A method for evaluating acupuncture stimulation based on real-time monitoring of multiple physiological signals fusion

Abstract

The present invention discloses an acupuncture stimulation evaluation method based on real-time monitoring of multiple physiological signals fusion, which relates to the field of acupuncture medical technology. Through the correlation analysis of the absolute value of the myoelectric attenuation slope and the standard deviation of the skin temperature phase difference, the "phase lock state" is accurately determined, breaking through the limitation of a single signal dimension, and the determination accuracy is improved to more than 90%; a linear prediction model is constructed based on clinical data, combined with the dynamic correction of the compensation coefficient when the β wave proportion exceeds the standard, to achieve personalized needle retention time prediction, effectively reducing the prediction deviation rate; when the cumulative phase lock time reaches the predicted value, the needle retention adjustment instruction is automatically triggered to form a "monitoring-prediction-execution" closed loop, reducing the ineffective treatment time by about 15%, and improving the efficiency of acupuncture operation; through the real-time correction of model parameters by β wave abnormality monitoring, the efficacy deviation caused by muscle tension or nerve imbalance is suppressed, the system false alarm rate is reduced to below 3%, and the robustness is significantly better than traditional methods.

Description

Technical Field

[0001] The present invention relates to the field of acupuncture medical technology, and in particular to an acupuncture stimulation evaluation method based on real-time monitoring of multiple physiological signals fusion. Background Art

[0002] As a traditional Chinese medicine treatment, acupuncture's efficacy assessment has long relied on the physician's subjective experience (such as the sensation of "getting qi") and patient complaints, lacking objective quantitative indicators. With the development of biosensor technology, existing technologies have begun to use single physiological signals (such as electromyography, skin temperature, or heart rate variability) to monitor efficacy. However, these methods have significant drawbacks:

[0003] The one-sidedness of single-signal analysis: EMG signals are easily affected by motion artifacts, and skin temperature signals have a delayed response to changes in local microcirculation. A single signal cannot fully reflect the complex physiological responses induced by acupuncture.

[0004] Inadequate quantitative assessment models: Existing methods often use fixed thresholds or empirical formulas to predict analgesia duration, failing to consider individual physiological differences and dynamic interference factors, leading to significant prediction bias.

[0005] Lack of ability to predict duration of effect: Traditional methods can only assess the immediate intensity of stimulation and cannot predict the duration of the acupuncture effect. As a result, the needle retention time is set based on empirical formulas (such as a fixed 30 minutes), which can easily lead to insufficient efficacy or excessive stimulation. Summary of the Invention

[0006] In view of the shortcomings of the existing technology, the present invention provides an acupuncture stimulation evaluation method based on real-time monitoring of multiple physiological signals fusion.

[0007] In order to achieve the above object, the technical solution of the present invention is as follows:

[0008] A method for evaluating acupuncture stimulation based on real-time monitoring of multiple physiological signals fusion includes the following steps:

[0009] Synchronously collect the myoelectric signal and skin temperature signal of the target area, wherein the sampling frequency of the myoelectric signal is a predetermined number of bits, and the sampling frequency of the skin temperature signal is a preset value;

[0010] Performing time domain feature extraction on the electromyographic signal and calculating the attenuation slope of the root mean square value within a preset time window, wherein a negative value of the attenuation slope represents a muscle relaxation rate;

[0011] Perform frequency domain feature extraction on the skin temperature signal, obtain 0.08-0.12 Hz frequency band oscillation signal through a bandpass filter, and calculate the standard deviation of the phase difference between adjacent peaks;

[0012] Associating the absolute value of the attenuation slope with the change direction of the phase difference standard deviation, and determining that a phase lock state has been entered when the absolute value of the attenuation slope continues to increase and the phase difference standard deviation is less than a preset threshold;

[0013] A phase-locked state duration prediction model is constructed based on historical data, and the duration prediction model satisfies:

[0014] Duration of effect = baseline value + compensation coefficient × (absolute value of current attenuation slope - preset baseline value), where the baseline value is a configurable constant and the compensation coefficient is calibrated to a constant within the range of 0.4-0.6 based on clinical trial data;

[0015] When the accumulated phase locking time reaches a preset proportion of the effect duration, a needle retention time adjustment instruction is generated and a reminder signal is output;

[0016] In the phase-locked state, the proportion of β wave components in the frequency range of 13-30 Hz in the power spectrum of the electromyographic signal is detected at every preset period. When the proportion of β wave exceeds the preset warning value, the compensation coefficient is updated. The update formula is:

[0017] Updated compensation coefficient = original compensation coefficient × (1-beta wave proportion exceeding the standard range / preset tolerance threshold);

[0018] The exceeding standard range is the absolute value of the difference between the current β wave ratio and the warning value.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] The time-domain attenuation slope of the electromyographic signal is dynamically correlated with the standard deviation of the frequency-domain phase difference of the skin temperature signal. The phase lock state is determined by the consistency of the change direction of the two, breaking through the limitations of single-signal analysis.

[0021] A linear prediction model was constructed based on clinical trial data, and the excess amplitude of the beta wave ratio was introduced to perform real-time correction of the compensation coefficient, significantly improving the model's adaptability to individual differences and enhancing prediction robustness.

[0022] By comparing the accumulated phase-locked time with the predicted value, the needle retention time adjustment instruction is triggered, forming a closed-loop control logic of "monitoring-prediction-correction-execution" to achieve dynamic optimization of acupuncture treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The disclosure of the present invention is described with reference to the accompanying drawings. It should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of protection of the present invention. In the drawings, the same reference numerals are used to refer to the same components. Among them:

[0024] Figure 1 A diagram showing the steps of the method of the present invention;

[0025] Figure 2 It is a workflow diagram of the present invention;

[0026] Figure 3 FIG. 1 is a diagram showing the adjustment process of the bandpass filter of the present invention;

[0027] Figure 4 A flow chart for updating the preset correction threshold value of the present invention;

[0028] Figure 5 This is a flow chart of feedback adjustment of the effect duration of the present invention. DETAILED DESCRIPTION

[0029] It is easy to understand that according to the technical solution of the present invention, without changing the essential spirit of the present invention, a person skilled in the art can propose a variety of interchangeable structural modes and implementation modes. Therefore, the following specific embodiments and drawings are only exemplary descriptions of the technical solution of the present invention and should not be regarded as the entire invention or as a limitation or restriction of the technical solution of the present invention.

[0030] Application Overview:

[0031] In traditional acupuncture efficacy evaluation systems, single physiological signal monitoring methods are unable to eliminate the inherent defects of motion artifacts and response lags. Separate analysis of time-frequency domain features limits the accuracy of physiological response pattern recognition, and fixed needle retention time models are difficult to adapt to individualized treatment needs. For example, in an acupuncture treatment room, when using a single electromyographic signal monitoring device, motion artifacts generated by the patient's limb micro-movements will significantly interfere with the calculation of the time-domain root mean square value, and the skin temperature signal acquisition device causes the frequency-domain phase difference standard deviation detection to lag behind actual physiological changes due to heat conduction delays. The system only uses the time-domain electromyographic attenuation slope or the frequency-domain skin temperature oscillation signal for separate analysis, which cannot capture the dynamic coupling relationship between muscle relaxation rate and microcirculatory regulation, resulting in misjudgment of the phase lock state. The treatment terminal relies on a fixed 30-minute needle retention time and fails to adjust the stimulation parameters according to real-time physiological characteristics, resulting in premature treatment interruption or overstimulation in some patients. If the above problems are not solved, the combined effects of motion artifacts and signal lag will reduce the effectiveness of multimodal physiological signal fusion, the time-frequency domain feature segmentation analysis will produce false physiological state identification results, and the needle retention time model that lacks dynamic prediction capabilities will lead to a mismatch between treatment parameters and patients' real-time responses, ultimately causing deviations in efficacy evaluation and increased treatment risks.

[0032] Faced with the above problems, the present applicant first realized that single physiological signal monitoring has inherent defects such as weak anti-interference ability and response delay, and the fragmented analysis of time-frequency domain features makes it difficult to capture the dynamic changes in physiological states. To this end, the present applicant attempts to eliminate the limitations of single signals through a multi-signal fusion mechanism, while establishing a collaborative analysis framework for time-frequency features to improve state recognition accuracy. Specifically, three solutions were explored: first, using dual-signal complementary acquisition, leveraging the rapid response characteristics of the electromyographic signal and the steady-state characteristics of the skin temperature signal to verify each other; second, constructing a joint criterion for the time-domain attenuation slope and the frequency-domain phase difference standard deviation to address the problem that single-dimensional features are susceptible to noise interference; and third, developing a dynamic prediction model based on real-time physiological characteristics to replace the traditional fixed-duration empirical formula. After clinical data verification, it was found that the electromyographic time-domain attenuation slope and the skin temperature frequency-domain phase difference standard deviation have an inverse correlation pattern. The joint judgment of the two can effectively identify physiological state transition points, and substituting the change trends of the two into a linear regression model can accurately predict the duration of the effect. Based on this, the dual-signal time-frequency feature fusion was finally determined as the core solution.

[0033] like Figure 1 As shown, in this regard, the present application proposes an acupuncture stimulation evaluation method based on multi-physiological signal fusion real-time monitoring, comprising the following steps:

[0034] Synchronously collect the electromyographic signals and skin temperature signals of the target area, where the sampling frequency of the electromyographic signals is a predetermined number of bits, and the sampling frequency of the skin temperature signals is a preset value; synchronously collecting the electromyographic signals and skin temperature signals of the target area means simultaneously acquiring two physiological signals, muscle electrical activity and skin temperature, which can be specifically achieved by using multi-channel biosensors and timestamp alignment technology, and ensuring the time consistency of the two signals through hardware synchronization or software time synchronization.

[0035] Time domain feature extraction is performed on the electromyographic signal, and the attenuation slope of the root mean square value within the preset time window is calculated. A negative attenuation slope represents the muscle relaxation rate. Calculating the attenuation slope of the root mean square value within the preset time window refers to performing sliding window processing on the electromyographic signal and calculating the downward trend of the root mean square value. Specifically, this can be achieved using a sliding window linear regression algorithm, and the muscle relaxation rate is quantified by fitting the slope of the data points within the window.

[0036] Frequency domain feature extraction is performed on the skin temperature signal. The oscillation signal in the 0.08-0.12 Hz frequency band is obtained through a band-pass filter, and the standard deviation of the phase difference between adjacent peaks is calculated. Calculating the standard deviation of the phase difference between adjacent peaks refers to extracting the time interval volatility of the oscillation period after filtering the skin temperature signal. Specifically, this can be achieved by using an extreme point detection algorithm combined with standard deviation calculation, and the stability of the autonomic nervous system regulation state is evaluated by the degree of phase difference dispersion.

[0037] The absolute value of the associated attenuation slope and the direction of change of the phase difference standard deviation are correlated. When the absolute value of the attenuation slope continues to increase and the phase difference standard deviation is less than the preset threshold, it is determined that the phase lock state has been entered; the absolute value of the associated attenuation slope and the direction of change of the phase difference standard deviation refer to the dynamic correlation analysis of the muscle relaxation rate and the synchronization of temperature fluctuations. Specifically, this can be achieved by using threshold comparison and trend judgment algorithms, and the critical points of the physiological state can be identified through the coordinated changes of two-way indicators.

[0038] A phase-locked state duration prediction model is constructed based on historical data. The duration prediction model satisfies:

[0039] Duration of effect = baseline value + compensation coefficient × (absolute value of current attenuation slope - preset baseline value), where the baseline value is a configurable constant and the compensation coefficient is calibrated to a constant within the range of 0.4-0.6 based on clinical trial data;

[0040] The duration prediction model refers to the estimation of the duration of analgesia by establishing a linear regression equation based on historical data. Specifically, it can be achieved by fitting the baseline value and compensation coefficient using the least squares method, and the prediction accuracy can be improved by calibrating the model parameters through clinical data.

[0041] When the cumulative phase-locked time reaches a preset proportion of the effect duration, a needle retention time adjustment instruction is generated and a reminder signal is output; generating a needle retention time adjustment instruction means triggering dynamic adjustment based on the deviation between the predicted time and the actual treatment progress, which can be specifically implemented using a proportional threshold comparison and signal output module, and optimizing the acupuncture stimulation duration through a real-time feedback mechanism.

[0042] In the phase-locked state, the proportion of β wave components in the frequency range of 13-30 Hz in the power spectrum of the electromyographic signal is detected at every preset period. When the proportion of β wave exceeds the preset warning value, the compensation coefficient is updated. The update formula is:

[0043] Updated compensation coefficient = original compensation coefficient × (1-beta wave proportion exceeding the standard range / preset tolerance threshold);

[0044] The excess magnitude is the absolute difference between the current beta wave ratio and the warning value. Updating the compensation coefficient dynamically modifies the prediction model parameters based on the excess magnitude of the beta wave ratio. This can be achieved using a difference calculation and scaling algorithm, adjusting the model to accommodate individual differences through frequency domain feature feedback.

[0045] The core innovation of this application lies in constructing a phase-locked state recognition mechanism for dual-signal fusion through dynamic correlation analysis of the time-domain attenuation slope of the electromyographic signal and the standard deviation of the frequency-domain phase difference of the skin temperature signal, and establishing an online updateable effect duration prediction model based on historical data. The model parameters are dynamically adjusted in combination with the feedback of the β-wave proportion to achieve real-time quantitative evaluation of the acupuncture stimulation effect and closed-loop control of the needle retention time.

[0046] like Figure 2 The figure below is a workflow diagram for this application. The working process and principle of this application are as follows: first, the EMG signal and skin temperature signal of the target area are synchronously acquired at different sampling frequencies. Time domain features are extracted from the EMG signal, and the attenuation slope of the root mean square value within a preset time window is calculated. A negative attenuation slope represents the rate of muscle relaxation. Frequency domain features are extracted from the skin temperature signal, and an oscillation signal in the 0.08-0.12 Hz frequency band is obtained through a bandpass filter. The standard deviation of the phase difference between adjacent peaks is calculated.

[0047] The absolute value of the attenuation slope is correlated with the direction of change of the phase difference standard deviation. When the absolute value of the attenuation slope continues to increase and the phase difference standard deviation is less than a preset threshold, the phase lock state is determined. A phase lock duration prediction model is constructed based on historical data. The model satisfies the requirement that the effect duration is equal to the baseline value plus the compensation coefficient multiplied by the difference between the current absolute value of the attenuation slope and the preset baseline value. The baseline value is a configurable constant, and the compensation coefficient is calibrated to a constant in the range of 0.4-0.6 based on clinical trial data.

[0048] When the cumulative phase-locked time reaches a preset proportion of the effect duration, a needle retention time adjustment instruction is generated and a reminder signal is output. While in the phase-locked state, the proportion of beta waves in the 13-30 Hz frequency band in the electromyographic signal power spectrum is detected at preset intervals. When the beta wave proportion exceeds the preset warning value, the compensation coefficient is updated. The update formula is the original compensation coefficient multiplied by 1 minus the ratio of the excess beta wave proportion to the preset tolerance threshold, where the excess is the absolute value of the difference between the current beta wave proportion and the warning value.

[0049] This multi-signal fusion approach eliminates the limitations of a single signal, and the collaborative analysis of time-frequency features improves state recognition accuracy. The dynamic prediction model replaces the traditional fixed-duration empirical formula, enabling personalized treatment based on real-time physiological characteristics.

[0050] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0051] The acquisition device simultaneously collects electromyographic (EMG) and skin temperature signals from acupuncture point areas. The sampling frequency for the EMG signal is set to 1000 Hz, and the sampling frequency for the skin temperature signal is set to 10 Hz. The root mean square (RMS) value of the collected EMG signal is calculated using a 500 ms time window. A sliding window linear regression algorithm is used to fit the attenuation curve, and the current attenuation slope is output. The skin temperature signal is filtered using a 0.08-0.12 Hz bandpass filter. Local maxima in the filtered signal are detected, and the standard deviation of the phase difference between adjacent maxima is calculated.

[0052] The phase lock status flag is triggered when the absolute value of the attenuation slope increases for three consecutive sampling cycles and the phase difference standard deviation is less than 0.2 radians for two consecutive sampling cycles. A duration prediction model was constructed using the least squares method to fit a linear regression equation based on data collected from 200 clinical samples. The baseline value in the model was set to 15 minutes, and the initial value of the compensation coefficient was set to 0.5.

[0053] In the phase-locked state, the percentage of myoelectric signal energy in the 13-30Hz frequency band relative to the total frequency band is measured every 30 seconds. If the beta wave percentage increases by more than 5% over three consecutive sampling cycles, it is considered a valid excess, and the compensation coefficient is updated according to the formula. When the cumulative phase-locked time reaches 80% of the predicted effect duration, the system generates a needle retention adjustment instruction and issues a reminder to the physician.

[0054] Through the above scheme, the present application realizes the real-time monitoring of the fusion of multiple physiological signals and improves the accuracy of acupuncture stimulation evaluation. The collaborative analysis of electromyographic signals and skin temperature signals eliminates the defect that a single signal is susceptible to interference, and the joint judgment of time-frequency domain features improves the accuracy of physiological state recognition. The dynamic prediction model based on real-time physiological characteristics replaces the traditional fixed needle retention time mode and realizes personalized adjustment of treatment parameters. The beta wave proportion monitoring mechanism further optimizes the model parameters, making the prediction results more in line with the actual response of the patient. This method provides a new technical means for the objective evaluation of the efficacy of acupuncture, which helps to improve the accuracy and effectiveness of acupuncture treatment.

[0055] In some of the above-mentioned schemes of the present application, when detecting the proportion of β wave components in the power spectrum of the electromyographic signal, the fixed cutoff frequency range of the bandpass filter may cause inaccurate extraction of skin temperature signal features, and cannot adapt to the frequency band offset caused by the dynamic changes of β waves in real time, thereby affecting the reliability of phase lock state determination.

[0056] like Figure 3 FIG. 1 is a diagram showing the adjustment process of the bandpass filter. The present application further proposes that, after the beta wave component ratio is detected, the cutoff frequency range of the bandpass filter is synchronously adjusted so that its center frequency dynamically shifts within the range of 0.1 Hz ± 5% as the beta wave ratio increases. Specifically,

[0057] When the beta wave ratio exceeds the preset warning value for two consecutive cycles:

[0058] The lower limit frequency of the band-pass filter was increased to 0.08 Hz × (1 + the amplitude of the excess β wave proportion / 10), and the upper limit frequency was reduced to 0.12 Hz × (1 - the amplitude of the excess β wave proportion / 15).

[0059] The bandpass filter's cutoff frequency adjustment is positively correlated with the extent to which the beta wave ratio exceeds the specified limit. The increase in the lower frequency limit is calculated by dividing the excess by 10, while the decrease in the upper frequency limit is calculated by dividing the excess by 15. As the beta wave ratio exceeds the specified limit, the lower frequency limit adjusts faster than the upper frequency limit, gradually narrowing the passband and shifting the center frequency toward around 0.1 Hz. The difference in the denominator values of the frequency adjustment parameters ensures asymmetric adjustment of the upper and lower frequency limits, preventing excessive shift in the passband range.

[0060] Specifically, when the beta wave ratio exceeds the warning value for two consecutive sampling periods, the current excess amplitude is calculated as the absolute difference between the current beta wave ratio and the warning value. This excess amplitude is substituted into the lower frequency adjustment formula to obtain a new lower frequency limit. For example, if the excess amplitude is 5%, the lower frequency limit is adjusted to 0.08Hz × 1.05 = 0.084Hz. Simultaneously, the excess amplitude is substituted into the upper frequency adjustment formula. For example, if the excess amplitude is 5%, the upper frequency limit is adjusted to 0.12Hz × 0.9667 = 0.116Hz. The adjusted passband range changes from the original 0.08-0.12Hz to 0.084-0.116Hz, and the center frequency shifts from 0.1Hz to within 0.1Hz ± 0.8%. Through dynamic adjustment, the bandpass filter can match the frequency domain characteristics of the skin temperature signal caused by beta wave anomalies, eliminating signal components that would be missed due to a fixed passband, thereby improving the calculation accuracy of the phase difference standard deviation. The asymmetric design of the upper and lower limit parameters during the frequency adjustment process effectively suppresses high-frequency noise interference while maintaining the stability of the center frequency, ensuring the accuracy of subsequent phase lock state determination and compensation coefficient update.

[0061] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0062] After the beta wave proportion is detected, the bandpass filter's cutoff frequency range is adjusted synchronously, so that its center frequency dynamically shifts within the range of 0.1Hz±5% as the beta wave proportion increases. Specifically, if the beta wave proportion exceeds the preset warning value for two consecutive cycles, the bandpass filter's lower limit frequency is increased to 0.08Hz×(1+the amount of excess beta wave proportion / 10), and the upper limit frequency is reduced to 0.12Hz×(1-the amount of excess beta wave proportion / 15).

[0063] For example, assuming the preset alert level is 30%, the currently detected beta wave percentage is 35%, and the excess is 5%. In this case, the lower frequency limit of the bandpass filter is adjusted to 0.08Hz × (1 + 5% / 10) = 0.084Hz, and the upper frequency limit is adjusted to 0.12Hz × (1 - 5% / 15) = 0.116Hz. This dynamic adjustment brings the filter's center frequency closer to 0.1Hz, more accurately capturing the characteristic oscillations in the skin temperature signal.

[0064] Furthermore, if the beta wave proportion continues to rise to 40% in the next detection cycle, and the excess exceeds the standard by 10%, the lower limit frequency of the bandpass filter will be further adjusted to 0.088Hz and the upper limit frequency will be adjusted to 0.112Hz. This continuous dynamic adjustment ensures that the filter can always capture the most relevant frequency components.

[0065] Through the above technical solution, the present application realizes the adaptive adjustment of the bandpass filter parameters. As a result, the frequency range of the filter can be dynamically optimized according to the changes in the proportion of beta waves, thereby improving the accuracy of capturing characteristic oscillations in the skin temperature signal. Specifically, when the proportion of beta waves increases, by narrowing the bandwidth of the filter and bringing its center frequency closer to 0.1Hz, the skin temperature oscillation signal related to acupuncture stimulation can be more accurately extracted, thereby improving the accuracy of the assessment of the acupuncture effect. At the same time, this adaptive adjustment mechanism also enhances the system's ability to adapt to the physiological differences of different individuals, making the evaluation method more widely applicable.

[0066] In some of the above-mentioned solutions of the present application, during the compensation coefficient updating process, due to the deviation between the predicted effect duration and the measured analgesic effect, the model prediction accuracy decreases, which in turn affects the effectiveness of subsequent parameter adjustments.

[0067] like Figure 4 As shown, it is a flow chart for updating the preset correction threshold value; the present application further proposes to recalculate the deviation rate between the predicted effect duration and the measured analgesic effect after the compensation coefficient is updated. The calculation formula of the deviation rate is: deviation rate = |predicted time-measured time| / measured time.

[0068] When the deviation rate exceeds the preset correction threshold, the beta wave's preset warning value is dynamically adjusted upwards in proportion to the deviation rate, and the compensation coefficient is reset to its initial value. The formula for updating the preset warning value is: Updated preset warning value = Original preset warning value × (1 + Deviation rate / 2).

[0069] The deviation rate calculation quantifies the difference between the predicted and measured times, providing a basis for subsequent parameter adjustments. When the deviation rate exceeds the correction threshold, the preset beta wave warning value is dynamically adjusted upward based on the deviation rate ratio to adapt the warning value to the actual deviation. The compensation coefficient is reset to its initial value to eliminate the cumulative coefficient error caused by an excessive beta wave ratio. In the preset warning value update formula, the deviation rate divided by 2 is used to control the warning value adjustment range to avoid overcorrection.

[0070] Specifically, the deviation rate calculation reflects the absolute deviation between the predicted and measured times through absolute value calculation, and is normalized by dividing by the measured time to eliminate the influence of individual differences on the deviation assessment. When the deviation rate exceeds the correction threshold, the beta wave warning value is adjusted based on a linear proportional relationship, so that the warning value increases as the deviation rate increases. The compensation coefficient reset operation interrupts the original update process to prevent the incorrect coefficient from affecting subsequent predictions. By simultaneously adjusting the warning value and resetting the compensation coefficient, adaptive correction of model parameters is achieved, ensuring the dynamic accuracy of the prediction model.

[0071] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0072] After the compensation coefficient is updated, the deviation rate between the predicted effect duration and the measured analgesic effect is recalculated. The calculation formula for the deviation rate is:

[0073] Deviation rate = |predicted time - measured time| / measured time;

[0074] When the deviation rate exceeds the preset correction threshold, the preset warning value of the beta wave is dynamically increased according to the deviation rate ratio, and the compensation coefficient is reset to the initial value. The preset warning value update formula is:

[0075] Updated preset warning value = original preset warning value × (1 + deviation rate / 2);

[0076] For example, if the original preset warning value is 0.3, the measured analgesic effect lasts for 45 minutes, while the predicted effect lasts for 60 minutes. The calculated deviation rate is:

[0077] |60-45| / 45 = 0.33;

[0078] If the preset correction threshold is 0.3, then the deviation rate of 0.33 exceeds the threshold. At this time, the preset warning value is updated:

[0079] Updated default warning value = 0.3×(1+0.33 / 2) = 0.3495;

[0080] At the same time, the compensation coefficient is reset to its initial value of 0.5. Through this dynamic adjustment mechanism, the system can continuously optimize the accuracy of the prediction model.

[0081] Through the above technical solution, this application achieves adaptive optimization of the prediction model. By comparing the deviation between the predicted time and the measured time, the beta wave warning value and compensation coefficient are dynamically adjusted, allowing the model to continuously self-correct according to the actual results. This closed-loop feedback mechanism improves the accuracy and stability of the prediction, enabling the system to adapt to the individual differences of different patients and the dynamic changes during the treatment process. At the same time, resetting the compensation coefficient to the initial value avoids the cumulative deviation of the parameters and ensures reliability during long-term use.

[0082] In some of the above-mentioned schemes of this application, when time domain features are extracted from electromyographic signals, the original signal may retain power frequency interference components, resulting in errors in the calculation of the root mean square value; at the same time, the fixed window segmentation method cannot adapt to the dynamic changes of the signal, affecting the fitting accuracy of the attenuation slope.

[0083] This application further proposes that time domain feature extraction of electromyographic signals includes:

[0084] After the EMG signal is filtered for power frequency interference, the signal sequence is segmented into preset window lengths, the root mean square value of each window is calculated, and the attenuation curve is fitted using a sliding window linear regression algorithm to output the current attenuation slope.

[0085] Among them, the power frequency interference filtering uses a 50Hz notch filter to eliminate AC interference; the preset window length is set to 200 milliseconds to match the physiological response cycle of muscle contraction; the root mean square value calculation uses the square root operation of the square sum of the sampling points in the window; the sliding window linear regression algorithm uses a sliding mechanism with a window overlap rate of 50%, and the slope is refitted after each slide.

[0086] Specifically, the power frequency interference filter first eliminates the 50Hz mains interference introduced during the signal acquisition process to ensure the accuracy of subsequent feature extraction. The filtered electromyographic signal is divided into multiple non-overlapping windows of 200 milliseconds in length. The root mean square value calculated in each window reflects the intensity changes of muscle electrical activity. When using sliding window linear regression, a scatter plot is constructed with the window sequence as the horizontal axis and the root mean square value as the vertical axis, and the slope value of the fitted line is calculated using the least squares method. When the window overlap rate is set to 50%, each new window retains the previous 100 milliseconds of historical data and adds 100 milliseconds of real-time data, improving the continuity of the slope fitting while ensuring computational efficiency. The root mean square value calculation error rate of this method is reduced by 12.7% compared to the traditional method, and the attenuation slope fitting speed is increased to output an updated result every 100 milliseconds.

[0087] As a preferred embodiment, the solution of the present application is specifically implemented as follows: After the electromyographic signal is subjected to power frequency interference filtering, the signal sequence is divided into 50ms window lengths, and the root mean square value of each window is calculated. A 200ms sliding window linear regression algorithm is used to fit the attenuation curve, and the current attenuation slope is output. A 50Hz notch filter is used for power frequency interference filtering, and the filter order is 4th order. When calculating the root mean square value, the signal is normalized to eliminate the influence of signal amplitude differences. Linear regression uses the least squares method, and the fitting equation is y=ax+b, where a is the attenuation slope.

[0088] Through the above technical solution, this application achieves precise extraction of the time-domain characteristics of electromyographic signals. Power frequency interference filtering eliminates the impact of power grid interference on signal quality. A sliding window linear regression algorithm overcomes the limitations of traditional fixed window analysis methods and improves the time resolution and accuracy of attenuation slope calculations. This provides a reliable data foundation for subsequent electromyographic signal-based assessments of acupuncture stimulation efficacy.

[0089] In some of the above-mentioned schemes of the present application, there is a problem of insufficient accuracy in identifying the dynamic oscillation pattern of the skin temperature signal during the frequency domain feature extraction process. Specifically, the traditional bandpass filtering operation fails to effectively separate the oscillation components of the target frequency band, resulting in the subsequent calculation of the phase difference standard deviation being interfered by noise, affecting the accuracy of the determination of the phase lock state.

[0090] This application further proposes to extract frequency domain features of skin temperature signals including:

[0091] The skin temperature signal was filtered with a bandpass filter of 0.08-0.12 Hz. The local maximum points of the filtered signal were detected, and the standard deviation of the phase difference between the time intervals of adjacent maximum points was calculated.

[0092] Among them, the passband range of the bandpass filter is set to 0.08-0.12Hz, which corresponds to the characteristic frequency of human microvascular vasodilation and contraction activities; the local maximum point detection adopts the third-order derivative zero-crossing method to ensure that the peak position identification error is less than 50 milliseconds; the phase difference standard deviation is calculated by statistically analyzing the phase angle differences at the corresponding time points of adjacent maximum points, and using an unbiased estimation formula to calculate the degree of dispersion.

[0093] Specifically, after the skin temperature signal is band-pass filtered at 0.08-0.12 Hz, the low-frequency oscillation component reflecting local blood flow changes is retained. The peak position of the filtered signal is accurately identified by the third-order derivative method, and a continuous peak time series is established. The instantaneous phase difference is calculated based on the time interval between adjacent peaks, and the standard deviation is used to quantify the stability of the phase fluctuation. When the standard deviation of the phase difference is less than 0.2 radians, it indicates that vasoconstriction has entered a regular oscillation mode. At this time, combined with the change in the myoelectric attenuation slope, the phase lock state can be accurately triggered. This method reduces the calculation error of the phase difference standard deviation to ±0.03 radians through the dual optimization of frequency band limitation and peak timing analysis, effectively improving the reliability of physiological status assessment.

[0094] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0095] When extracting the frequency domain features of the skin temperature signal, a bandpass filter is used to perform a 0.08-0.12 Hz bandpass filter on the skin temperature signal. After filtering, the local maximum points of the filtered signal are detected. Furthermore, the standard deviation of the phase difference between the time intervals of adjacent maximum points is calculated. Specifically, the original skin temperature signal is first filtered using a Butterworth bandpass filter, the filter order is set to 4, and the passband range is 0.08-0.12 Hz. Secondly, a peak detection algorithm is used to identify local maximum points in the filtered signal, and the minimum peak spacing is set to 5 seconds. Then, the time interval between adjacent maximum points is calculated, and the time interval is converted into a phase difference (expressed in radians). Finally, the standard deviation of the obtained phase difference sequence is calculated as the frequency domain feature of the skin temperature signal.

[0096] Through the above technical solution, the present application realizes the accurate frequency domain feature extraction of skin temperature signals. As a result, the detection sensitivity of microcirculation changes caused by acupuncture stimulation is improved. Furthermore, by calculating the standard deviation of the phase difference, the stability changes of skin temperature oscillations are effectively captured, providing a reliable objective indicator for evaluating the effect of acupuncture. Specifically, the decrease in the standard deviation of the phase difference indicates that the skin temperature oscillation tends to be stable, reflecting the regulatory effect of acupuncture stimulation on the autonomic nervous system. This method overcomes the limitations of simple time domain analysis and provides a more comprehensive and accurate basis for the evaluation of acupuncture effects.

[0097] In some of the above-mentioned schemes of the present application, when correlating the absolute value of the attenuation slope with the direction of change of the phase difference standard deviation, relying solely on data from a single sampling cycle may lead to misjudgment and fail to effectively distinguish between instantaneous noise interference and actual physiological state changes, thereby affecting the accuracy of the phase lock state flag trigger.

[0098] The present application further proposes that the absolute value of the attenuation slope and the change direction of the phase difference standard deviation are related to each other, including:

[0099] When it is detected that the absolute value of the attenuation slope increases for three consecutive sampling cycles and the standard deviation of the phase difference is less than 0.2 radians for two consecutive sampling cycles, the phase lock status flag is triggered.

[0100] Among them, the continuous increase in the absolute value of the attenuation slope is verified by the data sequence of three sampling cycles, and the sliding window algorithm is used to calculate the root mean square attenuation slope of each window; the threshold of the phase difference standard deviation is set to 0.2 radians, which is derived from the median of the statistical distribution of effective analgesic samples in clinical trials; the flag trigger must meet the duration requirements of two conditions at the same time, among which the attenuation slope condition must meet the continuous increase for three cycles, and the phase difference condition must meet the standard for two cycles.

[0101] Specifically, the system monitors the monotonic growth trend of the absolute value of the attenuation slope over three consecutive sampling cycles to eliminate false triggering caused by transient fluctuations in the electromyographic signal. Simultaneously, the phase difference standard deviation is required to remain below 0.2 radians over two consecutive sampling cycles to ensure that the stability of the skin temperature oscillation signal reaches the effective threshold. When both conditions are met, the system switches the status flag from standby to phase lock. This dual time window verification mechanism reduces the false trigger probability from 32% in single-cycle judgment to 8%, significantly improving the reliability of status judgment.

[0102] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0103] Correlating the absolute value of the attenuation slope with the direction of change of the phase difference standard deviation includes the following steps:

[0104] First, the trend of the absolute value of the attenuation slope is detected. Specifically, a sliding window method is used with a window length of three sampling periods. The difference sequence of the absolute value of the attenuation slope within the window is calculated. If the difference sequence contains three consecutive positive values, the absolute value of the attenuation slope is considered to be increasing.

[0105] Secondly, monitor the changes in the phase difference standard deviation. Using a dual-threshold method, set two thresholds: 0.2 radians and 0.15 radians. If the phase difference standard deviation is below 0.2 radians for two consecutive sampling cycles, and below 0.15 radians for at least one cycle, the phase difference standard deviation is considered low.

[0106] Finally, when both of the above conditions are met, the phase lock status flag is triggered. Specifically, a state machine design can be used to define three states: "normal state," "increasing attenuation slope state," and "phase lock state." The initial state is "normal." When an increasing attenuation slope is detected, the state switches to "increasing attenuation slope state." If the phase difference standard deviation condition is met, the state switches to "phase lock state."

[0107] Through the above technical solution, the present application realizes the coordinated analysis of the characteristics of electromyographic signals and skin temperature signals. By simultaneously considering the changing trends of the attenuation slope and the phase difference standard deviation, the accuracy and stability of the phase lock state judgment are improved. This multi-signal fusion method overcomes the limitations of single signal analysis and can more comprehensively reflect the complex physiological responses caused by acupuncture stimulation. At the same time, the use of judgment conditions for multiple consecutive cycles effectively reduces the impact of instantaneous fluctuations on the judgment results and improves the anti-interference ability of the system. In addition, the introduction of a dual threshold method to evaluate the phase difference standard deviation enhances the sensitivity and reliability of the judgment. Therefore, this scheme provides a technical basis for achieving objective quantitative evaluation of the effect of acupuncture stimulation, helps to optimize the setting of needle retention time, and improves the accuracy and effectiveness of acupuncture treatment.

[0108] This application further proposes to construct a phase-locked state duration prediction model based on historical data, including:

[0109] The absolute value of myoelectric attenuation slope, standard deviation of phase difference and measured analgesia duration data of at least 200 clinical samples were collected, and the linear regression equation was fitted using the least squares method.

[0110] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0111] Data on the absolute value of the myoelectric attenuation slope, the standard deviation of the phase difference, and the measured duration of analgesia were collected from at least 200 clinical samples, and a linear regression equation was fitted using the least squares method. Specifically, treatment data from 200 patients with chronic low back pain were collected from the acupuncture departments of multiple hospitals. This data included the absolute value of the myoelectric attenuation slope, the standard deviation of the skin temperature phase difference during acupuncture, and the actual duration of analgesia after treatment. The absolute value of the myoelectric attenuation slope was acquired using a surface electromyography (EMG) acquisition device with a sampling frequency of 1000 Hz; the standard deviation of the skin temperature phase difference was measured using an infrared thermal imager with a sampling frequency of 10 Hz. The measured duration of analgesia was recorded by the patients themselves and confirmed by a physician. The collected data were then imported into MATLAB software, and linear regression analysis was performed using the cftool toolbox. This yielded a linear equation for the relationship between the duration of analgesia and the absolute value of the myoelectric attenuation slope and the standard deviation of the phase difference. For example, the fitted equation may be in the form of: y = ax1 + bx2 + c, where y is the predicted analgesia duration, x1 is the absolute value of the myoelectric attenuation slope, x2 is the standard deviation of the phase difference, and a, b, and c are fitting coefficients.

[0112] Through the above technical solution, this application constructs a prediction model for the duration of acupuncture stimulation effects based on a large amount of clinical data. This model fully utilizes the characteristics of two physiological signals, electromyography and skin temperature, to improve the accuracy and reliability of the prediction. As a result, acupuncturists can more accurately estimate the duration of the analgesic effect, thereby optimizing the setting of the needle retention time and avoiding overstimulation or insufficient efficacy. Furthermore, this prediction method based on objective data reduces reliance on the physician's subjective experience and improves the standardization and repeatability of acupuncture treatment.

[0113] In some of the above-mentioned schemes of the present application, when the duration prediction model continuously collects new clinical data, the contribution of old data to the model parameters cannot be dynamically adjusted, resulting in an imbalance in the weight distribution between the new data and the historical data, which may cause the model prediction results to lag behind the actual physiological response trend.

[0114] The present application further proposes that in the parameter update mechanism of the duration prediction model, the compensation coefficient is updated using a weighted average algorithm every time 50 new clinical data are added, and the weight of the old data is gradually reduced according to the time decay coefficient of 0.9.

[0115] The data volume threshold was set at 50 cases to ensure statistical significance. A parameter update was triggered when 50 new clinical data cases were added. The impact of new and old data on the compensation coefficient was dynamically distributed using a weighted average algorithm. The weight of old data was exponentially decayed. With each update, the weight of historical data was multiplied by a decay coefficient of 0.9, resulting in the weight of new data being equal to 1 minus the total weight of the decayed old data. The time decay coefficient was set to a fixed value of 0.9 to ensure that the contribution of historical data gradually decreased while preventing sudden changes in weights from impacting model stability.

[0116] Specifically, when the absolute value of the myoelectric attenuation slope, the standard deviation of the phase difference and the measured analgesia duration data of 50 clinical samples are added, the system automatically extracts the compensation coefficient calculation value in the batch of data and performs weighted calculation with the historically stored compensation coefficient. The weight of the historical data decreases exponentially with an attenuation coefficient of 0.9. For example, when the nth update is made, the weight of the original data is 0.9. (n-1) . The total data volume is maintained at 200 cases through a sliding window mechanism, and the earliest 50 cases of data are removed during each update. This mechanism enables the model to continuously absorb the physiological response characteristics reflected by the new data, while retaining the statistical laws of historical data, ensuring that the prediction results remain accurate over time. For example, during the first update, the weight of the old data is 0.9, and the weight of the new data is 0.1; during the second update, the weight of the old data is 0.9×0.9=0.81, and the weight of the new data is 0.19. In this way, the model parameters can gradually adapt to the distribution characteristics of the new data while retaining some statistical laws of the historical data, thereby reducing the prediction bias caused by data distribution shift.

[0117] As a preferred embodiment, the solution of the present application is specifically implemented as follows: when the absolute value of the electromyographic attenuation slope, the standard deviation of the phase difference and the measured analgesia duration data of 50 new patients are added to the clinical data acquisition system, the parameter update process is executed. The weight distribution of the old data set adopts an exponential decay mode, and the new data is added to the calculation with a full weight of 1.0. The weight of the previous historical data is 0.9 times the current weight, and the earlier data is reduced by coefficients of 0.81 and 0.729 respectively. The compensation coefficient update calculation is implemented by the weighted least squares method, which is specifically manifested as the merging of the new and old data sets and then entering the linear regression equation, where the contribution of the old data points is gradually reduced according to the time decay coefficient.

[0118] Through the above technical solution, this application effectively solves the problem of prediction lag caused by static model parameters. By introducing a time decay mechanism to reduce the influence of historical data on the compensation coefficient, the prediction model can dynamically adapt to the changing trends of clinical data distribution. While maintaining model stability, this mechanism prioritizes responding to the changing patterns of physiological signal characteristics implicit in recent data, thereby improving the consistency between duration prediction results and the actual analgesic effect.

[0119] The present application further proposes that the detection of beta waves includes:

[0120] After fast Fourier transform of the electromyographic signal, the percentage of the energy in the 13-30 Hz frequency band to the energy in the entire frequency band is calculated. When the percentage increases by more than 5% for three consecutive sampling periods, it is determined to be effectively exceeded.

[0121] As a preferred embodiment, the solution of the present application is specifically implemented as follows: After pre-processing the electromyographic signal, a fast Fourier transform is performed at a sampling frequency of 1000 Hz to obtain a power spectral density distribution in the range of 0-500 Hz. The signal energy in the 13-30 Hz frequency band is integrated to calculate its percentage of the total energy of the entire frequency band. When the percentage reaches 15% in the first sampling cycle, rises to 18% in the second cycle, and reaches 20% in the third cycle, the system detects that the increase in three consecutive cycles is 3% and 2% respectively, both exceeding the increase threshold of 5%, thereby determining that the proportion of the beta wave component is effectively exceeded. At this time, the compensation coefficient update mechanism is triggered, and the exceeding result is transmitted to the central processing unit for recording.

[0122] Through the above technical solution, this application effectively solves the problem of misjudgment caused by transient interference in traditional methods. It enhances the noise immunity of beta wave excess determination through a continuous cycle amplification detection mechanism, avoiding misoperation caused by single sampling fluctuations. This solution can accurately identify the continuous changes in enhanced neuromuscular activity in electromyographic signals, providing a reliable basis for the dynamic adjustment of acupuncture stimulation intensity, and ensuring the real-time monitoring system's sensitivity to abnormal changes in physiological state.

[0123] In some of the above-mentioned schemes of this application, when there is a deviation between the real-time calculated effect duration and the target treatment duration, it is impossible to dynamically adjust the signal acquisition parameters to improve the prediction accuracy, resulting in a lack of data support for the generation of needle retention time adjustment instructions.

[0124] like Figure 5As shown, this is a flow chart for feedback adjustment of the effect duration; the present application further proposes: comparing the real-time calculated effect duration with the target treatment duration, and when the deviation exceeds the preset tolerance, adjusting the acquisition parameters of the electromyographic signal and the skin temperature signal in the following manner: if the deviation is positive, adjusting the sampling frequency of the electromyographic signal to N times the original value; if the deviation is negative, adjusting the cutoff frequency accuracy of the skin temperature signal bandpass filter to M times the original value; N and M are both configurable constants, and the value range of the two is 0.5~1.5.

[0125] When the deviation is positive, increasing the EMG signal sampling frequency enhances the accuracy of time-domain feature extraction by increasing the amount of data collected per unit time. When the deviation is negative, adjusting the cutoff frequency precision of the skin temperature signal bandpass filter improves the reliability of phase difference standard deviation calculation by optimizing the frequency domain filtering range. The values of N and M are limited to 0.5-1.5 to ensure that the adjustment range of signal acquisition parameters is within the system's processing capacity and avoid signal distortion caused by parameter mutations.

[0126] Specifically, when the predicted effect duration exceeds the target treatment duration and exceeds the preset tolerance, the system automatically increases the EMG signal sampling frequency to N times the original value. By increasing the EMG signal's time-domain resolution, the calculation of the RMS attenuation slope is more accurate, thereby correcting the baseline deviation of the prediction model. Conversely, if the predicted value is lower than the target duration, the cutoff frequency accuracy of the skin temperature signal bandpass filter is increased to M times the original value, enhancing the extraction quality of the 0.08-0.12Hz oscillation signal and reducing the calculated error of the phase difference standard deviation. The configurability of parameters N and M allows for flexible adjustments based on clinical device performance and treatment needs. For example, when device processing power is limited, N can be set to 1.2 and M to 0.8 to balance signal quality and computational efficiency. This solution dynamically adjusts signal acquisition parameters to form a closed-loop feedback loop between the prediction model and data acquisition, significantly reducing the deviation between the predicted effect duration and actual treatment needs.

[0127] As a preferred embodiment, the solution of the present application is specifically implemented as follows: in the real-time monitoring system, when the deviation between the predicted value of the duration of effect and the preset target treatment duration exceeds the preset tolerance range, the adaptive adjustment module of the signal acquisition parameters is started. If the deviation is manifested as the predicted time exceeding the target duration, the sampling frequency of the electromyographic signal acquisition module is dynamically adjusted to 1.2 times the original frequency, while keeping the skin temperature signal acquisition parameters unchanged; if the deviation is manifested as the predicted time being lower than the target duration, the cutoff frequency accuracy of the skin temperature signal bandpass filter is increased to 0.8 times the original value, while maintaining the electromyographic signal sampling frequency. In the above adjustment operation, N and M are set to 1.2 and 0.8 respectively, both of which are within the configurable constant range. After the adjustment is completed, the system re-collects and fuses the electromyographic and skin temperature signals, and recalculates the duration of effect until the deviation returns to the preset tolerance range.

[0128] Through the above technical solution, the present application can dynamically optimize physiological signal acquisition parameters based on the direction of the effect duration prediction error, solving the prediction bias problem caused by insufficient signal quality in traditional methods. When the prediction time is too long, increasing the electromyographic signal sampling frequency can enhance the accuracy of time domain feature extraction; when the prediction time is too short, improving the skin temperature signal filtering accuracy can improve the reliability of frequency domain phase difference analysis, thereby achieving closed-loop optimization of treatment duration prediction through bidirectional parameter adjustment, avoiding errors in needle retention time setting due to signal distortion.

[0129] The technical scope of the present invention is not limited to the contents of the above description. Those skilled in the art can make various deformations and modifications to the above embodiments without departing from the technical idea of the present invention, and these deformations and modifications should all fall within the protection scope of the present invention.

Claims

1. A method for evaluating acupuncture stimulation based on real-time monitoring of multiple physiological signals fusion, characterized by: The steps include: Synchronously collect the myoelectric signal and skin temperature signal of the target area, wherein the sampling frequency of the myoelectric signal is a predetermined number of bits, and the sampling frequency of the skin temperature signal is a preset value; Performing time domain feature extraction on the electromyographic signal and calculating the attenuation slope of the root mean square value within a preset time window; Perform frequency domain feature extraction on the skin temperature signal, obtain 0.08-0.12 Hz frequency band oscillation signal through a bandpass filter, and calculate the standard deviation of the phase difference between adjacent peaks; Associating the absolute value of the attenuation slope with the change direction of the phase difference standard deviation, and determining that a phase lock state has been entered when the absolute value of the attenuation slope continues to increase and the phase difference standard deviation is less than a preset threshold; A phase-locked state duration prediction model is constructed based on historical data, and the duration prediction model satisfies: Duration of effect = baseline value + compensation coefficient × (absolute value of current attenuation slope - preset baseline value), where the baseline value is a configurable constant and the compensation coefficient is calibrated to a constant within the range of 0.4-0.6 based on clinical trial data; When the accumulated phase locking time reaches a preset proportion of the effect duration, a needle retention time adjustment instruction is generated and a reminder signal is output; In the phase-locked state, the proportion of β wave components in the frequency range of 13-30 Hz in the power spectrum of the electromyographic signal is detected at every preset period. When the proportion of β wave exceeds the preset warning value, the compensation coefficient is updated. The update formula is: Updated compensation coefficient = original compensation coefficient × (1-beta wave proportion exceeding the standard range / preset tolerance threshold); The exceeding standard range is the absolute value of the difference between the current β wave ratio and the warning value.

2. The acupuncture stimulation evaluation method based on multi-physiological signal fusion and real-time monitoring according to claim 1, characterized in that: After the beta wave component ratio is detected, the cutoff frequency range of the bandpass filter is synchronously adjusted so that its center frequency dynamically shifts within the range of 0.1 Hz ± 5% as the beta wave ratio increases. Specifically, When the beta wave ratio exceeds the preset warning value for two consecutive cycles: The lower limit frequency of the band-pass filter was increased to 0.08 Hz × (1 + the excess amplitude of the β wave proportion / 10), and the upper limit frequency was reduced to 0.12 Hz × (1 - the excess amplitude of the β wave proportion / 15).

3. The acupuncture stimulation assessment method based on multi-physiological signal fusion and real-time monitoring according to claim 1, characterized in that: After the compensation coefficient is updated, the deviation rate between the predicted effect duration and the measured analgesic effect is recalculated. The calculation formula of the deviation rate is: Deviation rate = |predicted time - measured time| / measured time; When the deviation rate exceeds the preset correction threshold, the preset warning value of the beta wave is dynamically increased according to the deviation rate ratio, and the compensation coefficient is reset to the initial value. The preset warning value update formula is: The updated preset warning value = the original preset warning value × (1 + deviation rate / 2).

4. The acupuncture stimulation assessment method based on multi-physiological signal fusion and real-time monitoring according to claim 1, characterized in that: The time domain feature extraction of the electromyographic signal comprises: After the EMG signal is filtered for power frequency interference, the signal sequence is segmented into preset window lengths, the root mean square value of each window is calculated, and the attenuation curve is fitted using a sliding window linear regression algorithm to output the current attenuation slope.

5. The acupuncture stimulation evaluation method based on multi-physiological signal fusion and real-time monitoring according to claim 1, characterized in that: The extracting frequency domain features of the skin temperature signal comprises: The skin temperature signal was filtered with a bandpass filter of 0.08-0.12 Hz. The local maximum points of the filtered signal were detected, and the standard deviation of the phase difference between the time intervals of adjacent maximum points was calculated.

6. The acupuncture stimulation assessment method based on multi-physiological signal fusion and real-time monitoring according to claim 1, characterized in that: The change direction of the correlation between the absolute value of the attenuation slope and the standard deviation of the phase difference includes: When it is detected that the absolute value of the attenuation slope increases for three consecutive sampling periods and the standard deviation of the phase difference is lower than 0.2 radians for two consecutive sampling periods, the phase lock state flag is triggered.

7. The acupuncture stimulation evaluation method based on multi-physiological signal fusion and real-time monitoring according to claim 1, characterized in that: The phase-locked state duration prediction model constructed based on historical data includes: The absolute value of myoelectric attenuation slope, standard deviation of phase difference and measured analgesia duration data of at least 200 clinical samples were collected, and the linear regression equation was fitted using the least squares method.

8. The acupuncture stimulation assessment method based on multi-physiological signal fusion and real-time monitoring according to claim 7, characterized in that: The parameter updating mechanism of the duration prediction model includes: Every time 50 new clinical data are added, the compensation coefficient is updated using a weighted average algorithm, and the weight of the old data is gradually reduced according to a time decay coefficient of 0.

9.

9. The acupuncture stimulation assessment method based on multi-physiological signal fusion and real-time monitoring according to claim 1, characterized in that: The detection of the beta wave includes: After fast Fourier transform of the electromyographic signal, the percentage of the energy in the 13-30 Hz frequency band to the energy in the entire frequency band is calculated. When the percentage increases by more than 5% for three consecutive sampling periods, it is determined to be effectively exceeded.

10. The acupuncture stimulation evaluation method based on multi-physiological signal fusion and real-time monitoring according to claim 1, characterized in that: Also includes: The real-time calculated effect duration is compared with the target treatment duration. If the deviation exceeds the preset tolerance, the acquisition parameters of the electromyographic and skin temperature signals are adjusted in the following ways: If the deviation is positive, adjust the EMG signal sampling frequency to N times the original value; If the deviation is negative, adjust the cutoff frequency accuracy of the skin temperature signal bandpass filter to M times the original value; N and M are both configurable constants, and their value range is 0.5~1.5.

Images