Touch panel biological impedance physiological monitoring method and device and computer equipment

Abstract

The invention relates to the technical field of biological impedance monitoring, and discloses a touch panel biological impedance physiological monitoring method and device and computer equipment. The method comprises the steps that a user is guided to execute standardized operation to collect personalized artifact signals, and the personalized artifact signals are compared with universal template signals to obtain a deviation coefficient matrix; identifying a current operation state in a physiological monitoring mode, querying a general artifact impedance template, and performing amplitude scaling and fluctuation adjustment on a deviation coefficient and the template to obtain a personalized artifact baseline; performing differential operation on the measured impedance signal and the personalized artifact baseline to obtain a net physiological impedance signal, collecting a new artifact sample to calculate an instantaneous deviation coefficient, and updating a deviation coefficient matrix through a weighted average strategy; according to the method and the device, the technical problem of adaptability caused by individual difference of different user artifact modes and the technical problem that motion artifacts interfere with physiological signals are solved.

Description

Technical Field

[0001] This application relates to the field of bioimpedance monitoring technology, and in particular to a method, device and computer equipment for bioimpedance physiological monitoring of a touch panel. Background Technology

[0002] The touchpad bioimpedance physiological monitoring method is a non-invasive physiological parameter monitoring technology that integrates bioimpedance measurement technology into laptop touchpads. This method utilizes the physiological principle that human tissue exhibits different impedance characteristics to currents of different frequencies. By placing multiple monitoring devices in specific areas of the touchpad, when a user's finger touches these monitoring locations, the system applies a weak alternating current signal to the human tissue and measures the voltage response after passing through the tissue, thereby calculating the bioimpedance value. Since changes in heartbeat, blood flow, and blood oxygen saturation cause periodic fluctuations in the impedance of human tissue, by filtering, amplifying, and extracting features from the collected impedance signals, real-time monitoring of physiological parameters such as heart rate and blood oxygen saturation can be achieved. This allows the laptop touchpad to retain its original input function while adding the added value of health monitoring.

[0003] Existing touchpad bioimpedance physiological monitoring methods suffer from the following shortcomings: First, insufficient monitoring accuracy and stability. Due to strong electromagnetic interference in the touchpad environment, and significant variations in factors such as the contact pressure, contact area, and skin humidity between the user's finger and the electrode, the impedance measurement signal noise is relatively low, affecting the accuracy of physiological parameter extraction. Second, the layout and number of monitoring devices are not rationally set. Existing solutions often use electrodes in fixed positions, which cannot adapt to the differences in hand shapes and usage habits of different users. This can easily lead to situations where the finger cannot stably contact multiple monitoring positions simultaneously, reducing the convenience and success rate of monitoring. Third, a lack of effective motion artifact suppression mechanisms. When users perform normal operations on the touchpad, finger movement and pressure changes introduce a large number of motion artifacts, interfering with the real physiological signals. Existing technologies struggle to maintain continuous and stable monitoring effects in dynamic usage scenarios. They typically use fixed artifact elimination parameters or universal templates, which cannot adapt to individual differences among different users, leading to overcompensation or undercompensation in artifact elimination. Fourth, the ability to simultaneously monitor multiple physiological parameters is limited. Existing solutions mostly focus on the measurement of a single parameter, which is insufficient for health assessment indicators that require joint analysis of multiple parameters, thus limiting the application value of the system. Summary of the Invention

[0004] This application provides a method, device, and computer equipment for bioimpedance physiological monitoring of a touchpad. By constructing a pure motion artifact database and calculating a personalized deviation coefficient matrix, it achieves dynamic reconstruction from a general artifact template to a personalized artifact baseline, solving the adaptation problem caused by individual differences in artifact patterns among different users and the technical problem of motion artifacts interfering with physiological signals. This application continuously tracks changes in the user's physiological state through an online incremental update mechanism, solving the problem of dynamic degradation caused by the failure of the initial calibration deviation coefficient over time, and improving the measurement accuracy and stability in long-term continuous monitoring scenarios.

[0005] In a first aspect, this application provides a method for monitoring bioimpedance physiology on a touchpad, the method comprising: Secondly, this application provides a touchpad bioimpedance physiological monitoring device, the touchpad bioimpedance physiological monitoring device comprising: The acquisition module is used to acquire impedance change sequences corresponding to user sliding operations, clicking operations, and stationary hovering operations in non-monitoring mode. The impedance change sequences are classified according to operation mode based on moving speed and contact pressure. A pure motion artifact database is constructed from the impedance time series corresponding to each operation mode. The comparison module is used to guide users to execute standardized operation sequences, collect personalized artifact signals, and compare the amplitude and fluctuation of the personalized artifact signals with the general template signals of the corresponding operation modes in the pure motion artifact database to obtain a deviation coefficient matrix. The identification module is used to identify the current operating state in real time under physiological monitoring mode. Based on the current operating state, it queries a general artifact impedance template from the pure motion artifact database, and performs amplitude scaling and fluctuation adjustment on the deviation coefficients of the corresponding row in the deviation coefficient matrix and the general artifact impedance template to obtain a personalized artifact baseline. The update module is used to perform differential calculation between the measured impedance signal and the personalized artifact baseline to obtain the net physiological impedance signal. When it is detected that the user leaves the monitoring position and operates the touchpad again, new artifact samples are collected and the instantaneous deviation coefficient is calculated. The deviation coefficient matrix is ​​updated by a sliding window weighted average strategy.

[0006] Thirdly, a computer device is provided, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor invokes the instructions in the memory to cause the computer device to perform the above-described touchpad bioimpedance physiological monitoring method.

[0007] Fourthly, a computer-readable storage medium is provided, wherein instructions are stored therein, which, when executed on a computer, cause the computer to perform the above-described touchpad bioimpedance physiological monitoring method.

[0008] The technical solution provided in this application collects impedance change sequences corresponding to user sliding operations, clicking operations, and stationary hovering operations in non-monitoring mode. These impedance change sequences are categorized by operation mode based on movement speed and contact pressure. A pure motion artifact database is constructed from the impedance time series corresponding to each operation mode. This technical feature enables the system to actively learn and accumulate pure motion artifact features even when the user is not physiologically monitoring, establishing an accurate benchmark template library for subsequent artifact identification and elimination, and avoiding template contamination problems caused by the mixing of physiological signals and motion artifacts. The system guides the user to execute standardized operation sequences to collect personalized artifact signals. The amplitude and fluctuation of these personalized artifact signals are compared with the general template signals of the corresponding operation modes in the pure motion artifact database to obtain a deviation coefficient matrix. This technical feature bridges the gap from general templates to personalized applications by quantifying individual differences. The deviation coefficient matrix accurately describes the degree of deviation of the current user's artifact features from the general template, providing a personalized correction basis for dynamic reconstruction. In physiological monitoring mode, the current operating state is identified in real time. Based on the current operating state, a general artifact impedance template is queried from the pure motion artifact database. The deviation coefficients of the corresponding row in the deviation coefficient matrix are compared with the general artifact impedance template for amplitude scaling and fluctuation adjustment to obtain a personalized artifact baseline. This technology enables adaptive dynamic reconstruction of the artifact baseline. By multiplying the general template by the personalized deviation coefficient, the artifact baseline is made to accurately match the real artifact characteristics of the current user in the current operating state. Compared with the existing technology that uses fixed parameters or general templates, the matching accuracy of artifact elimination is significantly improved.

[0009] The net physiological impedance signal is obtained by differentially calculating the measured impedance signal and the personalized artifact baseline. This technique effectively separates the true physiological information through precise artifact elimination, laying a signal quality foundation for the accurate extraction of heart rate and blood oxygen parameters. When the user leaves the monitoring position and re-operates the touchpad, a new artifact sample is collected and the instantaneous deviation coefficient is calculated. The deviation coefficient matrix is ​​updated using a sliding window weighted averaging strategy. This technique solves the dynamic degradation problem of the initial deviation coefficient becoming invalid due to changes in the user's physiological state over time. The online incremental update mechanism ensures that the deviation coefficient matrix continuously tracks the user's current true state, guaranteeing the continuous effectiveness of artifact elimination during long-term monitoring. From the perspective of algorithmic contribution, the core innovation of the adaptive multi-channel impedance baseline dynamic reconstruction algorithm lies in establishing a correlation mapping relationship between the pure motion artifact database, personalized deviation coefficients, and dynamic artifact baselines. The algorithm realizes personalized scaling and fluctuation adjustment of the general template through the deviation coefficient matrix, enabling the artifact baseline to adaptively match the individual differences of different users and parameter changes in different operating states. In the specific application field of touchpad bioimpedance physiological monitoring, the algorithm effectively solves the three core technical challenges of large individual differences in motion artifacts, variable operating states, and dynamic changes in physiological states. Compared with traditional fixed parameter artifact elimination methods, it significantly improves the monitoring accuracy and robustness in dynamic usage scenarios. Attached Figure Description

[0010] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is a schematic diagram of one embodiment of the touchpad bioimpedance physiological monitoring method in this application. Figure 2 This is a schematic diagram illustrating the effect of dynamic artifact baseline parameter matching adjustment in the embodiments of this application; Figure 3 This is a schematic diagram of the detection of the peak value of the net physiological impedance signal in an embodiment of this application. Detailed Implementation

[0012] This application provides a method, apparatus, and computer device for monitoring bioimpedance physiological parameters using a touchpad. The terms "first," "second," "third," "fourth," etc. (if present)," in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0013] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the touchpad bioimpedance physiological monitoring method in this application includes: Step S1: In non-monitoring mode, collect the impedance change sequences corresponding to user sliding operation, clicking operation and stationary hovering operation. Classify the impedance change sequences according to the operation mode based on the moving speed and contact pressure. Construct a pure motion artifact database from the impedance time series corresponding to each operation mode. In the non-monitoring mode, pure motion artifact signals generated by touchpad operations are collected. In this mode, the user's finger is not placed in the preset position of the monitoring electrode but performs daily operations in any area of ​​the touchpad. The collected impedance changes are entirely due to the mechanical effects of finger contact pressure and movement speed, without containing physiological information. The system simultaneously collects impedance signals, pressure sensor data, and speed data output by the touchpad controller. The three types of data are aligned by timestamp to form a triplet sequence. Based on the speed data, swiping operations are divided into three intervals: slow, medium, and fast. Based on the pressure peak value, clicking operations are divided into three intervals: mild, moderate, and severe. Based on the pressure standard deviation, a stable interval for stationary hovering is selected, forming nine subdivided operation modes. The impedance time series of each mode, along with their corresponding pressure series, speed series, and operation labels, are stored to construct a pure motion artifact database. This database serves as a benchmark template library for subsequent identification and elimination of motion artifacts.

[0014] Step S2: Guide the user to execute a standardized operation sequence, collect personalized artifact signals, and compare the amplitude and fluctuation of the personalized artifact signals with the general template signals of the corresponding operation modes in the pure motion artifact database to obtain the deviation coefficient matrix; The process involves guiding users through a standardized sequence of actions via a personalized calibration procedure to collect personalized artifact signals. These actions include constant-speed sliding, target pressure clicking, and stationary hovering. The impedance mean and standard deviation extracted from the personalized sliding artifact signal are divided by the corresponding parameters of the general template for sliding operations in the pure motion artifact database to calculate the amplitude deviation coefficient and fluctuation deviation coefficient for the sliding operation. Similarly, the pulse peak value and pulse width are extracted from the personalized clicking artifact signal to calculate the deviation coefficient for the clicking operation, and the baseline value and drift slope are extracted from the personalized stationary artifact signal to calculate the deviation coefficient for the stationary operation. These deviation coefficients quantify the individual differences in the current user's artifact characteristics relative to the general template and are organized and stored as a deviation coefficient matrix.

[0015] Step S3: In physiological monitoring mode, identify the current operation status in real time, query the general artifact impedance template from the pure motion artifact database according to the current operation status, and adjust the amplitude scaling and fluctuation of the deviation coefficients in the corresponding row of the deviation coefficient matrix with the general artifact impedance template to obtain the personalized artifact baseline. The core algorithmic improvement of this invention is its adaptive multi-channel impedance baseline dynamic reconstruction mechanism, which solves the technical problem that a universal artifact template cannot adapt to the individual differences of different users. In physiological monitoring mode, impedance signals, pressure signals, and operational events are analyzed in real time. The system determines the stationary, click, or sliding state through first-order pressure differential analysis, and queries the corresponding universal artifact impedance template from a pure motion artifact database based on the identified current operational state label. The key innovation lies in applying the deviation coefficients of the corresponding rows in the deviation coefficient matrix to the universal template. First, the template's overall amplitude is scaled using amplitude-type deviation coefficients, and then the standard deviation of the template is adjusted according to fluctuation-type deviation coefficients to match the fluctuation characteristics, resulting in a personalized artifact baseline adapted to the current user's physiological characteristics. This dynamic reconstruction process transforms the universal template into a baseline that accurately reflects the individual artifact pattern, significantly improving the adaptation accuracy compared to existing artifact elimination methods that use fixed parameters.

[0016] Step S4: Perform differential calculation between the measured impedance signal and the personalized artifact baseline to obtain the net physiological impedance signal. When the user leaves the monitoring position and re-operates the touchpad, new artifact samples are collected and the instantaneous deviation coefficient is calculated. The deviation coefficient matrix is ​​updated using a sliding window weighted average strategy.

[0017] The process involves differentially calculating the real-time measured impedance signal with a personalized artifact baseline to eliminate motion artifacts and obtain a net physiological impedance signal containing heart rate and blood oxygenation information. To address the issue of initial deviation coefficients becoming invalid due to changes in the user's physiological state over time, an online incremental update mechanism is designed. When the system detects that the user has left the monitoring position and is re-operating the touchpad, new artifact samples are automatically collected. The feature parameters of the new samples are extracted and compared with a general template to calculate the instantaneous deviation coefficient. A sliding window weighted averaging strategy is used to fuse the current deviation coefficient matrix with the instantaneous deviation coefficient matrix, ensuring that the deviation coefficient continuously tracks changes in the user's actual physiological state and maintains the sustained effectiveness of artifact elimination during long-term monitoring. For example, if the user's finger temperature is low and skin impedance is high at the start of monitoring, and the finger temperature rises after a period of time, causing skin impedance to decrease, the initially calibrated deviation coefficient is no longer accurate. The online incremental update mechanism re-collects artifact samples of the current state to calculate new deviation coefficients. The updated deviation coefficient matrix accurately reflects the impedance characteristics after the temperature rise, thus maintaining the accuracy of artifact elimination.

[0018] In one specific embodiment, step S1 includes: Multiple metal electrodes are set in a preset area on the touch panel surface as a bioimpedance monitoring device. A sinusoidal AC excitation current is applied to the excitation electrode pair through a constant current source. The voltage signal between the measurement electrode pair is collected and the real-time impedance value is calculated. In non-monitoring mode, the user is guided to place their finger on any position on the touchpad to perform routine operations. The impedance signal, the contact pressure data output by the touchpad's built-in pressure sensor, and the finger movement speed data output by the touchpad controller are collected simultaneously. The three types of data are aligned by timestamp to form a triplet sequence. The sliding operation data in the triplet sequence is divided into speed intervals based on the movement speed, the click operation data is divided into pressure intervals based on the peak contact pressure, and the pressure stability interval is extracted from the static hovering data based on the pressure standard deviation. The impedance time series, corresponding pressure series, velocity series, and operation mode labels collected under each subdivided operation mode are stored separately to construct a pure motion artifact database.

[0019] Specifically, the metal electrodes are made of a material with good conductivity and are electrically connected to the touchpad circuit system. The first and second electrodes form the excitation electrode pair, and the third and fourth electrodes form the measurement electrode pair. The four electrodes are arranged in a rectangular pattern on the touchpad surface to form an effective current path. The sinusoidal AC excitation current applied to the excitation electrode pair by the constant current source is selected at a frequency of 50 kHz. At this frequency, the impedance characteristics of human tissue are sensitive to changes in physiological parameters without causing discomfort. The amplitude of the excitation current is controlled at 100 microamps to ensure safety. The voltage signal collected by the measurement electrode pair is differentially amplified and converted from analog to digital. According to Ohm's law, the voltage value is divided by the known amplitude of the excitation current to calculate the real-time impedance value. The sampling rate is set at 2 kHz to capture rapid fluctuations in the impedance signal. In non-monitoring mode, the user's finger is not placed at a specific position on the monitoring electrode but is used for everyday operations such as sliding and clicking in any area of ​​the touchpad. The impedance changes collected at this time are entirely due to the mechanical effects of finger contact pressure and movement speed, and do not include physiological information related to heart rate or blood oxygenation. These signals constitute pure motion artifacts.

[0020] Each element in the triplet sequence contains the impedance, pressure, and velocity values ​​at the same moment, ensuring strict synchronization of the three types of data in the time dimension. Velocity intervals are defined by setting velocity thresholds, extracting time periods in the triplet sequence where the velocity data is not zero. Velocities less than 20 mm / s are classified as slow sliding intervals, speeds between 20 and 50 mm / s as medium sliding intervals, and speeds greater than 50 mm / s as fast sliding intervals. Pressure intervals are defined by identifying the peak values ​​of the pressure curve, extracting time periods in the triplet sequence where the pressure exhibits a pattern of first rising and then falling. Peak pressure less than 200 g / s is classified as a light click interval, peak pressure between 200 and 400 g / s as a medium click interval, and peak pressure greater than 400 g / s as a heavy click interval. Stable pressure intervals are selected by calculating the standard deviation of pressure within a continuous time window; a standard deviation less than 10 g / s indicates stable pressure, corresponding to a static hovering state. Each record in the pure motion artifact database contains a complete impedance time series, pressure time series, velocity time series, and operation mode label that identifies the operation type and level under a specific operation mode. This database serves as a benchmark template library for subsequent artifact identification and elimination.

[0021] In one specific embodiment, step S3 includes: The impedance signal, pressure signal and operation event are jointly analyzed by a sliding window of set duration. The first-order difference of pressure within the window is extracted as the pressure change rate feature. The static state is determined by the absolute value of the pressure change rate of continuous sampling points, the click state is determined by the monotonic change pattern of pressure value, and the sliding state is determined by the periodic fluctuation of mouse movement event and pressure change rate. The current operation state label is obtained. Based on the current operation status label, extract the corresponding general artifact impedance time series template from the pure motion artifact database, and extract the amplitude-type deviation coefficient and fluctuation-type deviation coefficient of the corresponding row from the deviation coefficient matrix. Amplitude scaling and fluctuation adjustment are performed on the general artifact impedance time series template to obtain a personalized artifact baseline.

[0022] Specifically, the duration of the sliding window is set to the hundreds of milliseconds level. This duration captures instantaneous changes in the operation state while avoiding overly frequent state switching judgments. The first-order difference of the pressure signal within the window is calculated by subtracting the pressure values ​​of adjacent sampling points, i.e., the pressure value at the current moment minus the pressure value at the previous moment. This difference reflects the rate of pressure change over time, i.e., the pressure change rate. The static state is determined by checking whether the absolute value of the pressure change rate at multiple consecutive sampling points is less than a first judgment threshold. When the absolute value of the pressure change rate at consecutive sampling points is consistently below this threshold, it indicates that the pressure remains stable without significant fluctuations, and is judged as a static state. The click state is determined based on the monotonic change pattern of the pressure value. First, it checks whether there are consecutive sampling points with a pressure change rate greater than zero (i.e., monotonically increasing pressure). Then, it checks whether a pressure peak is followed by a pressure change rate less than zero (i.e., monotonically decreasing pressure). When the duration of the complete process of first increasing and then decreasing is within a set range, it is judged as a click state. The sliding state determination requires the simultaneous fulfillment of two conditions. The first condition is that the touchpad controller reports a mouse movement event, i.e., detects a change in finger position. The second condition is that the pressure change rate exhibits periodic fluctuations. These periodic fluctuations are determined by calculating the autocorrelation function of the pressure change rate sequence. When the autocorrelation function shows a peak at a specific lag time, it indicates the existence of periodicity. The frequency of these periodic fluctuations corresponds to the periodic changes in contact pressure when the finger slides on the touchpad surface. Through the above determination logic, the operation state label corresponding to the current sliding window is obtained. This label indicates whether the current state is static, clicked, or sliding.

[0023] Based on the current operation status label, the corresponding operation mode category is located in the pure motion artifact database. This category stores multiple impedance time series collected under this operation mode. These series are time-aligned and arithmetically averaged to obtain a general artifact impedance time series template. This template represents the average artifact characteristics of multiple users under this operation mode. Deviation coefficients are extracted from the deviation coefficient matrix according to the operation status label. For a sliding state, the amplitude and fluctuation deviation coefficients of the first row are extracted; for a click state, the deviation coefficients of the second row are extracted; and for a stationary state, the deviation coefficients of the third row are extracted. Amplitude scaling is achieved by multiplying the impedance value at each moment in the general artifact impedance time series template by the amplitude deviation coefficient. The overall amplitude of the scaled impedance template is adjusted to match the artifact amplitude characteristics of the current user. Fluctuation adjustment first calculates the standard deviation of the scaled impedance template, reflecting the dispersion of the template signal around the mean. Then, the standard deviation is multiplied by a fluctuation-type deviation coefficient to obtain the target standard deviation. Next, the mean of the scaled impedance template is subtracted to obtain a zero-mean signal. This zero-mean signal is then multiplied by the ratio of the target standard deviation to the original standard deviation to adjust the fluctuation amplitude. Finally, the mean is added back to obtain the fluctuation-adjusted impedance template. The impedance template after amplitude scaling and fluctuation adjustment becomes the personalized artifact baseline. This baseline accurately reflects the artifact characteristics of the current user under the current operating conditions, and can more accurately match individual differences compared to a generic template.

[0024] In one specific embodiment, after amplitude scaling and fluctuation adjustment of the general artifact impedance time series template, the method further includes: The mean pressure change rate and mean velocity within the sliding window are extracted as real-time feature parameters, and the standard pressure change rate and standard velocity are extracted from the pure motion artifact database as template feature parameters. The parameter scaling factor is calculated based on the real-time feature parameters and the template feature parameters. For sliding operations, the time axis of the personalized artifact baseline is scaled. For clicking operations, the amplitude of the personalized artifact baseline is scaled. For stationary operations, a linear drift component is superimposed on the personalized artifact baseline to obtain the dynamic artifact baseline.

[0025] Specifically, the average pressure change rate within the sliding window is obtained by calculating the arithmetic mean of all first-order pressure differences within the window. This mean reflects the average rate of pressure change at the current moment. The average speed is obtained by calculating the arithmetic mean of all speed samples within the window. This mean reflects the average speed of finger movement at the current moment. From the pure motion artifact database, the corresponding operation mode category is located based on the current operation state label. The standard pressure change rate and standard speed stored under this category are extracted as template feature parameters. The standard pressure change rate is the statistical mean of multiple pressure change rates collected under this operation mode, and the standard speed is the statistical mean of multiple speeds collected under this operation mode. For sliding operations, a speed scaling factor is calculated. This factor equals the real-time average speed divided by the standard speed. A speed scaling factor greater than 1 indicates that the current sliding speed is faster than the standard speed, and a speed scaling factor less than 1 indicates that the current sliding speed is slower than the standard speed. For click operations, a pressure scaling factor is calculated. This factor equals the real-time average pressure change rate divided by the standard pressure change rate. The pressure scaling factor reflects the degree of deviation of the pressure change rate of the current click action from the standard click. For static operation, the pressure difference is calculated. This difference is equal to the average real-time pressure change rate minus the standard pressure change rate. The pressure difference reflects the deviation of the pressure drift trend in the current static state from the standard static state.

[0026] The timeline scaling for sliding operations is achieved by changing the time sampling points of the personalized artifact baseline. Each moment in the original time series of the baseline is multiplied by a velocity scaling factor to obtain a new time series. When the velocity scaling factor is greater than 1, the timeline is compressed, meaning more artifact periods are included within the same duration; when the velocity scaling factor is less than 1, the timeline is stretched, meaning fewer artifact periods are included within the same duration. Impedance values ​​are resampled on the new time series using linear interpolation to obtain an artifact baseline that matches the current sliding velocity. The amplitude scaling for click operations is achieved by multiplying the impedance value at each moment of the personalized artifact baseline by a pressure scaling factor. When the pressure scaling factor is greater than 1, the overall amplitude of the artifact baseline is amplified; when the pressure scaling factor is less than 1, the overall amplitude of the artifact baseline is reduced. The scaled baseline amplitude matches the pressure variation characteristics of the current click operation. For the linear drift component superposition during static operation, the drift slope is first calculated based on the pressure difference. The drift slope equals the pressure difference multiplied by the standard drift coefficient for that operation mode stored in the database. Then, a linear drift function is constructed, which has a value of zero at the zero point of time and grows linearly with time; the slope is the drift slope. The value of this linear drift function is added point by point to the personalized artifact baseline to obtain a baseline containing the drift component. After the above parameter matching and adjustment for different operation states, a dynamic artifact baseline is obtained. This baseline not only adapts to individual user differences but also accurately matches the real-time parameter characteristics of the current operation.

[0027] Figure 2This is a schematic diagram illustrating the effect of dynamic artifact baseline parameter matching adjustment in the embodiments of this application; Figure 2 This paper presents a comparison of the effects of parameter matching adjustment on personalized artifact baselines based on real-time characteristic parameters. In the figure, the solid line represents the real-time measured impedance signal, the dashed line represents the general template baseline obtained from a pure motion artifact database, and the dotted-dash line represents the personalized artifact baseline obtained after amplitude scaling and fluctuation adjustment using the deviation coefficient matrix. The amplitude and fluctuation characteristics of the general template baseline deviate significantly from the measured signal, while the personalized artifact baseline after parameter matching adjustment can more accurately follow the changing trend of motion artifact components in the measured signal. This verifies the effectiveness of parameter matching strategies such as time axis scaling based on velocity scaling factor, amplitude scaling based on pressure scaling factor, and superimposing linear drift components.

[0028] In one specific embodiment, after obtaining the dynamic artifact baseline, the method further includes: The real-time measured impedance signal is differentially calculated with the dynamic artifact baseline to obtain the net physiological impedance signal after eliminating motion artifacts. The net physiological impedance signal is decomposed into a multi-scale wavelet decomposition. The wavelet coefficients corresponding to the heart rate frequency band are extracted and reconstructed to obtain the heart rate analysis signal. The wavelet coefficients corresponding to the low frequency band are extracted and reconstructed to obtain the blood oxygen analysis signal. Peak detection is performed on the heart rate analysis signal, and the number of peaks is counted to calculate the heart rate value. The baseline mean of the blood oxygen analysis signal is calculated and substituted into the blood oxygen concentration conversion formula to obtain the blood oxygen saturation.

[0029] Specifically, the difference between the real-time measured impedance signal and the dynamic artifact baseline is achieved through point-by-point subtraction. That is, the measured impedance value at each moment is subtracted from the corresponding dynamic artifact baseline value. The motion artifact component in the net physiological impedance signal obtained after difference is effectively eliminated, preserving the physiological impedance fluctuations caused by heartbeats and changes in blood oxygenation. Multi-scale wavelet decomposition uses discrete wavelet transform to decompose the net physiological impedance signal into wavelet coefficients in multiple frequency bands. The decomposition level is set to five levels. The frequency band corresponding to the detail coefficients of the first to third levels covers 0.5 Hz to 4 Hz, which includes the impedance pulse wave signal generated by heartbeats. The frequency band corresponding to the approximation coefficients of the fourth to fifth levels covers 0 Hz to 0.5 Hz, which includes low-frequency baseline fluctuations caused by respiratory movements and changes in blood oxygen saturation. The heart rate analysis signal is obtained by extracting the wavelet coefficients of the first to third levels and reconstructing them through inverse wavelet transform. This signal highlights the periodic impedance changes related to heart rate. The blood oxygenation analysis signal is obtained by extracting the wavelet coefficients from the fourth to fifth layers and reconstructing them by inverse wavelet transform. This signal highlights the slow baseline drift characteristics associated with blood oxygenation.

[0030] Peak detection of the heart rate analysis signal employs an adaptive threshold method. First, the mean and standard deviation of the signal are calculated. The threshold is set to the mean plus 1.5 times the standard deviation. When the signal value exceeds this threshold and adjacent sampling points show a trend of first rising and then falling, it is determined to be a peak. The number of peaks detected within a consecutive 60-second time window is the heart rate, and the heart rate value is equal to the heart rate converted to heart rate per minute. The baseline mean of the blood oxygen analysis signal is obtained by calculating the arithmetic mean of the signal values ​​within a 30-second sliding window. This mean reflects the average impedance baseline level of the current period, and the baseline level is negatively correlated with the concentration of oxyhemoglobin in the blood. The blood oxygen concentration conversion formula is: blood oxygen saturation = 98 minus the calibration coefficient multiplied by the difference between the baseline mean and the reference baseline value. The calibration coefficient is determined by linear regression of the standard pulse oximeter measurement value and the system's measured baseline value, and is set to 0.15. The reference baseline value is the standard baseline impedance of a healthy adult, taken as 2500 ohms. Substituting the current baseline mean into the formula yields the blood oxygen saturation percentage.

[0031] Figure 3 This is a schematic diagram of the detection of the peak value of the net physiological impedance signal in an embodiment of this application; Figure 3 This diagram illustrates the process of peak detection to extract heart rate parameters from net physiological impedance signals. The solid line in the figure represents the net physiological impedance signal obtained after motion artifact elimination through differential operations, while the dashed line represents the adaptive peak detection threshold, calculated based on the signal mean plus 1.5 times the standard deviation. The "peak" positions marked in the figure represent the detected impedance pulse wave peaks. A valid peak is defined as a signal value exceeding the detection threshold with adjacent sampling points showing an initial increase followed by a decrease. The heart rate value can be calculated by counting the number of peaks detected within a statistical time window. The time interval between adjacent peaks is the RR interval of the cardiac cycle. This diagram verifies the effectiveness of multi-scale wavelet decomposition for extracting heart rate frequency band signals and the adaptive threshold peak detection algorithm, demonstrating its ability to accurately identify the periodic impedance changes caused by cardiac pulsation.

[0032] In one specific embodiment, step S2 includes: Before the user uses it for the first time, start the personalized calibration program, guide the user to slide their finger back and forth in the center area of ​​the touchpad at a constant speed multiple times in non-monitoring mode, collect personalized sliding artifact signals, guide the user to click once at each of the multiple preset positions on the touchpad with the target pressure, collect personalized click artifact signals, and guide the user to keep their finger still in the center of the touchpad, collect personalized stillness artifact signals. The impedance mean and standard deviation are extracted from the personalized sliding artifact signal, and the general template impedance mean and standard deviation corresponding to the sliding operation are extracted from the pure motion artifact database. The amplitude deviation coefficient and fluctuation deviation coefficient of the sliding operation are calculated. The impedance pulse peak value and pulse width are extracted from the personalized click artifact signal, and the general template pulse peak value and pulse width corresponding to the click operation are extracted from the pure motion artifact database. The peak deviation coefficient and width deviation coefficient of the click operation are calculated. The impedance baseline value and baseline drift slope are calculated from the personalized still artifact signal. The general template baseline value and drift slope corresponding to the still operation are extracted from the pure motion artifact database. The baseline deviation coefficient and drift deviation coefficient of the still operation are calculated. The baseline deviation coefficient and drift deviation coefficient are organized into a deviation coefficient matrix.

[0033] Specifically, the personalized calibration program is automatically triggered when the user first activates the physiological monitoring function. The interface guides the user through visual or audio prompts to perform a standardized sequence of operations. The constant-speed gliding requires the user to maintain a uniform speed throughout the gliding process. The constant speed is set to 30 mm / s, and the interface displays the current speed value in real time to help the user maintain a uniform speed. Three round trips are set to collect a sufficient length of signal samples. The collected personalized gliding artifact signals reflect the influence of the user's current finger physiological structure on the gliding artifact pattern. The target pressure is set to 300 g / f. The interface provides real-time feedback on the current pressure value through a pressure indicator bar, guiding the user to adjust the click force to approach the target value. Multiple preset positions are available, including the upper left, upper right, lower left, lower right, and center positions of the touchpad. The collected personalized click artifact signals include the impedance pulse characteristics caused by the click action. The stationary hovering requires the user to lightly touch the touchpad surface with their finger, maintaining a contact pressure of 150 g / f without any movement. The holding time is set to 10 seconds to observe the stability and drift trend of the impedance baseline. The collected personalized stationary artifact signals mainly reflect the impedance baseline value and the slow drift characteristics of the baseline over time.

[0034] All impedance samples within a 30 mm / s velocity range are extracted from the personalized sliding artifact signal. The arithmetic mean of these samples is calculated, and the standard deviation of the samples around the mean is calculated to obtain the impedance standard deviation. The medium-speed sliding range is located from the pure motion artifact database, and the general template impedance time series stored under this category is extracted. The arithmetic mean of all impedance values ​​in the template series is calculated, and the standard deviation of the template series is calculated to obtain the general template impedance standard deviation. The amplitude deviation coefficient of the sliding operation is equal to the impedance mean of the personalized sliding artifact signal divided by the general template impedance mean, and the fluctuation deviation coefficient is equal to the impedance standard deviation of the personalized sliding artifact signal divided by the general template impedance standard deviation.

[0035] The impedance pulse corresponding to each click is identified from five clicks in the personalized click artifact signal. The peak value of the impedance pulse is obtained by detecting the maximum impedance value within the pulse time period, and the pulse width is obtained by calculating the time span from the pulse start to the end. The impedance pulse peak value is obtained by calculating the arithmetic mean of the pulse peak values ​​of the five clicks, and the impedance pulse width is obtained by calculating the arithmetic mean of the pulse widths of the five clicks. A moderate click interval category is located from the pure motion artifact database, and the general template impedance pulse features stored under this category are extracted, including the general template pulse peak value and the general template pulse width. The peak deviation coefficient of the click operation is equal to the impedance pulse peak value of the personalized click artifact signal divided by the general template pulse peak value, and the width deviation coefficient is equal to the impedance pulse width of the personalized click artifact signal divided by the general template pulse width.

[0036] The impedance baseline value is obtained by calculating the arithmetic mean of all impedance samples within a 10-second time window of the personalized still artifact signal. The slope of the fitted line obtained by linear fitting of the impedance time series is the baseline drift slope, which reflects the trend of impedance baseline change over time. The still operation category with a pressure range of 150 g / f is located from the pure motion artifact database, and the general template baseline value and general template drift slope stored under this category are extracted. The baseline deviation coefficient of the still operation is equal to the impedance baseline value of the personalized still artifact signal divided by the general template baseline value, and the drift deviation coefficient is equal to the baseline drift slope of the personalized still artifact signal divided by the general template drift slope. The amplitude deviation coefficient and fluctuation deviation coefficient of the sliding operation are organized into the first row of the deviation coefficient matrix, the peak deviation coefficient and width deviation coefficient of the click operation are organized into the second row, and the baseline deviation coefficient and drift deviation coefficient of the still operation are organized into the third row, forming a three-row, three-column deviation coefficient matrix.

[0037] In one specific embodiment, step S4 includes: The system monitors the user's behavior patterns. When it detects that the user's finger leaves the monitoring electrode position for more than a set time and resumes normal touchpad operation, it collects the user's real-time operation artifact signal in non-monitoring mode as a new artifact sample. The new artifact samples are identified and classified into sliding segments, click segments, or stationary segments according to the operation type. Impedance feature parameters are extracted for each segment. The impedance feature parameters are compared with the general template features of the corresponding operation mode in the pure motion artifact database, and the instantaneous deviation coefficient is calculated. A sliding window weighted average strategy is used to weight and fuse the currently used deviation coefficient matrix with the instantaneous deviation coefficient matrix constructed based on the instantaneous deviation coefficients, resulting in an updated deviation coefficient matrix.

[0038] Specifically, the behavior monitoring mode is determined by detecting the contact state between the finger and the monitoring electrodes. When the contact resistance of all four monitoring electrodes rises simultaneously above a set threshold, it is determined that the finger has left the monitoring position. The set duration is 5 seconds. If the touchpad controller reports a new operation event after the finger has been away for more than 5 seconds, it is determined that the user has resumed normal touchpad operation. At this time, the system automatically switches to non-monitoring mode to collect real-time operation artifact signals as new artifact samples for 10 seconds. The obtained new artifact samples contain artifact characteristics of the user's current physiological state. Operation type recognition is achieved by analyzing the pressure change rate and velocity data in the new artifact samples. When the velocity data is not zero and the duration exceeds 1 second, the time period is classified as a sliding segment. When the pressure shows a pulse pattern of first rising and then falling, the time period is classified as a clicking segment. When the pressure standard deviation is less than 10 g / L and the duration exceeds 2 seconds, the time period is classified as a stationary segment. The mean and standard deviation of impedance are extracted as impedance characteristic parameters for sliding segments, the peak value and pulse width of impedance pulses are extracted as impedance characteristic parameters for clicking segments, and the baseline value and baseline drift slope of impedance pulses are extracted as impedance characteristic parameters for stationary segments.

[0039] The impedance characteristic parameters of the sliding segment are compared with the mean and standard deviation of the general template impedance corresponding to the sliding operation in the pure motion artifact database. The instantaneous amplitude deviation coefficient is calculated as the mean impedance of the sliding segment divided by the mean impedance of the general template, and the instantaneous fluctuation deviation coefficient is calculated as the standard deviation of the sliding segment impedance divided by the standard deviation of the general template impedance. Similarly, the instantaneous deviation coefficients of the clicked segment and the stationary segment are calculated, and these instantaneous deviation coefficients are organized into an instantaneous deviation coefficient matrix. The sliding window weighted averaging strategy uses a forgetting factor with a value of 0.8. Each element in the updated deviation coefficient matrix is ​​equal to 0.8 multiplied by the corresponding element of the currently used deviation coefficient matrix plus 0.2 multiplied by the corresponding element of the instantaneous deviation coefficient matrix. This weighted fusion method allows the system to gradually adapt to changes in the user's physiological state while retaining historical calibration information. When the user places their finger back in the monitoring electrode position to resume physiological monitoring, the system automatically switches to the updated deviation coefficient matrix for subsequent artifact baseline reconstruction.

[0040] The above describes the touchpad bioimpedance physiological monitoring method in the embodiments of this application. The following describes the touchpad bioimpedance physiological monitoring device in the embodiments of this application. One embodiment of the touchpad bioimpedance physiological monitoring device in the embodiments of this application includes: The acquisition module is used to acquire impedance change sequences corresponding to user sliding operations, clicking operations, and stationary hovering operations in non-monitoring mode. The impedance change sequences are classified according to operation mode based on moving speed and contact pressure. A pure motion artifact database is constructed from the impedance time series corresponding to each operation mode. The comparison module is used to guide users to execute standardized operation sequences, collect personalized artifact signals, and compare the amplitude and fluctuation of the personalized artifact signals with the general template signals of the corresponding operation modes in the pure motion artifact database to obtain a deviation coefficient matrix. The identification module is used to identify the current operating state in real time under physiological monitoring mode. Based on the current operating state, it queries a general artifact impedance template from the pure motion artifact database, and performs amplitude scaling and fluctuation adjustment on the deviation coefficients of the corresponding row in the deviation coefficient matrix and the general artifact impedance template to obtain a personalized artifact baseline. The update module is used to perform differential calculation between the measured impedance signal and the personalized artifact baseline to obtain the net physiological impedance signal. When it is detected that the user leaves the monitoring position and operates the touchpad again, new artifact samples are collected and the instantaneous deviation coefficient is calculated. The deviation coefficient matrix is ​​updated by a sliding window weighted average strategy.

[0041] This invention also provides a computer device, which may be a server. The computer device includes a processor, memory, display screen, input device, network interface, and database connected via a system bus. The processor in this computer design provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device is used to store the data corresponding to this embodiment. The network interface of the computer device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements the above-described method.

[0042] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of the touchpad bioimpedance physiological monitoring method.

[0043] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0044] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0045] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. 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 of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring bioimpedance physiology on a touchpad, characterized in that, The method includes: Step S1: In non-monitoring mode, collect the impedance change sequences corresponding to user sliding operation, clicking operation and stationary hovering operation. Classify the impedance change sequences according to the operation mode based on the moving speed and contact pressure. Construct a pure motion artifact database from the impedance time series corresponding to each operation mode. Step S2: Guide the user to execute a standardized operation sequence, collect personalized artifact signals, and compare the amplitude and fluctuation of the personalized artifact signals with the general template signals of the corresponding operation modes in the pure motion artifact database to obtain the deviation coefficient matrix; Step S3: In physiological monitoring mode, identify the current operation status in real time, query the general artifact impedance template from the pure motion artifact database according to the current operation status, and adjust the amplitude scaling and fluctuation of the deviation coefficients in the corresponding row of the deviation coefficient matrix with the general artifact impedance template to obtain a personalized artifact baseline. Step S4: Perform differential calculation between the measured impedance signal and the personalized artifact baseline to obtain the net physiological impedance signal. When the user leaves the monitoring position and re-operates the touchpad, new artifact samples are collected and the instantaneous deviation coefficient is calculated. The deviation coefficient matrix is ​​updated using a sliding window weighted average strategy.

2. The method for monitoring bioimpedance physiology of a touchpad according to claim 1, characterized in that, Step S1 includes: Multiple metal electrodes are set in a preset area on the touch panel surface as a bioimpedance monitoring device. A sinusoidal AC excitation current is applied to the excitation electrode pair through a constant current source. The voltage signal between the measurement electrode pair is collected and the real-time impedance value is calculated. In non-monitoring mode, the user is guided to place their finger on any position on the touchpad to perform routine operations. The impedance signal, the contact pressure data output by the touchpad's built-in pressure sensor, and the finger movement speed data output by the touchpad controller are collected simultaneously. The three types of data are aligned by timestamp to form a triplet sequence. The sliding operation data in the triplet sequence is divided into speed ranges based on the moving speed, the click operation data is divided into pressure ranges based on the peak contact pressure, and the pressure stability range is extracted from the static hovering data based on the pressure standard deviation. The impedance time series, corresponding pressure series, velocity series, and operation mode labels collected under each subdivided operation mode are stored separately to construct the pure motion artifact database.

3. The method for monitoring bioimpedance physiology of a touchpad according to claim 1, characterized in that, Step S3 includes: The impedance signal, pressure signal and operation event are jointly analyzed by a sliding window of set duration. The first-order difference of pressure within the window is extracted as the pressure change rate feature. The static state is determined by the absolute value of the pressure change rate of continuous sampling points, the click state is determined by the monotonic change pattern of pressure value, and the sliding state is determined by the periodic fluctuation of mouse movement event and pressure change rate. The current operation state label is obtained. Based on the current operation status label, extract the corresponding general artifact impedance time series template from the pure motion artifact database, and extract the amplitude-type deviation coefficient and fluctuation-type deviation coefficient of the corresponding row from the deviation coefficient matrix; The amplitude scaling and fluctuation adjustment of the general artifact impedance time series template are performed to obtain the personalized artifact baseline.

4. The method for monitoring bioimpedance physiology of a touchpad according to claim 3, characterized in that, After performing amplitude scaling and fluctuation adjustment on the general artifact impedance time series template, the method further includes: The average pressure change rate and average velocity within the sliding window are extracted as real-time feature parameters, and the standard pressure change rate and standard velocity are extracted from the pure motion artifact database as template feature parameters. Based on the real-time feature parameters and the template feature parameters, a parameter scaling factor is calculated. For sliding operations, the time axis of the personalized artifact baseline is scaled. For clicking operations, the amplitude of the personalized artifact baseline is scaled. For stationary operations, a linear drift component is superimposed on the personalized artifact baseline to obtain a dynamic artifact baseline.

5. The method for monitoring bioimpedance physiology of a touchpad according to claim 4, characterized in that, After obtaining the dynamic artifact baseline, the following is also included: The real-time measured impedance signal is differentially calculated with the dynamic artifact baseline to obtain the net physiological impedance signal after eliminating motion artifacts. The net physiological impedance signal is decomposed into a multi-scale wavelet decomposition. Wavelet coefficients corresponding to the heart rate frequency band are extracted and reconstructed to obtain the heart rate analysis signal. Wavelet coefficients corresponding to the low frequency band are extracted and reconstructed to obtain the blood oxygen analysis signal. Peak detection is performed on the heart rate analysis signal, and the number of peaks is counted to calculate the heart rate value. The baseline mean of the blood oxygen analysis signal is calculated and substituted into the blood oxygen concentration conversion formula to obtain the blood oxygen saturation.

6. The method for monitoring bioimpedance physiology of a touchpad according to claim 1, characterized in that, Step S2 includes: Before the user uses it for the first time, start the personalized calibration program, guide the user to slide their finger back and forth in the center area of ​​the touchpad at a constant speed multiple times in non-monitoring mode, collect personalized sliding artifact signals, guide the user to click once at each of the multiple preset positions on the touchpad with the target pressure, collect personalized click artifact signals, and guide the user to keep their finger still in the center of the touchpad, collect personalized stillness artifact signals. The impedance mean and standard deviation are extracted from the personalized sliding artifact signal, and the general template impedance mean and standard deviation corresponding to the sliding operation are extracted from the pure motion artifact database. The amplitude deviation coefficient and fluctuation deviation coefficient of the sliding operation are calculated. The impedance pulse peak value and pulse width are extracted from the personalized click artifact signal, and the general template pulse peak value and pulse width corresponding to the click operation are extracted from the pure motion artifact database. The peak deviation coefficient and width deviation coefficient of the click operation are calculated. The impedance baseline value and baseline drift slope are calculated from the personalized still artifact signal. The general template baseline value and drift slope corresponding to the still operation are extracted from the pure motion artifact database. The baseline deviation coefficient and drift deviation coefficient of the still operation are calculated. The baseline deviation coefficient and drift deviation coefficient are organized into the deviation coefficient matrix.

7. The method for monitoring bioimpedance physiology of a touchpad according to claim 1, characterized in that, Step S4 includes: The system monitors the user's behavior patterns. When it detects that the user's finger leaves the monitoring electrode position for more than a set time and resumes normal touchpad operation, it collects the user's real-time operation artifact signal in non-monitoring mode as a new artifact sample. The new artifact samples are identified and divided into sliding segments, click segments, or stationary segments according to the operation type. Impedance feature parameters are extracted from each segment. The impedance feature parameters are compared with the general template features of the corresponding operation mode in the pure motion artifact database, and the instantaneous deviation coefficient is calculated. A sliding window weighted average strategy is used to weight and fuse the currently used deviation coefficient matrix with the instantaneous deviation coefficient matrix constructed based on the instantaneous deviation coefficients to obtain the updated deviation coefficient matrix.

8. A touchpad bioimpedance physiological monitoring device, characterized in that, For implementing the touchpad bioimpedance physiological monitoring method as described in any one of claims 1-7, the touchpad bioimpedance physiological monitoring device comprises: The acquisition module is used to acquire impedance change sequences corresponding to user sliding operations, clicking operations, and stationary hovering operations in non-monitoring mode. The impedance change sequences are classified according to operation mode based on moving speed and contact pressure. A pure motion artifact database is constructed from the impedance time series corresponding to each operation mode. The comparison module is used to guide users to execute standardized operation sequences, collect personalized artifact signals, and compare the amplitude and fluctuation of the personalized artifact signals with the general template signals of the corresponding operation modes in the pure motion artifact database to obtain a deviation coefficient matrix. The identification module is used to identify the current operating state in real time under physiological monitoring mode. Based on the current operating state, it queries a general artifact impedance template from the pure motion artifact database, and performs amplitude scaling and fluctuation adjustment on the deviation coefficients of the corresponding row in the deviation coefficient matrix and the general artifact impedance template to obtain a personalized artifact baseline. The update module is used to perform differential calculation between the measured impedance signal and the personalized artifact baseline to obtain the net physiological impedance signal. When it is detected that the user leaves the monitoring position and operates the touchpad again, new artifact samples are collected and the instantaneous deviation coefficient is calculated. The deviation coefficient matrix is ​​updated by a sliding window weighted average strategy.

9. A computer device, characterized in that, The device includes a memory and a processor, the memory storing a computer program that can run on the processor, and the processor executing the computer program to implement the touchpad bioimpedance physiological monitoring method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is run by the processor, it causes the processor to perform the touchpad bioimpedance physiological monitoring method as described in any one of claims 1 to 7.

Images