Clinical application of combination of traditional Chinese medicine and modern physiotherapy technology to treatment of chronic fatigue syndrome

Abstract

The invention discloses a clinical application for treating chronic fatigue syndrome by combining traditional Chinese medicine with modern physiotherapy technology, which comprises the following steps: continuously collecting meridian point electric signals, electroencephalogram signals and peripheral nerve electrophysiological signals of a patient, and synchronously recording traditional Chinese medicine taking information; inputting the signals into a meridian-nerve electrophysiology coupling model to calculate a coupling characteristic parameter set; calling a fatigue factor time sequence prediction algorithm, and predicting a future fatigue factor change trend in combination with historical data; adjusting chronic fatigue brain-body interaction intervention system parameters and a traditional Chinese medicine taking scheme according to a prediction result; alternately implementing electroencephalogram adjustment and meridian point electrical stimulation intervention according to the adjusted parameters; after intervention, signal acquisition and analysis are periodically repeated, and a result optimization scheme is compared to form a closed loop. According to the application, personalized treatment is achieved through multi-signal integration analysis and dynamic parameter adjustment, the advantages of long-acting conditioning of traditional Chinese medicine and rapid intervention of modern physical therapy are combined, and the treatment accuracy and effect are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of chronic fatigue syndrome treatment technology, and in particular to the clinical application of traditional Chinese medicine combined with modern physiotherapy techniques in the treatment of chronic fatigue syndrome. Background Technology

[0002] Chronic fatigue syndrome (CFS) is a condition characterized by persistent or recurrent fatigue. Its long course and accompanying physical and psychological symptoms severely impact patients' daily lives and work abilities. Current clinical treatments, while traditional Chinese medicine (TCM) alone can improve symptoms by regulating the body's state, suffer from slow onset of action and reliance on subjective feelings for efficacy assessment. While modern physiotherapy techniques can quickly relieve local discomfort through physical stimulation, they struggle to provide precise intervention tailored to individual patient differences. With advancements in medical technology, the clinical demand for combining TCM with modern physiotherapy is increasing. There is an urgent need to construct a comprehensive treatment system that integrates multi-dimensional physiological signal acquisition, feature analysis, trend prediction, and dynamic intervention to achieve systematic and personalized treatment of CFS, thereby improving treatment effectiveness and patient compliance.

[0003] Existing technologies for treating chronic fatigue syndrome have two significant drawbacks. Firstly, current combined treatment plans lack in-depth coupling analysis of the patient's physiological signals. They often collect only single types of physiological signals without systematically linking the intrinsic connections between meridian and acupoint electrical signals, electroencephalogram (EEG) signals, and peripheral nerve electrophysiological signals. This fails to accurately capture the complex physiological changes associated with the condition, resulting in a lack of scientific and comprehensive basis for adjusting treatment plans. Secondly, existing technologies lack dynamic treatment adjustment mechanisms. Most plans are fixed after being determined, failing to optimize intervention parameters and TCM dosage plans in real time based on the temporal changes in the patient's fatigue state. Furthermore, they lack a closed-loop process of "collection-analysis-prediction-intervention-reassessment," making it difficult to adapt to the dynamic changes in the patient's condition, thus affecting the accuracy and long-term effectiveness of treatment. Summary of the Invention

[0004] In order to overcome the shortcomings and deficiencies of existing technologies, this invention provides a clinical application of traditional Chinese medicine combined with modern physiotherapy techniques for the treatment of chronic fatigue syndrome.

[0005] The technical solution adopted in this invention is the clinical application of traditional Chinese medicine combined with modern physiotherapy in the treatment of chronic fatigue syndrome, including the following steps: S1, using a multi-channel physiological signal acquisition device to continuously acquire the meridian acupoint electrical signals, electroencephalogram (EEG) signals, and peripheral nerve electrophysiological signals of patients with chronic fatigue syndrome. The acquisition duration is set to 24 hours, and the sampling frequency is adjusted to 500Hz. Simultaneously, the daily dosage and frequency of traditional Chinese medicine taken by the patient, as well as the physiotherapy intervention period, are recorded; S2, the meridian acupoint electrical signals, EEG signals, and peripheral nerve electrophysiological signals acquired in S1 are input into a meridian-neurophysiological coupling model. The model calculates the coupling strength, phase synchronization, and frequency response characteristics between different signals, and outputs a set of coupling feature parameters, which includes 12 feature indicators; S3, based on the coupling feature parameter set output in S2, a fatigue factor time-series prediction algorithm is called, using the patient's physiological signal historical data from the past 3 months as training samples, to predict the trend of fatigue factor changes in the patient over the next 7 days. Output fatigue factor time-series prediction curves, with a curve time interval set to 1 hour; S4, based on the fatigue factor time-series prediction curves output in S3 and combined with the patient's traditional Chinese medicine (TCM) medication regimen, adjust the intervention parameters of the chronic fatigue brain-body interaction intervention system. The intervention parameters include electrical stimulation intensity, stimulation frequency, intervention duration, and intervention acupoint combination. Among them, the electrical stimulation intensity adjustment range is 0.1-5mA, and the stimulation frequency adjustment range is 1-100Hz; S5, using the chronic fatigue brain-body interaction intervention system according to the intervention parameters adjusted in S4, implement EEG modulation intervention and meridian acupoint electrical stimulation intervention on the patient. The two intervention methods are alternated, with a single intervention duration set to 30 minutes and a daily intervention frequency set to 2 times; S6, on the 1st, 3rd, and 7th days after the intervention in S5, repeat the signal acquisition and coupling feature calculation process of S1-S2, compare the changes in the coupling feature parameter set before and after the intervention, and adjust the intervention parameters in S4 and the patient's TCM medication dosage again based on the comparison results to form a closed-loop treatment process.

[0006] Furthermore, when calculating the coupling strength in S2, the meridian-neuroelectrophysiological coupling model uses the following model formula: ,in, This represents the electrophysiological coupling strength between meridians and nerves, with a value ranging from 0 to 1; Emeritian. This represents the amplitude of the electrical signal at the meridian acupoint of the i-th acquisition point; Energy, This represents the amplitude of the neurophysiological signal at the i-th acquisition point; This represents the phase difference between the electrical signal of the meridian acupoint and the neurophysiological signal at the i-th acquisition point. This represents the total number of data collection points, with a value of 24.

[0007] Furthermore, when predicting the fatigue factor in S3, the fatigue factor time-series prediction algorithm uses the following model formula: ,in, express The fatigue factor value at any given time; This represents the initial value of the fatigue factor at time t; This represents the fatigue factor attenuation coefficient, with a value range of [value missing]. Indicates the prediction time interval This represents the coupling characteristic influence coefficient, with a value range of 0.5-2.0; This represents the normalized value of the j-th coupled feature parameter at time t; This indicates the number of coupling feature parameters, with a value of 12.

[0008] Furthermore, when adjusting the electrical stimulation intensity in S4, the chronic fatigue brain-body interaction intervention system uses the following model formula: ,in, This indicates the adjusted intensity of electrical stimulation; This represents the baseline intensity of electrical stimulation, with a value ranging from 0.1 to... This represents the fatigue factor influence coefficient, with a value range of 1.0-3.0; This represents the fatigue factor value at the current moment; This represents the fatigue factor threshold, with a value of [value missing]. This represents the maximum fatigue factor, which is 1.0.

[0009] Furthermore, when calculating phase synchronization in S2, the meridian-neuroelectrophysiological coupling model uses the following model formula: ,in, This represents the phase synchronization index, with a value range of [value range missing]. This represents the total number of data collection points, with values ​​ranging from [value 1] to [value 2]. Represents the imaginary unit; This represents the phase of the electrical signal at the meridian acupoint of the i-th acquisition point; This represents the phase of the neurophysiological signal at the i-th acquisition point.

[0010] Furthermore, when the fatigue factor time-series prediction algorithm corrects the prediction curve in S3, it adopts the following model formula: ,in, This represents the corrected predicted fatigue factor value; This represents the predicted fatigue factor value before correction; This represents the effect coefficient of traditional Chinese medicine dosage, with a value range of [value range missing]. Indicates the patient's current dosage of traditional Chinese medicine. This indicates the average dosage of traditional Chinese medicine taken by the patient over the past 3 months.

[0011] Further, S2 includes the following sub-steps: S21, performing signal filtering on the meridian acupoint electrical signals, EEG signals, and peripheral nerve electrophysiological signals collected in S1, using a bandpass filter to retain signal components in the 5-50Hz frequency band, filtering out power frequency interference and baseline drift signals, setting the filter order to 8th order and the transition band width to 2Hz during the filtering process; S22, aligning the three types of filtered signals according to the acquisition timestamps to ensure that the three types of signal data at the same time point correspond one-to-one, with the timestamp error controlled within ±1ms, and forming a multi-dimensional signal matrix after alignment, where the number of rows in the matrix corresponds to the number of sampling points corresponding to the total acquisition time, and the number of columns... The number is 3; S23, input the multi-dimensional signal matrix into the meridian-neuro-electrophysiological coupling model, call the built-in feature extraction module of the model, and calculate the coupling strength, phase synchronization and frequency response features of each row of data in the matrix point by point. The sliding window method is used in the calculation process, the window size is set to 100 sampling points and the window step size is set to 50 sampling points; S24, arrange the feature values ​​obtained by point by point in chronological order, remove outliers to form a coupling feature parameter set. The outlier judgment criterion is the value that deviates from the mean by 3 times the standard deviation. After removing outliers, linear interpolation is performed on the parameter set to complete it, ensuring that the parameter set data is continuous and without missing data.

[0012] Further, S3 includes the following sub-steps: S31, extracting historical physiological signal data from the patient's medical record database for the past 3 months. This data includes daily meridian-neuro-electrophysiological coupling characteristic parameters and corresponding measured values ​​of fatigue factors. The data sampling frequency is consistent with S1, both being 500Hz, and the data duration is no less than 90 days; S32, dividing the extracted historical data into a training set and a validation set in a 7:3 ratio. The training set is used for model training of the fatigue factor time-series prediction algorithm, and the validation set is used for verifying the algorithm's prediction accuracy. The division process uses random sampling to ensure uniform sample distribution; S33 The coupled feature parameter set output from S2 and the training set data are input into the fatigue factor time series prediction algorithm. The weight parameters in the algorithm are adjusted by gradient descent to minimize the error between the algorithm's predicted value and the measured fatigue factor value in the training set. The number of training iterations is set to 1000, and the learning rate is set to 0.001. In S34, the trained algorithm is tested using validation set data. The mean square error between the predicted value and the measured value is calculated. If the mean square error is less than 0.01, the algorithm is called to predict the fatigue factor change trend for the next 7 days. If the mean square error is greater than or equal to 0.01, the amount of training set data is increased and the algorithm is retrained.

[0013] Further, S4 includes the following sub-steps: S41, performing feature analysis on the fatigue factor time-series prediction curve output by S3, extracting the peak value, trough value, and slope of change in the curve. The peak value corresponds to the time when the patient's fatigue level is highest each day, the trough value corresponds to the time when the fatigue level is lowest, and the slope of change reflects the rate of change in fatigue level. The analysis process uses numerical differentiation to calculate the curve slope; S42, determining the intervention period of the chronic fatigue brain-body interaction intervention system based on the curve feature analysis results, setting the intervention period 1 hour before the peak value of the fatigue factor to avoid intervention during the trough value period, and determining the number of daily interventions based on the number of peak values. If there are 2 peak values ​​per day, then 2 interventions are set; S43, combining... The patient's current TCM treatment plan includes dosage, frequency, and drug components. Different drug components correspond to different intervention parameter adjustment coefficients. For example, TCM containing ginseng has a coefficient of 1.2, and TCM containing astragalus has a coefficient of 1.1. S44: Based on the fatigue factor curve characteristics and TCM component coefficients, the electrical stimulation intensity, stimulation frequency, intervention duration, and intervention acupoint combination of the intervention system are adjusted. The electrical stimulation intensity is linearly adjusted according to the difference between the peak and trough values ​​of the fatigue factor. The stimulation frequency is segmented according to the slope of the curve. When the absolute value of the slope is greater than 0.05, the frequency is set to 50-100Hz, and when the absolute value of the slope is less than or equal to 0.05, the frequency is set to 1-50Hz.

[0014] Further, S5 includes the following sub-steps: S51, the intervention parameters adjusted in S4 are input into the control module of the chronic fatigue brain-body interaction intervention system. The control module parses the parameters, converting the electrical stimulation intensity and stimulation frequency parameters into hardware drive signals, and the intervention duration and intervention acupoint combination parameters into timing control signals. During the parsing process, the parameter range is verified to ensure that the parameters are within the safe threshold. S52, the EEG modulation module of the intervention system generates the corresponding EEG stimulation signal. This signal is a sine wave signal, the signal amplitude is determined by the electrical stimulation intensity parameter, and the signal frequency is determined by the stimulation frequency parameter. The EEG modulation module applies the stimulation signal to the patient's prefrontal cortex region through electrode pads. The electrode pads are made of silver-silver chloride material and have impedance control. The current is controlled below 5kΩ; S53, the meridian acupoint stimulation module of the intervention system generates a meridian electrical stimulation signal. This signal is a square wave signal with a duty cycle set to 50%. The combination of stimulation acupoints is selected according to the parameters determined in S4, usually including Zusanli, Sanyinjiao, and Hegu acupoints. The stimulation time for each acupoint is allocated to 10 minutes, and the acupoint stimulation sequence is from the lower limb to the upper limb; S54, the intervention is implemented by alternating between EEG regulation intervention and meridian acupoint electrical stimulation intervention. First, 15 minutes of EEG regulation intervention is performed, followed by 15 minutes of meridian acupoint electrical stimulation intervention. The total duration of a single intervention is 30 minutes. The patient's physiological signals are monitored in real time during the intervention. If abnormal signals occur, the intervention is paused and resumed after the signals return to normal.

[0015] Beneficial Effects: This invention proposes a clinical application of traditional Chinese medicine combined with modern physiotherapy techniques for the treatment of chronic fatigue syndrome. It continuously collects electrical signals from meridian acupoints, electroencephalograms (EEGs), and peripheral nerve electrophysiological signals for 24 hours. Combined with a meridian-neurophysiological coupling model, it performs deep coupling analysis of multiple signals, systematically correlating the coupling strength and phase synchronization characteristics between different signals. This solves the problem of existing technologies that only collect single physiological signals and cannot capture complex physiological changes, providing a comprehensive scientific basis for adjusting treatment plans. Simultaneously, it uses a fatigue factor time-series prediction algorithm to predict the patient's future fatigue trend. Combined with a chronic fatigue brain-body interaction intervention system, it dynamically adjusts intervention parameters such as electrical stimulation intensity and frequency with the traditional Chinese medicine dosage plan. Through multiple signal acquisitions and parameter comparisons after intervention, a closed-loop treatment process is formed, effectively compensating for the shortcomings of existing technologies that lack dynamic adjustment mechanisms and cannot adapt to changes in the condition. Furthermore, this application achieves synergistic cooperation between traditional Chinese medicine conditioning and modern physiotherapy intervention. It leverages the long-term effects of traditional Chinese medicine conditioning while relying on modern physiotherapy techniques for rapid and precise intervention, significantly improving the systematization and personalization of treatment, further enhancing treatment efficacy and patient compliance. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the clinical application of the present invention. Detailed Implementation

[0017] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] like Figure 1 As shown, the clinical application of traditional Chinese medicine combined with modern physiotherapy techniques in the treatment of chronic fatigue syndrome includes the following steps: S1 uses a multi-channel physiological signal acquisition device to continuously acquire the electrical signals of meridian acupoints, electroencephalogram (EEG) signals, and peripheral nerve electrophysiological signals of patients with chronic fatigue syndrome. The acquisition time is set to 24 hours, the sampling frequency is adjusted to 500Hz, and the daily dosage, frequency of administration of traditional Chinese medicine, and physiotherapy intervention period of the patients are recorded. Specifically, S1, as the foundational data acquisition stage of the entire treatment process, obtains raw data that comprehensively reflects the patient's physiological state and treatment-related information, providing a reliable basis for subsequent model analysis, algorithm prediction, and intervention parameter adjustment. The pathogenesis of chronic fatigue syndrome is closely related to meridian function and neurophysiological activity. Collecting only a single type of signal cannot fully present the patient's physical state; therefore, it is necessary to simultaneously collect meridian acupoint electrical signals, electroencephalogram (EEG) signals, and peripheral nerve electrophysiological signals, while also recording information on traditional Chinese medicine usage. This ensures that the data dimensions cover physiological indicators and treatment intervention factors, avoiding deviations in subsequent analysis results due to missing data, and laying a data foundation for precision treatment. In the specific implementation process, a multi-channel physiological signal acquisition device is used for signal acquisition. The device sampling frequency is set to 500Hz, which can effectively capture subtle changes in meridian and neurophysiological signals, avoiding high-frequency signal loss or low-frequency signal interference. The acquisition duration is set to 24 hours, covering the patient's diurnal physiological rhythm changes and reflecting differences in physiological state at different times, avoiding sample limitations caused by short-term acquisition. During the data collection process, the collection electrodes must be precisely attached to the preset meridian acupoints and EEG collection sites to ensure signal acquisition stability. At the same time, a dedicated person is assigned to record the patient's daily dosage of traditional Chinese medicine (such as 5g, 10g, etc.), frequency of administration (such as 2 or 3 times a day), and the timing of physical therapy intervention (such as 9 am, 4 pm, etc.). The recorded data must be synchronized to the storage system in real time to ensure that the physiological signals and treatment information at each time point correspond one-to-one. After the data collection is completed, a preliminary integrity check is performed on the data, and invalid data with obvious abnormalities is removed.

[0019] S2 inputs the meridian acupoint electrical signals, EEG signals and peripheral nerve electrophysiological signals collected in S1 into the meridian-nerve electrophysiological coupling model. The model calculates the coupling strength, phase synchronization and frequency response characteristics between different signals and outputs a set of coupling feature parameters, which contains 12 feature indicators. Specifically, S2 is a crucial step in converting physiological signals into feature parameters. It analyzes various types of collected signals using a meridian-neuroelectrophysiological coupling model, uncovering the intrinsic correlations between different signals and outputting a set of coupling feature parameters that quantify the patient's meridian and nerve function status. This provides core input variables for subsequent fatigue factor prediction. Patients with chronic fatigue syndrome exhibit an abnormal coupling relationship between meridian electrical signals and neuroelectrophysiological signals. Single signal analysis cannot reflect this correlation, but the coupling model can transform the raw signals into clinically significant feature indicators by calculating coupling strength, phase synchronization, and frequency response characteristics. These indicators directly reflect the degree of correlation between the patient's fatigue state and physiological function, avoiding the limitations of relying solely on subjective symptom assessment and improving the objectivity and scientific rigor of subsequent analyses. In practice, the meridian acupoint electrical signals, EEG signals, and peripheral nerve electrophysiological signals collected in S1 are first imported into the signal preprocessing module of the meridian-neuroelectrophysiological coupling model. The module first aligns the three types of signals along their time axes to ensure data matching at the same time point, with alignment errors controlled within ±1ms to avoid time deviations affecting the coupling analysis results. The model then calls the coupling strength calculation module to analyze the amplitude changes and correlations of different signals within the same time window to obtain the coupling strength value; it calls the phase synchronization calculation module to compare the phase difference changes between signals to determine the phase synchronization index; and it calls the frequency response feature module to analyze the peak value and attenuation trend of the signal response in different frequency bands to extract the frequency response parameters. Finally, the above calculation results are integrated to form a coupling feature parameter set containing 12 feature indicators. Each indicator is labeled with a corresponding time node. The parameter set needs to be validated to ensure that the values ​​of each indicator are within a reasonable physiological range. After successful validation, it is output to the next step.

[0020] S3, based on the coupled feature parameter set output by S2, calls the fatigue factor time series prediction algorithm, using the patient's physiological signal history data from the past 3 months as training samples, to predict the trend of fatigue factor changes in the patient in the next 7 days, and outputs the fatigue factor time series prediction curve, with the curve time interval set to 1 hour. Specifically, S3, based on the set of coupled feature parameters output by S2, combines historical data to predict the future trend of fatigue factors in patients, providing a forward-looking basis for adjusting subsequent intervention plans and avoiding the lag caused by treatment plans based solely on the current state. Fatigue factors are key indicators that quantify the degree of fatigue in patients, and their changing trends can predict the direction of the evolution of the patient's fatigue state in advance. Through the fatigue factor time-series prediction algorithm, the correlation between coupled feature parameters and fatigue factors can be transformed into a predictable time-series curve. Doctors can identify the risk of fatigue aggravation in advance based on the curve, adjust treatment strategies in a timely manner, and prevent further development of fatigue symptoms. At the same time, training the algorithm with historical data can improve the accuracy of prediction and ensure that the prediction results can effectively guide clinical intervention. In the implementation process, the historical physiological signal data of the past 3 months are first extracted from the patient's medical record database. This data must include meridian acupoint electrical signals, electroencephalogram signals, peripheral nerve electrophysiological signals, and corresponding measured values ​​of fatigue factors, consistent with the dimensions collected by S1. The historical data sampling frequency is also 500Hz to ensure consistency with the current data format. Subsequently, the set of coupled feature parameters output from S2 is used as the input variable for the algorithm. The coupled feature parameters and measured fatigue factor values ​​from historical data are used as training samples and imported into the training module of the fatigue factor time-series prediction algorithm. The algorithm learns the changing patterns of feature parameters and fatigue factors from historical data to establish a prediction model. After model training, a prediction time interval of 1 hour is set, and fatigue factor values ​​for the next 7 days are predicted point-by-point, generating a fatigue factor time-series prediction curve. The curve needs to be labeled with the predicted value and corresponding confidence interval for each time point. After prediction, the curve is smoothed to eliminate abnormal fluctuations and ensure that the curve clearly reflects the overall trend of fatigue factor changes. Finally, the curve is output to step S4 for intervention parameter adjustment.

[0021] S4. Based on the fatigue factor time-series prediction curve output by S3 and combined with the patient's traditional Chinese medicine treatment plan, adjust the intervention parameters of the chronic fatigue brain-body interaction intervention system. The intervention parameters include electrical stimulation intensity, stimulation frequency, intervention duration and intervention acupoint combination. Among them, the electrical stimulation intensity adjustment range is 0.1-5mA and the stimulation frequency adjustment range is 1-100Hz. Specifically, S4 combines the prediction curve with the TCM (Traditional Chinese Medicine) dosage regimen to adjust the intervention parameters of the chronic fatigue brain-body interaction intervention system, achieving personalized optimization of the treatment plan and avoiding the poor treatment effect caused by using fixed intervention parameters. Patients with chronic fatigue syndrome exhibit significant individual differences; the trends of fatigue factors, responses to TCM, and sensitivity to physical therapy interventions vary among patients. Interventions based solely on general parameters cannot meet individual treatment needs. S4, by combining the fatigue trends reflected in the prediction curve with TCM dosage information, can specifically adjust intervention parameters, ensuring the intervention plan matches the patient's future fatigue state and synergizes with TCM treatment, improving the accuracy and synergistic effect of treatment. In practice, the fatigue factor time-series prediction curve output by S3 is first obtained. The peak periods, trough periods, and rates of change of fatigue factors in the curve are analyzed to determine the key time nodes for intervention (e.g., the optimal intervention period is one hour before the peak). Simultaneously, the daily average level of fatigue factors in the curve is statistically analyzed as the basis for adjusting the intervention intensity. Then retrieve the patient's TCM treatment plan recorded in S1, including dosage, frequency of administration and drug components. Different drug components correspond to different intervention parameter adjustment coefficients (e.g., TCM containing Qi-tonifying components corresponds to a coefficient of 1.2, and TCM containing blood-nourishing components corresponds to a coefficient of 1.1). The coefficients need to be preset based on clinical research data and stored in the system database. Subsequently, based on the fatigue factor level and the adjustment coefficient of traditional Chinese medicine, the intervention parameters of the chronic fatigue brain-body interaction intervention system were adjusted: the electrical stimulation intensity adjustment range was set to 0.1-5mA, with higher fatigue factors requiring higher stimulation intensity (e.g., 3mA for a fatigue factor of 0.8, and 1.5mA for 0.5); the stimulation frequency adjustment range was set to 1-100Hz, with higher frequency settings for faster fatigue factor changes (e.g., 80Hz for a 0.1 increase per hour, and 40Hz for a 0.05 increase per hour); the intervention duration was fixed at 30 minutes to ensure sufficient intervention effect and avoid overstimulation; the combination of intervention acupoints was selected based on the patient's abnormal meridian signals (e.g., commonly used acupoints such as Zusanli and Sanyinjiao). After the parameters were adjusted, an intervention parameter table was generated, which clearly specified the specific parameter values ​​for each intervention period. After confirmation by the clinician, the table was transmitted to the intervention system.

[0022] S5, through the chronic fatigue brain-body interaction intervention system according to the intervention parameters adjusted in S4, implements EEG modulation intervention and meridian acupoint electrical stimulation intervention on the patient. The two intervention methods are alternated, the duration of a single intervention is set to 30 minutes, and the number of interventions per day is set to 2. Specifically, S5 is the execution phase of the intervention plan adjusted from S4, transforming it into actual treatment procedures. Through the chronic fatigue brain-body interaction intervention system, it implements electroencephalography (EEG) regulation and meridian acupoint electrical stimulation intervention to directly address the patient's fatigue state, leveraging the rapid relief effect of modern physiotherapy techniques. Simultaneously, it synergizes with traditional Chinese medicine treatment to enhance the overall therapeutic effect. Patients with chronic fatigue syndrome often exhibit abnormal brain electrical activity and meridian dysfunction. EEG regulation intervention can improve the brain's neurophysiological state and alleviate mental fatigue; meridian acupoint electrical stimulation intervention can regulate the flow of qi and blood in the meridians and improve physical fatigue. Alternating these two intervention methods can simultaneously act on both the mental and physical levels, avoiding the limitations of a single intervention method. Furthermore, the intervention is implemented according to the adjusted parameters to ensure that the intensity and frequency of the intervention match the patient's condition, avoiding insufficient or excessive intervention. During implementation, the intervention parameter table determined by S4 is first imported into the control unit of the chronic fatigue brain-body interaction intervention system. The control unit parses the parameters, converting the electrical stimulation intensity and frequency into hardware drive signals, and the intervention duration and acupoint combination into timing control logic. After parsing, the parameters are verified for safety to ensure that the electrical stimulation intensity is within the safe range of 0.1-5mA and that the stimulation frequency does not exceed the human body's tolerance threshold. After verification, the system enters standby mode. During the preset intervention period, the system automatically starts the intervention process: first, EEG modulation intervention is implemented, outputting electrical stimulation signals of the set frequency and intensity through prefrontal cortex electrodes for 15 minutes. During this period, changes in EEG signals are monitored in real time. If abnormal fluctuations occur (such as the amplitude suddenly exceeding the normal range), the system automatically pauses for 5 seconds and then restarts. If the abnormality persists, the intervention is stopped. After EEG modulation, the system automatically switches to meridian acupoint electrical stimulation intervention, attaching electrodes to preset acupoints and outputting electrical stimulation signals of the corresponding frequency and intensity for 15 minutes. During the stimulation, the meridian acupoint electrical signals are monitored to ensure that the stimulation effect meets the target. The number of daily interventions is set to 2, implemented in the morning and afternoon before the peak of fatigue factors. After each intervention, the system automatically records the intervention duration, actual parameter values ​​and real-time physiological signal changes of the patient, generates an intervention execution report, and stores it in the patient's treatment file for subsequent evaluation.

[0023] S6, on days 1, 3, and 7 after the intervention in S5, repeat the signal acquisition and coupling feature calculation process of S1-S2, compare the changes in the coupling feature parameter set before and after the intervention, and adjust the intervention parameters in S4 and the dosage of traditional Chinese medicine taken by the patient again based on the comparison results to form a closed-loop treatment process.

[0024] Specifically, S6 is a key step in achieving closed-loop optimization of the treatment process. By periodically repeating signal acquisition and analysis after intervention, comparing changes in coupling characteristic parameters before and after intervention, the treatment effect is evaluated, and the intervention plan and dosage of traditional Chinese medicine are adjusted. This avoids the effect decay caused by a static treatment plan and ensures that the treatment is always in a state of dynamic optimization. The treatment of chronic fatigue syndrome is a long-term process, and the patient's response to treatment will change over time. The initial intervention plan may no longer be applicable due to changes in the patient's physiological state. Through periodic evaluation and plan adjustment, S6 can promptly identify problems with poor treatment effect, optimize intervention parameters and dosage of traditional Chinese medicine, and form a complete closed loop of "acquisition-analysis-prediction-intervention-evaluation-optimization", improving the sustainability and effectiveness of treatment. In specific implementation, the S6 process is initiated on the 1st, 3rd, and 7th days after the initial intervention in S5. The evaluation on the 1st day mainly observes the short-term intervention effect: first, the signal acquisition process of S1 is repeated, with the acquisition duration still 24 hours and the sampling frequency 500Hz, and the information on the dosage of traditional Chinese medicine on that day is recorded simultaneously; then, the coupling characteristic parameter calculation process of S2 is repeated to obtain the set of coupling characteristic parameters after intervention. The post-intervention parameter set is then compared with the pre-intervention parameter set (S2 output), focusing on analyzing changes in core indicators such as coupling strength and phase synchronization (e.g., a 0.2 increase in coupling strength and a 0.15 improvement in phase synchronization after intervention). If the indicator changes are as expected (e.g., approaching the normal physiological range), the intervention parameters and TCM dosage determined in S4 are maintained. If the indicators show no significant change or worsen, the parameter adjustment phase begins. The evaluation process on days 3 and 7 is the same as on day 1, but the focus differs: day 3 focuses on analyzing the mid-term effect; if the indicators continue to improve, the intervention intensity can be appropriately reduced (e.g., the electrical stimulation intensity is reduced by 0.5 mA); day 7 conducts a phased effect evaluation, adjusting the TCM dosage (e.g., if the fatigue factor continues to decrease by 0.3, the daily dosage can be reduced by 10%) and intervention parameters (e.g., the stimulation frequency is reduced by 20 Hz) based on the parameter change trend over 7 days. The adjustment range must be determined according to the parameter change range to avoid excessive adjustments that could lead to a rebound in the condition. After each assessment and adjustment, the new intervention parameters and TCM dosages are updated into the treatment plan, forming a new closed-loop starting point to ensure that the treatment plan is dynamically optimized according to the patient's condition.

[0025] Preferably, when calculating the coupling strength in S2, the meridian-neuroelectrophysiological coupling model uses the following model formula: ,in, This represents the electrophysiological coupling strength between meridians and nerves, with a value ranging from 0 to 1; Emeritian. This represents the amplitude of the electrical signal at the meridian acupoint of the i-th acquisition point; Energy, This represents the amplitude of the neurophysiological signal at the i-th acquisition point; This represents the phase difference between the electrical signal of the meridian acupoint and the neurophysiological signal at the i-th acquisition point. This represents the total number of data collection points, with a value of 24.

[0026] Specifically, the meridian-neuroelectrophysiological coupling model accurately reflects the synergistic state of a patient's meridian and nerve functions by quantifying the correlation between different physiological signals, providing reliable characteristic evidence for subsequent fatigue factor prediction. During implementation, the total number of collection points is first determined, fixed at 24 to cover key meridian acupoints and nerve monitoring sites, ensuring data representativeness. Then, the amplitude of the meridian acupoint electrical signal and the amplitude of the neuroelectrophysiological signal at each collection point are extracted, both in microvolts. Simultaneously, the phase difference between the two signals is measured in radians. The coupling strength value is obtained by summing the product of the amplitudes of all collection points and the cosine of the phase difference, and then dividing by the product of the square root of the sum of the squares of the meridian signal amplitudes and the square root of the sum of the squares of the nerve signal amplitudes. This value is controlled between 0 and 1, where 0 indicates no coupling and 1 indicates complete coupling. The calculation results are compared with reference values ​​for the normal population; if the deviation exceeds 20%, it is marked as abnormal and serves as an important basis for subsequent intervention and adjustment.

[0027] Preferably, the fatigue factor time-series prediction algorithm uses the following model formula when predicting the fatigue factor in S3: ,in, express The fatigue factor value at any given time; This represents the initial value of the fatigue factor at time t; This represents the fatigue factor attenuation coefficient, with a value range of [value missing]. Indicates the prediction time interval This represents the coupling characteristic influence coefficient, with a value range of 0.5-2.0; This represents the normalized value of the j-th coupled feature parameter at time t; This indicates the number of coupling feature parameters, with a value of 12.

[0028] Specifically, the fatigue factor time-series prediction algorithm in S3 establishes a correlation model of fatigue factor changes over time to achieve a forward-looking judgment of future fatigue status, avoiding delays in intervention plans. During implementation, the prediction time interval is first determined to be 1 hour to ensure timely capture of changes in fatigue status; the fatigue factor decay coefficient is set between 0.01 and 0.1 per hour, adjusted according to patient age, with higher values ​​for younger patients and lower values ​​for older patients; the coupling feature influence coefficient is controlled between 0.5 and 2.0, set based on past treatment response rates, with lower values ​​for patients with good responses. In calculation, the initial fatigue factor value at the current moment is used as a base, multiplied by an exponential function (the exponent being the product of the decay coefficient and the negative value of the time interval), and then multiplied by the coupling feature influence coefficient and the sum of the normalized values ​​of all coupling feature parameters multiplied by the time interval to obtain the fatigue factor value at the future moment. Normalization requires converting the coupling feature parameters to a range of 0-1 to ensure balanced weighting of each parameter. The prediction results need to be corrected based on the patient's daily activity intensity; patients with high activity levels require an additional 5%-10% correction value.

[0029] Preferably, when adjusting the electrical stimulation intensity in S4, the chronic fatigue brain-body interaction intervention system uses the following model formula: ,in, This indicates the adjusted intensity of electrical stimulation; This represents the baseline intensity of electrical stimulation, with a value ranging from 0.1 to... This represents the fatigue factor influence coefficient, with a value range of 1.0-3.0; This represents the fatigue factor value at the current moment; This represents the fatigue factor threshold, with a value of [value missing]. This represents the maximum fatigue factor, which is 1.0.

[0030] Specifically, the adjustment logic of electrical stimulation intensity in the S4 section of the chronic fatigue brain-body interaction intervention system matches the appropriate stimulation intensity based on the patient's real-time fatigue state, avoiding insufficient stimulation or excessive damage. During implementation, the baseline electrical stimulation intensity is set at 0.1-0.5 mA, determined based on the patient's skin sensitivity, with a lower value used for sensitive patients; the fatigue factor influence coefficient is controlled between 1.0-3.0, with a larger coefficient indicating higher fatigue levels. First, the current fatigue factor value is obtained and compared with the fatigue factor threshold (fixed at 0.3) and the maximum fatigue factor value (fixed at 1.0). The ratio of the difference between the current value and the threshold to the difference between the maximum value and the threshold is calculated, multiplied by the fatigue factor influence coefficient, added to 1, and then multiplied by the baseline intensity to obtain the adjusted electrical stimulation intensity. During the calculation process, the patient's skin impedance needs to be monitored in real time. When the impedance exceeds 5 kΩ, the intensity needs to be reduced by 10%-15%. Adjustments are also made based on the intervention time; the intensity at night needs to be 20% lower than during the day to ensure that sleep quality is not affected.

[0031] Preferably, when calculating phase synchronization in S2, the meridian-neuroelectrophysiological coupling model uses the following model formula: ,in, This represents the phase synchronization index, with a value range of [value range missing]. This represents the total number of data collection points, with values ​​ranging from [value 1] to [value 2]. Represents the imaginary unit; This represents the phase of the electrical signal at the meridian acupoint of the i-th acquisition point; This represents the phase of the neurophysiological signal at the i-th acquisition point.

[0032] Specifically, the phase synchronization of the meridian-neuroelectrophysiological coupling model in S2 is calculated. By analyzing the signal phase relationship, the coordination between meridian and nerve functions is determined, providing a basis for judging the cause of fatigue. During implementation, the total number of collection points is fixed at 24, consistent with weight 2, to ensure data continuity. First, the phase of the meridian acupoint electrical signal and the phase of the neuroelectrophysiological signal at each collection point are measured, both in radians. The phase difference between the two points is calculated, and then each phase difference is substituted into an exponential function (the exponent is the product of the imaginary unit and the phase difference) to obtain the phase synchronization value for each collection point. The phase synchronization values ​​of all collection points are summed and divided by the total number of collection points to obtain the phase synchronization index, which ranges from 0 to 1, where 0 indicates complete asynchrony and 1 indicates complete synchronization. After calculation, statistics are collected over time periods. If the synchronization index is below 0.4 for three consecutive hours, it is marked as meridian-neuroelectrophysiological coordination abnormality, and targeted acupoint stimulation is added in subsequent interventions, with the stimulation duration increased by 10 minutes compared to the conventional approach.

[0033] Preferably, when the fatigue factor time-series prediction algorithm corrects the prediction curve in S3, it adopts the following model formula: ,in, This represents the corrected predicted fatigue factor value; This represents the predicted fatigue factor value before correction; This represents the effect coefficient of traditional Chinese medicine dosage, with a value range of [value range missing]. Indicates the patient's current dosage of traditional Chinese medicine. This indicates the average dosage of traditional Chinese medicine taken by the patient over the past 3 months.

[0034] Specifically, in the correction phase of the fatigue factor time-series prediction algorithm in S3, the prediction results are optimized by incorporating the patient's TCM (Traditional Chinese Medicine) usage to improve accuracy. During implementation, the TCM dosage influence coefficient is set between 0.2 and 0.8, adjusted according to the TCM components; higher values ​​are used for TCM containing Qi-tonifying components, and lower values ​​are used for those containing heat-clearing components. First, the patient's current TCM dosage is obtained and compared with the average dosage over the past three months. The proportion of the difference between the current and average dosages is calculated, multiplied by the dosage influence coefficient, and then added to 1. This is then multiplied by the original fatigue factor prediction value to obtain the corrected prediction value. During calculation, it is necessary to ensure that the average dosage statistical period is no less than 90 days to avoid the impact of short-term fluctuations. If the patient has recently adjusted their TCM prescription, the correction margin needs to be increased by 10%-15%. Simultaneously, the patient's liver and kidney function indicators are considered; for patients with renal insufficiency, the coefficient needs to be reduced by 20%-25% to prevent slow drug metabolism from causing prediction bias. The corrected result needs to be compared with the actual monitored fatigue score; if the error exceeds 10%, the coefficient is readjusted.

[0035] Preferably, step S2 includes the following sub-steps: S21, performing signal filtering on the meridian acupoint electrical signals, EEG signals, and peripheral nerve electrophysiological signals collected in S1, using a bandpass filter to retain signal components in the 5-50Hz frequency band, filtering out power frequency interference and baseline drift signals, setting the filter order to 8th order and the transition band width to 2Hz during the filtering process; S22, aligning the three types of filtered signals according to the acquisition timestamps to ensure that the three types of signal data at the same time point correspond one-to-one, with the timestamp error controlled within ±1ms, and forming a multi-dimensional signal matrix after alignment, where the number of rows in the matrix corresponds to the number of sampling points corresponding to the total acquisition time, and the number of columns... S23: Input the multi-dimensional signal matrix into the meridian-neuro-electrophysiological coupling model, call the built-in feature extraction module of the model, and calculate the coupling strength, phase synchronization and frequency response features of each row of data in the matrix point by point. The sliding window method is used in the calculation process, with the window size set to 100 sampling points and the window step size set to 50 sampling points. S24: Arrange the feature values ​​obtained by point by point in chronological order, remove outliers to form a coupling feature parameter set. The outlier judgment criterion is a value that deviates from the mean by 3 times the standard deviation. After removing outliers, linear interpolation is performed on the parameter set to complete it, ensuring that the parameter set data is continuous and without missing data.

[0036] Specifically, the implementation process of S2 involves multi-step signal processing and model calculation to transform the raw physiological signals into an effective set of coupled feature parameters, providing accurate input for subsequent prediction and intervention. During implementation, the three signals collected in S1 are first bandpass filtered with an 8th-order filter and a transition band width of 2Hz, retaining signals in the 5-50Hz frequency band to filter out power frequency interference and baseline drift. Then, the signals are aligned according to the acquisition timestamps, with the timestamp error controlled within ±1ms, forming a multi-dimensional signal matrix with rows equal to the number of sampling points and columns equal to 3. The matrix is ​​then input into the coupling model, and the feature extraction module uses a sliding window method to calculate features. The window size is 100 sampling points, and the step size is 50 sampling points, analyzing the coupling strength, phase synchronization, and frequency response features point by point. Finally, the feature values ​​are arranged in chronological order, outliers deviating from the mean by three times the standard deviation are removed, and the data is completed through linear interpolation, forming a complete set of coupled feature parameters to ensure that the parameters accurately reflect the correlation between meridians and neurophysiology.

[0037] Preferably, step S3 includes the following sub-steps: S31, extracting historical physiological signal data from the patient's medical record database for the past 3 months. This data includes daily meridian-neuroelectrophysiological coupling characteristic parameters and corresponding measured values ​​of fatigue factors. The data sampling frequency is consistent with S1, both being 500Hz, and the data duration is no less than 90 days; S32, dividing the extracted historical data into a training set and a validation set in a 7:3 ratio. The training set is used for model training of the fatigue factor time-series prediction algorithm, and the validation set is used for verifying the algorithm's prediction accuracy. The division process uses random sampling to ensure uniform sample distribution; S33. The coupled feature parameter set output from S2 and the training set data are input into the fatigue factor time series prediction algorithm. The weight parameters in the algorithm are adjusted by gradient descent to minimize the error between the algorithm's predicted value and the measured fatigue factor value in the training set. The number of training iterations is set to 1000, and the learning rate is set to 0.001. In S34, the trained algorithm is tested using validation set data. The mean square error between the predicted value and the measured value is calculated. If the mean square error is less than 0.01, the algorithm is called to predict the fatigue factor change trend for the next 7 days. If the mean square error is greater than or equal to 0.01, the amount of training set data is increased and the algorithm is retrained.

[0038] Specifically, the implementation process of S3 aims to train and validate the fatigue factor time-series prediction algorithm using historical data, ensuring the reliability of the prediction results to guide adjustments to the intervention plan. During implementation, historical physiological signal data from the past three months are extracted from the patient's medical record database, including daily coupled feature parameters and measured fatigue factor values, with a sampling frequency of 500Hz and a data duration of at least 90 days. Then, the training and validation sets are randomly divided at a 7:3 ratio to ensure uniform sample distribution. The coupled feature parameter set from S2 and the training set are input into the algorithm, and the weight parameters are adjusted using gradient descent. The iteration count is set to 1000, and the learning rate to 0.001, minimizing the error between the predicted and measured values. Subsequently, the algorithm is tested using the validation set, and the mean squared error is calculated. If it is less than 0.01, the algorithm is used to predict the fatigue factor trend for the next 7 days; if it is greater than or equal to 0.01, the training set data is increased for retraining, ensuring that the algorithm's accuracy meets clinical needs and avoiding inappropriate intervention plans due to prediction bias.

[0039] Preferably, step S4 includes the following sub-steps: S41, performing feature analysis on the fatigue factor time-series prediction curve output by S3, extracting the peak value, trough value, and slope of change in the curve. The peak value corresponds to the time when the patient's fatigue level is highest each day, the trough value corresponds to the time when the fatigue level is lowest, and the slope of change reflects the rate of change in fatigue level. The analysis process uses numerical differentiation to calculate the curve slope; S42, determining the intervention period of the chronic fatigue brain-body interaction intervention system based on the curve feature analysis results, setting the intervention period 1 hour before the peak value of the fatigue factor to avoid intervention during the trough value period, and determining the number of daily interventions based on the number of peak values. If there are 2 peak values ​​per day, then 2 interventions are set; S43, combining... The patient's current TCM treatment plan includes dosage, frequency, and drug components. Different drug components correspond to different intervention parameter adjustment coefficients. For example, TCM containing ginseng has a coefficient of 1.2, and TCM containing astragalus has a coefficient of 1.1. S44: Based on the fatigue factor curve characteristics and TCM component coefficients, the electrical stimulation intensity, stimulation frequency, intervention duration, and intervention acupoint combination of the intervention system are adjusted. The electrical stimulation intensity is linearly adjusted according to the difference between the peak and trough values ​​of the fatigue factor. The stimulation frequency is segmented according to the slope of the curve. When the absolute value of the slope is greater than 0.05, the frequency is set to 50-100Hz, and when the absolute value of the slope is less than or equal to 0.05, the frequency is set to 1-50Hz.

[0040] Specifically, the core of the S4 intervention parameter adjustment process is to personalize the parameters of the chronic fatigue brain-body interaction intervention system by combining the fatigue factor prediction curve with the traditional Chinese medicine (TCM) treatment plan. During implementation, the S3 prediction curve is first analyzed using numerical differentiation to extract the peak value (the moment of highest fatigue), trough value (the moment of lowest fatigue), and slope (the rate of change in fatigue). Then, the intervention period is determined based on the curve characteristics, set to one hour before the peak value, and the number of interventions is determined according to the number of daily peak values ​​(e.g., twice daily for two peak values). Subsequently, the patient's TCM dosage, frequency, and components are obtained, and adjustment coefficients are set according to the components (e.g., coefficient 1.2 for ginseng and 1.1 for astragalus). Finally, the parameters are adjusted based on the curve characteristics and TCM coefficients. The electrical stimulation intensity is adjusted linearly according to the difference between the fatigue factor peak and trough values, and the stimulation frequency is adjusted segmentally according to the slope (50-100Hz for slope absolute value > 0.05, 1-50Hz for slope ≤ 0.05). The intervention duration and acupoint combination are set in accordance with clinical standards to ensure a high degree of matching between the parameters and the patient's condition.

[0041] Preferably, step S5 includes the following sub-steps: S51, the intervention parameters adjusted in S4 are input into the control module of the chronic fatigue brain-body interaction intervention system. The control module analyzes the parameters, converting the electrical stimulation intensity and stimulation frequency parameters into hardware drive signals, and the intervention duration and intervention acupoint combination parameters into timing control signals. During the analysis process, the parameter range is verified to ensure that the parameters are within the safe threshold. S52, the EEG modulation module of the intervention system generates a corresponding EEG stimulation signal. This signal is a sine wave signal, the signal amplitude is determined by the electrical stimulation intensity parameter, and the signal frequency is determined by the stimulation frequency parameter. The EEG modulation module applies the stimulation signal to the patient's prefrontal cortex region through electrode pads. The electrode pads are made of silver-silver chloride material and have impedance control. The current is controlled below 5kΩ; S53, the meridian acupoint stimulation module of the intervention system generates a meridian electrical stimulation signal. This signal is a square wave signal with a duty cycle set to 50%. The combination of stimulation acupoints is selected according to the parameters determined in S4, usually including Zusanli, Sanyinjiao, and Hegu acupoints. The stimulation time for each acupoint is allocated to 10 minutes, and the acupoint stimulation sequence is from the lower limb to the upper limb; S54, the intervention is implemented by alternating between EEG regulation intervention and meridian acupoint electrical stimulation intervention. First, 15 minutes of EEG regulation intervention is performed, followed by 15 minutes of meridian acupoint electrical stimulation intervention. The total duration of a single intervention is 30 minutes. The patient's physiological signals are monitored in real time during the intervention. If abnormal signals occur, the intervention is paused and resumed after the signals return to normal.

[0042] Specifically, the core of the S5 intervention implementation process is to translate the parameters adjusted in S4 into actual intervention operations. This is achieved through alternating EEG modulation and meridian acupoint stimulation, ensuring precise intervention and safety. During implementation, the intervention parameters from S4 are first input into the intervention system control module. The module analyzes the parameters, converting the electrical stimulation intensity and frequency into hardware drive signals, and the intervention duration and acupoint combination into timing control signals. Simultaneously, it verifies that the parameters are within safe thresholds (electrical stimulation intensity 0.1-5mA, frequency 1-100Hz). Then, the EEG modulation module is activated, generating a sinusoidal stimulation signal applied to the prefrontal cortex via silver-silver chloride electrodes, with the electrode impedance controlled below 5kΩ. Next, the meridian acupoint stimulation module is activated, generating a square wave signal with a 50% duty cycle applied to preset acupoints. Each acupoint is stimulated for 10 minutes, proceeding from the lower limb to the upper limb. The two interventions alternate for 15 minutes each to complete a single 30-minute intervention. Physiological signals are monitored in real time during the process. If any abnormalities occur, the intervention is paused and resumed after the signal recovers, ensuring safety and compliance with parameter settings.

[0043] The meridian-neuroelectrophysiological coupling model in this invention is a core tool for analyzing the correlation characteristics between meridian and neurophysiological signals in patients with chronic fatigue syndrome. Specifically, it integrates the patient's meridian acupoint electrical signals, electroencephalogram (EEG) signals, and peripheral nerve electrophysiological signals to uncover the intrinsic correlation among these three. In terms of implementation, the three raw physiological signals collected in step S1 are first received. The signals are then bandpass filtered (8th order filter, 2Hz transition band width, retaining the 5-50Hz frequency band) and aligned with the timestamps (error ±1ms) to form a multi-dimensional signal matrix. Next, the built-in feature extraction module is invoked, and a sliding window method (100 sampling points per window, 50 sampling points per step) is used to calculate the coupling strength, phase synchronization, and frequency response characteristics point by point. The coupling strength is calculated by correlating the product of signal amplitudes with the phase difference (result range 0-1), and the phase synchronization is determined by analyzing the phase difference exponential function (result range 0-1). Finally, outliers deviating from the mean by three times the standard deviation are removed, and the result is completed by linear interpolation to form a set of coupling feature parameters. The model transforms raw physiological signals into quantifiable feature indicators, providing objective input for subsequent fatigue factor prediction. Its significance lies in breaking the limitations of single signal analysis, accurately capturing the synergistic state of the patient's meridians and nerve functions, providing a scientific basis for judging the degree of fatigue and formulating intervention plans, and avoiding the bias of relying on subjective symptom assessment.

[0044] The fatigue factor time-series prediction algorithm in this invention is an algorithmic tool used to prospectively judge the future fatigue status trend of patients. Specifically, it is an algorithm that predicts the evolution of fatigue factors over time based on the patient's historical physiological data and current coupled feature parameters. In terms of implementation, firstly, historical physiological signal data of the patient over the past 3 months (including daily coupled feature parameters and measured values ​​of fatigue factors, sampling frequency 500Hz, duration not less than 90 days) is extracted from the medical record database and randomly divided into training set and validation set in a 7:3 ratio; then, the coupled feature parameter set output from step S2 and the training set are input into the algorithm, and the weight parameters are adjusted by gradient descent method (1000 iterations, learning rate 0.001) to minimize the error between the predicted value and the measured value; then, the algorithm is tested with the validation set. If the mean square error is less than 0.01, the fatigue factor change in the next 7 days (1-hour time interval) is predicted based on the current coupled feature parameters, and a time-series prediction curve is generated. If the error does not meet the standard, the amount of training set data is increased and the algorithm is retrained; at the same time, the prediction results can be optimized by combining the dosage of traditional Chinese medicine (corrected by a dosage influence coefficient of 0.2-0.8). The algorithm aims to predict the evolution of a patient's fatigue state in advance, providing a forward-looking basis for adjusting intervention plans. Its significance lies in avoiding the lag caused by treatment plans being based solely on the current state, helping doctors to identify the risk of fatigue worsening in advance, optimize strategies in a timely manner, improve the initiative and precision of treatment, and reduce the possibility of further development of fatigue symptoms.

[0045] The chronic fatigue brain-body interactive intervention system of this invention is a hardware operating platform for realizing precise intervention by combining traditional Chinese medicine with modern physiotherapy. Specifically, it is an integrated intervention device that combines fatigue prediction results with traditional Chinese medicine solutions to implement electroencephalogram (EEG) regulation and meridian acupoint electrical stimulation. In terms of implementation, it first receives the intervention parameters adjusted in step S4 (including electrical stimulation intensity 0.1-5mA, frequency 1-100Hz, duration 30 minutes, acupoint combination, etc.). The control module analyzes the parameters, converts the electrical stimulation-related parameters into hardware drive signals, converts the timing and acupoint parameters into control signals, and verifies whether the parameters are within the safe threshold. Then, the EEG regulation module is activated to generate a sinusoidal stimulation signal, which is applied to the patient's frontal lobe through silver-silver chloride electrode pads (impedance <5kΩ). At the same time, the meridian acupoint stimulation module is activated to generate a square wave signal with a duty cycle of 50%, which is applied to preset acupoints in the order from lower limb to upper limb (10 minutes of stimulation for each acupoint). The two interventions are performed alternately (15 minutes each), and the patient's physiological signals are monitored in real time during the process. If an abnormality occurs, the process is paused and resumed after the signal recovers. The system's function is to transform personalized intervention plans into actual treatment operations, enabling the precise application of modern physiotherapy techniques. Its significance lies in its synergy with traditional Chinese medicine (long-term conditioning with traditional Chinese medicine + rapid intervention by the system) to compensate for the shortcomings of single treatments. At the same time, it relies on dynamic parameter adjustments to adapt to individual patient differences, ensuring the safety and effectiveness of interventions. This promotes the transformation of chronic fatigue syndrome treatment from "generalized" to "personalized," improving overall treatment effectiveness and patient compliance.

[0046] The clinical application of traditional Chinese medicine combined with modern physiotherapy techniques in the treatment of chronic fatigue syndrome has a primary advantage: it achieves in-depth integrated analysis of multi-dimensional physiological signals, accurately overcoming the shortcomings of existing technologies that only collect single physiological signals and cannot capture complex physiological changes. It continuously collects patients' meridian acupoint electrical signals, electroencephalogram (EEG) signals, and peripheral nerve electrophysiological signals for 24 hours, and then uses a meridian-neurophysiological coupling model to systematically analyze these signals, uncovering the coupling strength, phase synchronization, and frequency response characteristics between different signals, forming a comprehensive set of coupling characteristic parameters. This multi-signal integrated analysis method breaks through the limitations of single-signal acquisition, more accurately reflecting the correlation between the patient's physical state and fatigue level, providing a scientific and comprehensive basis for adjusting subsequent treatment plans, and effectively solving the problem of insufficient evidence in existing technologies.

[0047] Secondly, this application possesses dynamic and personalized treatment adjustment capabilities, overcoming the shortcomings of existing technologies that lack dynamic adjustment mechanisms and cannot adapt to changes in the patient's condition. Based on the feature parameter set output by the meridian-neuro-electrophysiological coupling model, it calls a fatigue factor time-series prediction algorithm, combined with the patient's historical physiological signal data, to predict the trend of fatigue factor changes over a future period, generating a clear fatigue factor time-series prediction curve. Then, based on the prediction curve and the patient's traditional Chinese medicine treatment plan, it specifically adjusts parameters such as the electrical stimulation intensity, stimulation frequency, intervention duration, and acupoint combination of the chronic fatigue brain-body interaction intervention system, ensuring that the intervention plan matches the patient's real-time condition. Furthermore, the signal acquisition and analysis process is repeated periodically after intervention, and parameters are optimized again based on the comparison results, forming a complete closed loop, ensuring that the treatment plan always adapts to changes in the patient's condition and improving treatment accuracy.

[0048] Finally, this application achieves efficient synergy between traditional Chinese medicine (TCM) and modern physiotherapy, fully leveraging the advantages of both treatment methods and further compensating for the shortcomings of existing single-treatment technologies. TCM can regulate the patient's overall physical condition, exerting a long-term conditioning effect and improving basic bodily functions; modern physiotherapy, through a chronic fatigue brain-body interaction intervention system, directly acts on meridians, acupoints, and the brain via electrical stimulation, quickly relieving fatigue symptoms and achieving short-term intervention effects. The combination of the two solves the problem of slow onset of action of TCM alone and compensates for the deficiency of modern physiotherapy alone in fundamentally regulating the body, forming a synergistic treatment model of "long-term conditioning + rapid intervention." This significantly improves the systematization level of treatment. Furthermore, because the treatment plan can be dynamically adjusted according to the individual patient's signal characteristics and changes in condition, it greatly enhances the personalization of treatment, ultimately effectively improving treatment outcomes and patient compliance.

[0049] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," "link," and "fix" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0050] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. The clinical application of traditional Chinese medicine combined with modern physiotherapy techniques in the treatment of chronic fatigue syndrome, characterized in that, Includes the following steps: S1: A multi-channel physiological signal acquisition device continuously collects meridian acupoint electrical signals, electroencephalogram (EEG) signals, and peripheral nerve electrophysiological signals from patients with chronic fatigue syndrome. The acquisition duration is set to 24 hours, and the sampling frequency is adjusted to 500Hz. Simultaneously, the daily dosage and frequency of traditional Chinese medicine (TCM) administration and the duration of physiotherapy intervention are recorded. S2: The meridian acupoint electrical signals, EEG signals, and peripheral nerve electrophysiological signals collected in S1 are input into a meridian-neurophysiological coupling model. The model calculates the coupling strength, phase synchronization, and frequency response characteristics between different signals, outputting a set of coupling feature parameters containing 12 feature indicators. S3: Based on the coupling feature parameter set output in S2, a fatigue factor time-series prediction algorithm is invoked. Using the patient's physiological signal history data from the past 3 months as training samples, the algorithm predicts the trend of fatigue factor changes in the patient over the next 7 days, outputting a fatigue factor time-series prediction curve. The curve time interval is set to 1 second. S1: Hour; S4: Based on the fatigue factor time-series prediction curve output from S3, and combined with the patient's traditional Chinese medicine (TCM) medication regimen, adjust the intervention parameters of the chronic fatigue brain-body interaction intervention system. The intervention parameters include electrical stimulation intensity, stimulation frequency, intervention duration, and intervention acupoint combination. The electrical stimulation intensity adjustment range is 0.1-5mA, and the stimulation frequency adjustment range is 1-100Hz; S5: Using the chronic fatigue brain-body interaction intervention system, implement EEG modulation intervention and meridian acupoint electrical stimulation intervention on the patient according to the intervention parameters adjusted in S4. The two intervention methods are alternated. The duration of a single intervention is set to 30 minutes, and the number of interventions per day is set to 2; S6: On the 1st, 3rd, and 7th days after the intervention in S5, repeat the signal acquisition and coupling feature calculation process of S1-S2, compare the changes in the coupling feature parameter set before and after the intervention, and adjust the intervention parameters in S4 and the patient's TCM medication dosage again based on the comparison results to form a closed-loop treatment process.

2. The clinical application of traditional Chinese medicine combined with modern physiotherapy techniques for treating chronic fatigue syndrome according to claim 1, characterized in that, When calculating the coupling strength in S2, the meridian-neuroelectrophysiological coupling model uses the following model formula: ,in, This represents the electrophysiological coupling strength between meridians and nerves, with a value ranging from 0 to 1; Emeritian. This represents the amplitude of the electrical signal at the meridian acupoint of the i-th acquisition point; Energy, This represents the amplitude of the neurophysiological signal at the i-th acquisition point; This represents the phase difference between the electrical signal of the meridian acupoint and the neurophysiological signal at the i-th acquisition point. This represents the total number of data collection points, with a value of 24.

3. The clinical application of traditional Chinese medicine combined with modern physiotherapy techniques in the treatment of chronic fatigue syndrome according to claim 1, characterized in that, The fatigue factor time-series prediction algorithm uses the following model formula when predicting the fatigue factor in S3: ,in, express The fatigue factor value at any given time; This represents the initial value of the fatigue factor at time t; This represents the fatigue factor attenuation coefficient, with a value range of [value missing]. Indicates the prediction time interval This represents the coupling characteristic influence coefficient, with a value range of 0.5-2.0; This represents the normalized value of the j-th coupled feature parameter at time t; This indicates the number of coupling feature parameters, with a value of 12.

4. The clinical application of traditional Chinese medicine combined with modern physiotherapy techniques in the treatment of chronic fatigue syndrome according to claim 1, characterized in that, When adjusting the electrical stimulation intensity in S4, the chronic fatigue brain-body interaction intervention system uses the following model formula: ,in, This indicates the adjusted intensity of electrical stimulation; This represents the baseline intensity of electrical stimulation, with a value ranging from 0.1 to... This represents the fatigue factor influence coefficient, with a value range of 1.0-3.0; This represents the fatigue factor value at the current moment; This represents the fatigue factor threshold, with a value of [value missing]. This represents the maximum fatigue factor, which is 1.

0.

5. The clinical application of traditional Chinese medicine combined with modern physiotherapy techniques in the treatment of chronic fatigue syndrome according to claim 1, characterized in that, When calculating phase synchronization in S2, the meridian-neuro-electrophysiological coupling model uses the following model formula: ,in, This represents the phase synchronization index, with a value range of [value range missing]. This represents the total number of data collection points, with values ​​ranging from [value 1] to [value 2]. Represents the imaginary unit; This represents the phase of the electrical signal at the meridian acupoint of the i-th acquisition point; This represents the phase of the neurophysiological signal at the i-th acquisition point.

6. The clinical application of traditional Chinese medicine combined with modern physiotherapy techniques in the treatment of chronic fatigue syndrome according to claim 1, characterized in that, When the fatigue factor time-series prediction algorithm corrects the prediction curve in S3, it uses the following model formula: ,in, This represents the corrected predicted fatigue factor value; This represents the predicted fatigue factor value before correction; This represents the effect coefficient of traditional Chinese medicine dosage, with a value range of [value range missing]. Indicates the patient's current dosage of traditional Chinese medicine. This indicates the average dosage of traditional Chinese medicine taken by the patient over the past 3 months.

7. The clinical application of traditional Chinese medicine combined with modern physiotherapy techniques in the treatment of chronic fatigue syndrome according to claim 1, characterized in that, S2 includes the following steps: S21, performing signal filtering on the meridian acupoint electrical signals, electroencephalogram (EEG) signals, and peripheral nerve electrophysiological signals acquired in S1. A bandpass filter is used to retain signal components in the 5-50Hz frequency band, while filtering out power frequency interference and baseline drift signals. During the filtering process, the filter order is set to 8th order, and the transition band width is 2Hz. S22, aligning the three types of filtered signals according to the acquisition timestamps to ensure that the three types of signal data at the same time point correspond one-to-one, with the timestamp error controlled within ±1ms. After alignment, a multi-dimensional signal matrix is ​​formed, with the number of rows corresponding to the number of sampling points corresponding to the total acquisition time, and the number of columns being 3. S23, input the multi-dimensional signal matrix into the meridian-neuro-electrophysiological coupling model, call the built-in feature extraction module of the model, and calculate the coupling strength, phase synchronization and frequency response features of each row of data in the matrix point by point. The sliding window method is used in the calculation process, with the window size set to 100 sampling points and the window step size set to 50 sampling points. S24, arrange the feature values ​​obtained by point by point in chronological order, remove outliers to form a coupling feature parameter set. The outlier judgment criterion is a value that deviates from the mean by 3 times the standard deviation. After removing outliers, linear interpolation is performed on the parameter set to complete it, ensuring that the parameter set data is continuous and without missing data.

8. The clinical application of traditional Chinese medicine combined with modern physiotherapy techniques for treating chronic fatigue syndrome according to claim 1, characterized in that, S3 includes the following steps: S31, extracting historical physiological signal data from the patient's medical record database for the past 3 months. This data includes daily meridian-neuroelectrophysiological coupling characteristic parameters and corresponding measured values ​​of fatigue factors. The data sampling frequency is consistent with S1, both being 500Hz, and the data duration is no less than 90 days; S32, dividing the extracted historical data into a training set and a validation set in a 7:3 ratio. The training set is used for model training of the fatigue factor time-series prediction algorithm, and the validation set is used for verifying the algorithm's prediction accuracy. The division process uses random sampling to ensure uniform sample distribution; S33, ...

2. The output set of coupled feature parameters and the training set data are input into the fatigue factor time series prediction algorithm. The weight parameters in the algorithm are adjusted by gradient descent to minimize the error between the algorithm's predicted value and the measured value of fatigue factor in the training set. The number of training iterations is set to 1000 and the learning rate is set to 0.

001. S34. The trained algorithm is tested using the validation set data. The mean square error between the predicted value and the measured value is calculated. If the mean square error is less than 0.01, the algorithm is called to predict the trend of fatigue factor change in the next 7 days. If the mean square error is greater than or equal to 0.01, the amount of training set data is increased and the algorithm is retrained.

9. The clinical application of traditional Chinese medicine combined with modern physiotherapy techniques for treating chronic fatigue syndrome according to claim 1, characterized in that, S4 includes the following steps: S41, performing feature analysis on the fatigue factor time-series prediction curve output by S3, extracting the peak value, trough value and slope of change in the curve, the peak value corresponds to the time when the patient's fatigue level is the highest each day, the trough value corresponds to the time when the fatigue level is the lowest, and the slope of change reflects the rate of change of fatigue level. The analysis process uses numerical differentiation to calculate the slope of the curve. S42. Based on the curve characteristic analysis results, determine the intervention period of the chronic fatigue brain-body interaction intervention system. Set the intervention period 1 hour before the peak of the fatigue factor to avoid intervention during the trough period. The number of daily interventions is determined based on the number of peak occurrences. If there are 2 peaks per day, set 2 interventions. S43. Combine the patient's current TCM treatment plan to obtain the dosage, frequency, and drug components. Different drug components correspond to different intervention parameter adjustment coefficients. For example, TCM containing ginseng corresponds to a coefficient of 1.2, and TCM containing astragalus corresponds to a coefficient of 1.

1. S44. Based on the fatigue factor curve characteristics and TCM component coefficients, adjust the electrical stimulation intensity, stimulation frequency, intervention duration, and intervention acupoint combination of the intervention system. The electrical stimulation intensity is linearly adjusted according to the difference between the peak and trough of the fatigue factor. The stimulation frequency is segmented according to the slope of the curve. When the absolute value of the slope is greater than 0.05, the frequency is set to 50-100Hz. When the absolute value of the slope is less than or equal to 0.05, the frequency is set to 1-50Hz.

10. The clinical application of traditional Chinese medicine combined with modern physiotherapy techniques in the treatment of chronic fatigue syndrome according to claim 1, characterized in that, S5 includes the following steps: S51, the intervention parameters adjusted in S4 are input into the control module of the chronic fatigue brain-body interaction intervention system. The control module analyzes the parameters, converting the electrical stimulation intensity and stimulation frequency parameters into hardware drive signals, and the intervention duration and intervention acupoint combination parameters into timing control signals. During the analysis process, the parameter range is verified to ensure that the parameters are within a safe threshold. S52, the EEG modulation module of the intervention system generates a corresponding EEG stimulation signal. This signal is a sine wave signal, with the signal amplitude determined by the electrical stimulation intensity parameter and the signal frequency determined by the stimulation frequency parameter. The EEG modulation module applies the stimulation signal to the patient's prefrontal cortex region through electrode pads. The electrode pads are made of silver-silver chloride material, and the impedance is controlled within a certain range. Below 5kΩ; S53, generate meridian electrical stimulation signals through the meridian acupoint stimulation module of the intervention system. The signal is a square wave signal with a duty cycle set to 50%. The combination of stimulation acupoints is selected according to the parameters determined in S4, usually including Zusanli, Sanyinjiao, and Hegu acupoints. The stimulation time for each acupoint is allocated to 10 minutes, and the acupoint stimulation sequence is from the lower limb to the upper limb; S54, implement the intervention by alternating between EEG regulation intervention and meridian acupoint electrical stimulation intervention. First, perform 15 minutes of EEG regulation intervention, then perform 15 minutes of meridian acupoint electrical stimulation intervention. The total duration of a single intervention is 30 minutes. The patient's physiological signals are monitored in real time during the intervention. If abnormal signals occur, the intervention is paused and resumed after the signals return to normal.

Images