.g., 0.1 mL), rather than according to... Adjust the proportions to avoid frequent fine-tuning that could cause air pressure oscillations. For example, when a basketball jumps... (High intensity deviation), only a single increase The device is compact and fits snugly in one step. The number of continuous pressure applications is limited (e.g., a maximum of 3 times) to prevent excessive pressure during exercise from causing blood flow obstruction in the auricle.

[0063] like Figure 1 As shown, in one embodiment, the ear massage module includes a control layer, within which multiple distributed vibration units are disposed, the vibration intensity of the multiple vibration units being equal to the current massage intensity.

[0064] In this embodiment, it should be noted that the core of the ear massage module lies in the integration of an independently controllable multi-point vibration array to achieve precise dynamic regulation of traditional Chinese medicine auricular therapy: dozens of micro-vibration units are embedded in the control layer, precisely arranged according to the anatomical structure of the ear (e.g., heart acupoint units are densely distributed in the concha cavity, and digestive units are distributed in the crus of the helix). The vibration intensity of each unit is synchronized with the current massage intensity value in real time, but differential stimulation is achieved through region-specific frequency modulation. For example, when it is detected that the user's digestive function is inhibited due to prolonged sitting, the helix crus unit is automatically switched to a 15Hz mid-frequency pulse (to promote gastrointestinal motility), while maintaining a 10Hz low-frequency vibration in the concha cavity (to stabilize the heart rate).

[0065] Furthermore, each unit can be equipped with a built-in pressure feedback sensor. When a change in local skin impedance is detected (such as vasoconstriction in the concha after exercise), the amplitude in that area is automatically increased to compensate for energy loss and ensure constant stimulation efficiency. Taking a hypertensive patient after a morning run as an example: while reducing the overall massage intensity, the baseline amplitude of the area where blood pressure-related acupoints (such as the auricular groove and sympathetic acupoints) are located is specifically maintained and regulated to achieve protective treatment.

[0066] like Figure 4 As shown, in one embodiment, a communication module is also included, which is connected to an interactive terminal for setting the basic massage intensity and viewing historical monitoring data.

[0067] In this embodiment, it should be noted that the communication module constructs a closed-loop medical decision-making system. Its core function is to realize the personalized setting of basic parameters through the interactive terminal. Clinicians can set the user's exclusive massage intensity baseline through the terminal (such as setting the upper limit for coronary heart disease patients to 60% of that for healthy people) and dynamically adjust it in conjunction with the medication time (such as prohibiting strong stimulation of the concha cavity 2 hours after taking antihypertensive drugs).

[0068] Furthermore, it has achieved multi-dimensional health data fusion analysis. The terminal integrates historical monitoring data and acupoint stimulation records, and generates an auricular acupoint therapy-visceral function response map with relevant TCM content, and pushes treatment suggestions, transforming TCM experience into data-driven decision-making.

[0069] The preferred embodiments of this disclosure have been described in detail above with reference to the accompanying drawings. However, this disclosure is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this disclosure, various simple modifications can be made to the technical solutions of this disclosure, and these simple modifications all fall within the protection scope of this disclosure.

[0070] It should also be noted that the various specific technical features described in the above embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, this disclosure will not describe the various possible combinations separately.

[0071] Furthermore, various different embodiments of this disclosure can be combined in any way, as long as they do not violate the spirit of this disclosure, they should also be regarded as the content disclosed in this disclosure.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. An adaptive ear health monitoring device based on multimodal sensors, characterized in that, include: The data acquisition module is used to acquire the previous monitoring cycle of the current monitoring cycle as a reference monitoring cycle, and to acquire the average acceleration and average massage intensity of the x, y, and z axes within the reference monitoring cycle. The first correction module is used to obtain the motion intensity of the period to be processed based on the motion model and the average acceleration of the x, y, and z axes within the reference monitoring period, and to obtain the first correction parameter based on the motion intensity of the period to be processed. The photoplethysmography (PPG) monitoring module is used to acquire standard monitoring parameters, and to acquire current monitoring parameters based on the first correction parameter and the standard monitoring parameters. Within the current monitoring cycle, it acquires multiple sets of current blood pressure data, current blood oxygen data, and current heart rate data according to the current monitoring parameters. The second correction module is used to obtain the second correction parameters based on the parameter model, the exercise intensity of the period to be processed, multiple sets of current blood pressure data, current blood oxygen data and current heart rate data; The ear massage module is used to obtain the current massage intensity based on the second correction parameter and the average massage intensity, and to perform ear massage according to the current massage intensity.

2. The adaptive ear health monitoring device based on a multimodal sensor according to claim 1, characterized in that, The first correction module is also used for: Multiple distinct and consecutive numerical ranges are defined, each corresponding to a different first correction parameter; Obtain the numerical range into which the motion intensity of the cycle to be processed falls, and obtain the first correction parameter corresponding to that numerical range.

3. The adaptive ear health monitoring device based on a multimodal sensor according to claim 1, characterized in that, The photoplethysmography (PPG) monitoring module is also used for: Obtain the standard light source intensity and standard sampling frequency based on standard monitoring parameters; Configure the current light source intensity according to the first correction parameter and the standard light source intensity, and configure the current sampling frequency according to the first correction parameter and the standard sampling frequency; The current monitoring parameters are generated based on the current light source intensity and the current sampling frequency.

4. The adaptive ear health monitoring device based on a multimodal sensor according to claim 1, characterized in that, The motion model used to obtain the motion intensity of the processing period based on the average acceleration of the x, y, and z axes within the reference monitoring period is expressed as follows: ;in, The motion intensity of the cycle to be processed. For reference, the data sampling points within the monitoring period, For the x-axis acceleration of the t-th data sampling point within the reference monitoring period, For the reference monitoring period, the y-axis acceleration of the t-th data sampling point, The z-axis acceleration is the data sampling point t within the reference monitoring period.

5. The adaptive ear health monitoring device based on a multimodal sensor according to claim 1, characterized in that, The second correction module is also used for: The proportion of exercise influence is obtained based on the exercise intensity and intensity threshold of the cycle to be processed, the proportion of blood pressure influence is obtained based on multiple sets of current blood pressure data and upper limit blood pressure threshold, the proportion of blood oxygen influence is obtained based on multiple sets of current blood oxygen data and blood oxygen threshold, and the proportion of heart rate influence is obtained based on multiple sets of current heart rate data and upper limit heart rate threshold. The second correction parameter is obtained based on the parameter model, the proportion of influence of exercise, the proportion of influence of blood pressure, the proportion of influence of blood oxygen, and the proportion of influence of heart rate.

6. The adaptive ear health monitoring device based on a multimodal sensor according to claim 1, characterized in that, The parameter model used to obtain the second correction parameter based on the parameter model, the exercise intensity of the period to be processed, multiple sets of current blood pressure data, current blood oxygen data, and current heart rate data is expressed as follows: ;in, This is the second correction parameter. This is the scaling factor. The motion intensity of the cycle to be processed. The intensity threshold, This represents the number of sampling time points within the current monitoring period. This refers to the blood pressure data at the i-th sampling time point within the current monitoring period. This is the upper limit threshold for blood pressure. This refers to the blood oxygen data at the i-th sampling time point within the current monitoring period. The lowest threshold for blood oxygenation. This refers to the heart rate data at the i-th sampling time point within the current monitoring period. This is the upper limit threshold for heart rate.

7. The adaptive ear health monitoring device based on a multimodal sensor according to claim 1, characterized in that, It also includes an inflatable adhesive layer disposed on the surface of the ear massage module and in contact with the skin, and a tightness adjustment module, the tightness adjustment module being used for: Obtain the average acceleration along the x, y, and z axes at the current moment, and obtain the motion intensity at the current moment based on the average acceleration along the x, y, and z axes at the current moment; The motion trend value is obtained based on the motion intensity of the period to be processed and the motion intensity at the current moment; Adjust the internal pressure of the inflatable bonding layer according to the magnitude of the motion trend value.

8. The adaptive ear health monitoring device based on a multimodal sensor according to claim 7, characterized in that, The adjustment of the pressure of the inflatable bonding layer according to the magnitude of the movement trend is expressed as follows: , ;in, For the volume increase of the inflatable bonding layer, For a single volume standard increment, This is a trend in sports.

9. The adaptive ear health monitoring device based on a multimodal sensor according to claim 1, characterized in that, The ear massage module includes a control layer, within which multiple distributed vibration units are provided, and the vibration intensity of the multiple vibration units is equal to the current massage intensity.

10. The adaptive ear health monitoring device based on a multimodal sensor according to claim 1, characterized in that, It also includes a communication module, which is connected to an interactive terminal, used to set the basic massage intensity and view historical monitoring data.

Images