Intelligent electric pulse physiotherapy control method based on AI adaptive feedback

Abstract

This invention discloses an intelligent electro-pulse therapy control method based on AI adaptive feedback, relating to the field of electro-pulse therapy control. The method includes the following steps: collecting biofeedback signals from the therapy user and simultaneously collecting the user's subjective feedback data; converting the subjective feedback data into effective subjective data and performing preliminary analysis of the feedback analysis data between different biofeedback signals and subjective feedback data; deeply analyzing the influence of different biofeedback signals on the subjective feedback data, and simultaneously constructing historical reference analysis schemes and predictive reference analysis schemes to calibrate the impact on the feedback analysis data, obtaining feedback reference data; and intelligently controlling the intelligent electro-pulse therapy based on the feedback reference data. This invention addresses the problem that existing electro-pulse therapy control technologies lack intelligent and accurate acquisition of pain perception, resulting in an inability to adjust the intensity of the electro-pulse in a timely and accurate manner.

Description

Technical Field

[0001] This invention relates to the field of electro-pulse therapy control technology, specifically to an intelligent electro-pulse therapy control method based on AI adaptive feedback. Background Technology

[0002] Electro-pulse therapy control technology refers to a type of medical and rehabilitation engineering technology that uses electronic devices to generate electrical pulses with specific parameters, apply them to the surface of the human body or specific parts, and achieve the expected physiological effects (such as analgesia, muscle activation, circulation promotion, and nerve regulation) through precise control and dynamic adjustment of these pulse parameters.

[0003] When adjusting the intensity of electrical pulses, electro-pulse therapy control technology typically requires reference to the user's pain perception. However, existing technologies often rely on the user's pain perception to adjust the pulse intensity, lacking intelligence. Alternatively, they may predict pain levels based on single data points, resulting in insufficient accuracy. For instance, patent application CN107308544A discloses a "physiotherapy device and its control method," but this method lacks a clear description of the pain threshold. Without detailed explanation, it usually requires user feedback on pain levels, necessitating full user cooperation and demonstrating insufficient intelligence. Therefore, existing electro-pulse therapy control technologies suffer from insufficient intelligence and accuracy in acquiring pain perception, leading to delays and inaccurate adjustments to pulse intensity. Summary of the Invention

[0004] This invention aims to at least partially address one of the technical problems in the prior art. It collects biofeedback signals from physiotherapy users, simultaneously collecting their subjective feedback data. The subjective feedback data is then converted into effective subjective data. Based on the biofeedback signals and effective subjective data, feedback analysis is performed on the interaction between different biofeedback signals and subjective feedback data. Pain feedback reference data is extracted from the resistive feedback analysis graph based on the user's real-time biofeedback signals. A historical reference analysis scheme is then constructed. Based on the pain feedback reference data, the user's electromyographic feedback analysis points are calibrated, and the user's pain perception is predicted to obtain feedback reference data. Alternatively, a predictive reference analysis scheme is constructed, calibrating the user's electromyographic feedback analysis points based on the pain feedback reference data and predicting the user's pain perception to obtain feedback reference data. Finally, intelligent control of intelligent electrical pulse physiotherapy is performed based on the feedback reference data. This addresses the problem that existing electrical pulse physiotherapy control technologies are not intelligent or accurate enough in acquiring pain perception, resulting in the inability to adjust the electrical pulse intensity in a timely and accurate manner.

[0005] To achieve the above objectives, this application provides an intelligent electro-pulse therapy control method based on AI adaptive feedback, comprising the following steps:

[0006] Collect biofeedback signals from physiotherapy users, and simultaneously collect subjective feedback data from users;

[0007] The subjective feedback data was converted into effective subjective data, and the feedback analysis data between different biological feedback signals and subjective feedback data was preliminarily analyzed.

[0008] We conducted an in-depth analysis of the impact of different biofeedback signals on subjective feedback data, and constructed historical reference analysis schemes and predictive reference analysis schemes to calibrate the feedback analysis data to obtain feedback reference data.

[0009] Intelligent control of intelligent electro-pulse therapy is achieved based on feedback reference data.

[0010] Furthermore, biofeedback signals from physiotherapy users are collected, along with subjective feedback data. The biofeedback signals include electromyography (EMG) values ​​and skin resistance, while the subjective feedback data includes pain perception, which includes no sensation, sensation, strong sensation, and pain sensation.

[0011] Furthermore, the process of converting subjective feedback data into effective subjective data and conducting preliminary analysis of the feedback between different biofeedback signals and subjective feedback data includes the following sub-steps:

[0012] Transform subjective feedback data into effective subjective data;

[0013] Feedback analysis data is based on the interaction between different biofeedback signals and subjective feedback data, using both biofeedback signals and effective subjective data analysis.

[0014] Furthermore, the subjective feedback data is converted into effective subjective data specifically by assigning values ​​to the pain perception of no sensation, sensation, strong sensation, and pain sensation, respectively, with values ​​of 1, 2, 3, and 4.

[0015] Furthermore, the feedback analysis data based on biofeedback signals and effective subjective data analysis between different biofeedback signals and subjective feedback data includes the following sub-steps:

[0016] Two-dimensional coordinate systems were established with electromyography (EMG) and skin resistance as the X-axis and pain perception as the Y-axis, respectively, and named EMG feedback analysis diagram and resistance feedback analysis diagram.

[0017] The biofeedback signals were entered into the electromyography feedback analysis chart and the electrical resistance feedback analysis chart according to the subjective feedback data.

[0018] The coordinate points in the electromyography feedback analysis diagram and the resistance feedback analysis diagram are named the electromyography feedback analysis point and the resistance feedback analysis point, respectively.

[0019] The electromyographic feedback analysis points and the electrical resistance feedback analysis points are the feedback analysis data.

[0020] Furthermore, the impact of different biofeedback signals on subjective feedback data is analyzed in depth. Simultaneously, historical reference analysis schemes and predictive reference analysis schemes are constructed to calibrate the feedback analysis data, resulting in feedback reference data. This process includes the following sub-steps:

[0021] Pain feedback reference data is extracted from the resistive feedback analysis graph based on real-time biofeedback signals from users.

[0022] A historical reference analysis scheme is constructed, which calibrates the user's electromyographic feedback analysis points based on pain feedback reference data and predicts the user's pain perception to obtain feedback reference data;

[0023] A predictive reference analysis scheme is constructed to calibrate the user's electromyographic feedback analysis points based on pain feedback reference data and predict the user's pain perception, thereby obtaining feedback reference data.

[0024] Furthermore, extracting pain feedback reference data from the resistive feedback analysis graph based on the user's real-time biofeedback signal includes the following sub-steps:

[0025] The user's real-time biofeedback signal is acquired and named the real-time monitoring feedback signal. The electromyography value and skin resistance in the real-time monitoring feedback signal are named the reference electromyography and reference resistance, respectively.

[0026] Perform function fitting on the resistance feedback analysis points, and name the curve corresponding to the function obtained by function fitting the resistance feedback curve.

[0027] The electromyography feedback analysis point where the X-axis equals the reference electromyography is named the effective reference point of electromyography. The skin resistance in the biofeedback signal to which the effective reference point of electromyography belongs is obtained and named the auxiliary resistance. The coordinate point in the resistance feedback curve where the X-axis equals the auxiliary resistance is named the auxiliary reference real point.

[0028] Obtain the minimum and maximum values ​​of X in the resistance feedback curve and label them as Q1 and Q2 respectively. Calculate the median of Q1 and Q2 to obtain R. Name the coordinate point in the resistance feedback curve where the X-axis is equal to R as the auxiliary reference virtual point and name the coordinate point in the resistance feedback curve where the X-axis is equal to the reference resistance as the target reference point. The auxiliary reference real point, the auxiliary reference virtual point and the target reference point are collectively referred to as pain feedback reference data.

[0029] Determine if an effective electromyographic reference point exists. If so, output the historical reference analysis signal; otherwise, output the predictive reference analysis signal.

[0030] If a historical reference analysis signal is output, the historical reference analysis scheme is executed; if a predictive reference analysis signal is output, the predictive reference analysis scheme is executed.

[0031] Furthermore, the historical reference analysis scheme includes the following sub-steps:

[0032] The effective electromyographic reference points are numbered in top-to-bottom order and designated by the symbol FA. n Let be an expression, where n is a non-zero natural number and n is the index of FA. n The corresponding Y-axis value is labeled TF. n , FA n The auxiliary resistor is marked as XF n XF n The corresponding auxiliary reference real point is labeled PF. n , will PF n The Y-axis value is denoted as YF. n Mark the Y-axis value of the target reference point as YC;

[0033] Assuming the predicted pain perception of the user is TS, the formula is used to... Calculate TS and FA n The values ​​of TT during comparison, where TT n Representing TS and FA n The value when compared;

[0034] Calculate TT n The average value is used to obtain the predicted value of the user's pain perception, which is named the predicted pain perception. The predicted pain perception is the feedback reference data.

[0035] Furthermore, the predictive reference analysis scheme includes the following sub-steps:

[0036] The electromyography feedback analysis points are fitted with a function, and the curve corresponding to the function obtained by the function fitting is named the electromyography feedback curve.

[0037] Obtain the coordinates of the X-axis point on the electromyography feedback curve that is equal to the reference electromyography point, and name it the electromyography mean feedback point. Mark the Y-axis value of the electromyography mean feedback point as FF, and mark the Y-axis value of the auxiliary reference virtual point as K.

[0038] Through formula The TS is calculated and solved to obtain the predicted value of the user's pain perception, which is named pain prediction perception.

[0039] Furthermore, intelligent control of intelligent electrical pulse therapy based on feedback reference data includes the following sub-steps:

[0040] If the predicted pain perception is greater than or equal to the value assigned when the pain perception is pain, then the intensity signal is reduced.

[0041] If the predicted pain perception is less than the value assigned when the pain perception is strong, then an intensity increase signal is output.

[0042] The intensity of the electrical pulse is adjusted down and up based on the intensity decrease signal and intensity increase signal until the pain prediction perception is greater than or equal to the value assigned when the pain perception is strong, and less than the value assigned when the pain perception is painful.

[0043] The beneficial effects of this invention are as follows: This invention collects biofeedback signals from physiotherapy users, simultaneously collecting their subjective feedback data. The subjective feedback data is then converted into effective subjective data. Based on the biofeedback signals and effective subjective data, feedback analysis is performed on the relationship between different biofeedback signals and subjective feedback data. Furthermore, pain feedback reference data is extracted from the resistive feedback analysis graph based on the user's real-time biofeedback signals. The advantage lies in the fact that, among biofeedback signals, electromyography (EMG) values ​​best reflect the user's pain perception, but this is not absolute. When users have the same EMG value, their pain perception is not entirely identical; rather, there is a dynamic range. That is, the user's pain perception is not only reflected in the EMG value but also in multiple data points. Among these, the response of EMG values ​​and skin resistance to changes in pain perception is particularly significant. Therefore, analyzing EMG values ​​and skin resistance to extract pain feedback reference data provides a data basis for subsequent pain perception prediction, improving the accuracy and effectiveness of electro-pulse physiotherapy control.

[0044] This invention constructs a historical reference analysis scheme to calibrate the user's electromyography (EMG) feedback analysis points based on pain feedback reference data and predict the user's pain perception, thereby obtaining feedback reference data. Alternatively, it constructs a predictive reference analysis scheme to calibrate the user's EMG feedback analysis points based on pain feedback reference data and predict the user's pain perception, thereby obtaining feedback reference data. Finally, it uses the feedback reference data to intelligently control the intelligent electrical pulse therapy. The advantage lies in the fact that by calibrating the EMG values ​​and skin resistance, the user's pain perception can be accurately predicted, thereby enabling adaptive intelligent control of the electrical pulse intensity during the therapy process, improving the accuracy and intelligence of the electrical pulse therapy control. Attached Figure Description

[0045] Figure 1 This is a flowchart of the steps of the method of the present invention;

[0046] Figure 2 This is a schematic diagram of the electromyographic feedback analysis diagram of the present invention;

[0047] Figure 3 This is a schematic diagram of the conductivity feedback analysis diagram of the present invention;

[0048] Figure 4 This is a schematic diagram of the conductivity feedback curve of the present invention;

[0049] Figure 5 This is a schematic diagram of the effective reference point for electromyography in this invention;

[0050] Figure 6 This is a schematic diagram of the auxiliary reference real point, auxiliary reference imaginary point, and target reference point of the present invention. Detailed Implementation

[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] Example 1, please refer to Figure 1 As shown, this application provides an intelligent electro-pulse therapy control method based on AI adaptive feedback, including the following steps:

[0053] Step S1: Collect biofeedback signals from the physiotherapy user and collect subjective feedback data from the user. The biofeedback signals include electromyography (EMG) values ​​and skin conductance, while the subjective feedback data includes pain perception, which includes no pain, sensation, strong sensation, and pain.

[0054] In practice, the collection of biofeedback signals from physiotherapy users is divided into two parts. In the first part, a sensory test is conducted on the user. During the test, the intensity of the electrical pulses is slowly increased from low to high, and electromyography (EMG) values ​​and skin conductance are recorded in real time. Simultaneously, the user's pain perception is obtained. Since pain perception is highly subjective and individual tolerance varies, a personalized approach is necessary for each user. Because the levels of pain perception cannot be precisely distinguished and depend entirely on the user's personal experience, pain perception is divided into four levels: no sensation, sensation, strong sensation, and painful sensation. No sensation means the electrical pulse cannot be felt; sensation means the electrical pulse can be felt, but not intensely. At this stage, the therapeutic effect is relatively weak. Strong sensation means that a strong electrical pulse can be felt, but it does not cause pain. This is the optimal therapeutic effect. Pain sensation means that pain can be felt, indicating that the intensity of the electrical pulse is high and exceeds the user's tolerance. In the perception test, the user's electromyography (EMG) value, skin conductance, and pain perception are collected. Then, each user's personalized feedback analysis data is analyzed. The second part is the actual therapeutic treatment. During the actual therapeutic treatment, only the user's EMG value and skin conductance need to be monitored. There is no need to ask the user about their pain perception. At this time, the user's real-time EMG value and skin conductance, combined with the feedback analysis data, can be used to predict the user's pain perception.

[0055] Step S2 involves converting subjective feedback data into valid subjective data and conducting preliminary analysis of the feedback data between different biofeedback signals and subjective feedback data. Step S2 includes the following sub-steps:

[0056] Step S201: Convert subjective feedback data into effective subjective data; Converting subjective feedback data into effective subjective data specifically involves assigning values ​​to the pain perception of no sensation, sensation, strong sensation, and pain sensation, respectively, to 1, 2, 3, and 4.

[0057] In practice, the degree of perception of electrical pulses is increasing from no sensation to sensation, strong sensation to pain. Therefore, they are assigned values ​​of 1, 2, 3 and 4 respectively to quantify and analyze no sensation, sensation, strong sensation and pain.

[0058] Step S202: Analyze the feedback data between different biological feedback signals and subjective feedback data based on the biological feedback signals and effective subjective data.

[0059] Step S202 includes the following sub-steps:

[0060] Please see Figures 2 to 3 As shown, in step S2021, a two-dimensional coordinate system is established with electromyography value and skin conductance as the X-axis and pain perception as the Y-axis, and named electromyography feedback analysis diagram and conductance feedback analysis diagram, respectively.

[0061] Step S2022: The biofeedback signals are entered into the electromyography feedback analysis chart and the electrical conductivity feedback analysis chart according to the subjective feedback data.

[0062] Step S2023: Name the coordinate points in the electromyography feedback analysis diagram and the conductance feedback analysis diagram as electromyography feedback analysis points and conductance feedback analysis points, respectively.

[0063] Step S2024, the electromyography feedback analysis points and the conductance feedback analysis points are the feedback analysis data;

[0064] In specific implementation, the electromyography feedback analysis diagram and the conductivity feedback analysis diagram are constructed as follows: Figure 2 as well as Figure 3 As shown, Figure 2 and Figure 3 The coordinate points in the diagram represent the feedback analysis data, which is obtained from the perception test analysis mentioned in step S1. The feedback analysis data for each individual is different.

[0065] Step S3 involves in-depth analysis of the impact of different biofeedback signals on subjective feedback data, and the construction of historical reference analysis schemes and predictive reference analysis schemes to calibrate the feedback analysis data to obtain feedback reference data. Step S3 includes the following sub-steps:

[0066] Step S301: Extract pain feedback reference data from the conductivity feedback analysis graph based on the user's real-time biofeedback signal;

[0067] Step S301 includes the following sub-steps:

[0068] Step S3011: Obtain the user's real-time biofeedback signal, name it the real-time monitoring feedback signal, and name the electromyography value and skin conductance in the real-time monitoring feedback signal as reference electromyography and reference conductance, respectively.

[0069] Please see Figure 4 As shown, in step S3012, a function is fitted to the conductivity feedback analysis point, and the curve corresponding to the function obtained by the function fitting is named the conductivity feedback curve.

[0070] Please see Figures 5 to 6 As shown, in step S3013, the electromyography feedback analysis point where the X-axis is equal to the reference electromyography is named the effective reference point of electromyography, the skin conductance in the biofeedback signal to which the effective reference point of electromyography belongs is obtained and named as the auxiliary conductance, and the coordinate point in the conductance feedback curve where the X-axis is equal to the auxiliary conductance is named as the auxiliary reference real point.

[0071] Step S3014: Obtain the minimum and maximum values ​​of X in the conductance feedback curve, and label them as Q1 and Q2 respectively. Calculate the median of Q1 and Q2 to obtain R. Name the coordinate point in the conductance feedback curve where the X-axis is equal to R as the auxiliary reference virtual point, and name the coordinate point in the conductance feedback curve where the X-axis is equal to the reference conductance as the target reference point. The auxiliary reference real point, the auxiliary reference virtual point, and the target reference point are collectively referred to as pain feedback reference data.

[0072] Step S3015: Determine whether there is an effective electromyographic reference point. If so, output the historical reference analysis signal; otherwise, output the predictive reference analysis signal.

[0073] Step S3016: If a historical reference analysis signal is output, then the historical reference analysis scheme is executed; if a predictive reference analysis signal is output, then the predictive reference analysis scheme is executed.

[0074] In practice, reference electromyography (EMG) and reference conductance were obtained at 45 μV and 25 μS, respectively. The conductance feedback curve was obtained through regression model analysis, as shown below. Figure 4 As shown, the electromyography (EMG) feedback analysis point with an X-axis value of 45 μV is named the effective EMG reference point. The effective EMG reference point is obtained as follows: Figure 5 As shown, Figure 5 There are two effective electromyographic reference points in the visual field, but because pain perception is only integer and has no decimals, the Y-axis values ​​are limited, resulting in multiple overlapping coordinate points. Figure 5 In the middle, there are actually two effective electromyographic reference points for each of the two coordinate points, namely... Figure 5 There are a total of four effective electromyographic reference points, namely (45,2), (45,2), (45,3), and (45,3). In this embodiment, they are labeled as α1, α2, α3, and α4, respectively. The auxiliary conductivities corresponding to α1, α2, α3, and α4 are 20, 22, 27, and 29, respectively. Figure 4 The coordinate points with X values ​​of 20, 22, 27, and 29 in the curve are named auxiliary reference real points. Q1 and Q2 are obtained as 4 μS and 41 μS respectively, and R is calculated to be 22.5 μS. The coordinate point with X value of 22.5 is named the auxiliary reference virtual point. The coordinate point with X-axis equal to 25 μS in the conductance feedback curve is named the target reference point. The auxiliary reference real points, auxiliary reference virtual points, and target reference points are obtained as follows: Figure 6 As shown, Figure 6 The black coordinate points are auxiliary reference virtual points, the gray coordinate points are auxiliary reference real points, and the white coordinate points are target reference points. Since there are 4 effective electromyographic reference points, historical reference analysis signals are output, and the historical reference analysis scheme is executed.

[0075] Step S302: Construct a historical reference analysis scheme, calibrate the user's electromyographic feedback analysis points based on pain feedback reference data, and predict the user's pain perception to obtain feedback reference data;

[0076] Step S302 includes the following sub-steps:

[0077] Step S3021: Number the effective electromyographic reference points in top-to-bottom order, using the symbol FA. n Let be an expression, where n is a non-zero natural number and n is the index of FA. n The corresponding Y-axis value is labeled TF. n , FA n The auxiliary conductance is marked as XF n XF n The corresponding auxiliary reference real point is labeled PF. n , will PF n The Y-axis value is denoted as YF. n Mark the Y-axis value of the target reference point as YC;

[0078] Step S3022, assuming the predicted pain perception of the user is TS, use the formula Calculate TS and FA n The values ​​of TT during comparison, where TT n Representing TS and FA n The value when compared;

[0079] Step S3023, calculate TT n The average value is used to obtain the predicted value of the user's pain perception, which is named the predicted pain perception. The predicted pain perception is the feedback reference data.

[0080] In practice, the numbers FA1 to FA4 are obtained, corresponding to α1, α2, α3, and α4 respectively, where 1 ≤ n ≤ 4. TF1 to TF4 are obtained as 3, 3, 2, and 2 respectively, and XF1 to XF4 are obtained as 29, 27, 22, and 20 respectively. Figure 6The auxiliary reference points are arranged from left to right as PF4 to PF1, yielding YF1 to YF4 values ​​of 2.13, 1.94, 1.59, and 1.47 respectively, and YC of 1.79. Since pain perception is reflected in both electromyography (EMG) and skin conductance, and a single factor cannot accurately predict pain perception in physiotherapy users, it is necessary to calibrate pain perception based on EMG values ​​using skin conductance. While the effective EMG reference points have the same EMG value, they exhibit different pain perceptions due to differences in skin conductance. These differences in skin conductance lead to variations in the location of the effective EMG reference points. Assuming the user's pain perception is TS, the positional relationship between TS and the effective EMG reference points approaches that of... Figure 6 The positional relationship between the auxiliary reference real point and the target reference point is calculated, and since the auxiliary reference real point corresponds to different effective electromyographic reference points, the calculation... The values ​​for TT1 to TT4 are 2.52, 2.77, 2.25 and 2.44 respectively. The results are rounded to two decimal places, and the average value is taken to get a pain prediction perception of 2.495. Rounding to the nearest integer, the pain prediction perception is 2, which means that the pain is perceived.

[0081] Step S303: Construct a predictive reference analysis scheme, calibrate the user's electromyographic feedback analysis points based on pain feedback reference data, and predict the user's pain perception to obtain feedback reference data;

[0082] Step S303 includes the following sub-steps:

[0083] Step S3031: Perform function fitting on the electromyography feedback analysis points, and name the curve corresponding to the function obtained by function fitting as the electromyography feedback curve.

[0084] Step S3032: Obtain the coordinate point on the electromyography feedback curve where the X-axis is equal to the reference electromyography point, name it the electromyography mean feedback point, mark the Y-axis value of the electromyography mean feedback point as FF, and mark the Y-axis value of the auxiliary reference virtual point as K.

[0085] Step S3033, through formula The TS is calculated and solved to obtain the predicted value of the user's pain perception, which is named pain prediction perception;

[0086] In practice, the principle of the predictive reference analysis scheme is the same as that of the historical reference analysis scheme. The difference lies in the fact that when the reference electromyography (EMG) and reference conductance are 45 μV and 25 μS respectively, there is no historical data available for reference. In this case, it is necessary to integrate existing data and find approximate values ​​as the analysis object. The EMG feedback curve obtained by function fitting approaches the value when the skin conductance is at the average level. For example, the Y-axis value of the coordinate point X=45 μV on the EMG feedback curve approaches the average level of the Y-axis of the four effective EMG reference points, that is, it approaches the median of the skin conductance of the four effective EMG reference points. The values ​​of reference EMG without effective EMG reference points are similar. However, since there are no effective EMG reference points, Q1 and Q2 are directly used as references to analyze R. R is in Figure 6 The position of the TS in the curve is close to that of the TS on the electromyographic feedback curve. The FF is 2.48μV and K is 1.64. The calculated TS is 2.71. Rounding to the integer, the TS is 3, which means the pain prediction perception is 3, i.e., strong sensation. The two analysis results are different because the prediction reference analysis scheme does not have accurate reference data, but uses approximate data as a reference, which has a certain error. However, the pain prediction perception obtained by the historical reference analysis scheme is 2.495. If we increase it by 0.005, it becomes 2.5. At this time, rounding to the integer, the pain prediction perception is 3, which shows that the error is very small.

[0087] Step S4 involves intelligently controlling the intelligent electrical pulse therapy based on feedback reference data; Step S4 includes the following sub-steps:

[0088] Step S401: If the pain prediction perception is greater than or equal to the value assigned when the pain perception is pain sensation, then output an intensity decrease signal;

[0089] Step S402: If the pain prediction perception is less than the value assigned when the pain perception is strong, then output an intensity increase signal.

[0090] Step S403: Based on the intensity decrease signal and intensity increase signal, the intensity of the electrical pulse is adjusted down and increased until the pain prediction perception is greater than or equal to the value assigned when the pain perception is strong, and less than the value assigned when the pain perception is pain.

[0091] In practice, when the pain perception is strong, the physiotherapy user has a strong sensation of the electrical pulse but does not feel pain. The physiotherapy effect is best at this time. Therefore, the pain perception is maintained at a strong level, and the intensity of the electrical pulse is adjusted in real time based on the pain prediction perception, without requiring the user to provide pain perception in real time.

[0092] Example 2: This application provides an electronic device, which may include a processor, a communication interface, a memory, and a communication bus. The processor, communication interface, and memory communicate with each other via the communication bus. The memory stores computer-readable instructions, and the processor can call these instructions. When the processor executes a computer-readable instruction, it performs steps such as those in the AI-based adaptive feedback intelligent electro-pulse physiotherapy control method to achieve the following functions: collecting biofeedback signals from the physiotherapy user and simultaneously collecting the user's subjective feedback data; converting the subjective feedback data into valid subjective data and performing preliminary analysis of the feedback analysis data between different biofeedback signals and subjective feedback data; deeply analyzing the influence of different biofeedback signals on the subjective feedback data, and simultaneously constructing historical reference analysis schemes and predictive reference analysis schemes to calibrate the feedback analysis data to obtain feedback reference data; and intelligently controlling the intelligent electro-pulse physiotherapy based on the feedback reference data.

[0093] Furthermore, when the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0094] Example 3: This application also provides a computer program product, which includes a computer program stored on a computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the AI-based adaptive feedback intelligent electro-pulse physiotherapy control method provided by the above methods. This method includes: collecting biofeedback signals from physiotherapy users and simultaneously collecting subjective feedback data from users; converting the subjective feedback data into effective subjective data and initially analyzing the feedback analysis data between different biofeedback signals and subjective feedback data; deeply analyzing the influence of different biofeedback signals on subjective feedback data, and simultaneously constructing a historical reference analysis scheme and a predictive reference analysis scheme to calibrate the influence of the feedback analysis data to obtain feedback reference data; and intelligently controlling the intelligent electro-pulse physiotherapy based on the feedback reference data.

[0095] Example 4: This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it performs the steps of the AI-based adaptive feedback intelligent electro-pulse physiotherapy control method described above to achieve the following functions: collecting biofeedback signals from physiotherapy users and simultaneously collecting subjective feedback data from users; converting subjective feedback data into effective subjective data and performing preliminary analysis of feedback analysis data between different biofeedback signals and subjective feedback data; deeply analyzing the influence of different biofeedback signals on subjective feedback data, and simultaneously constructing historical reference analysis schemes and predictive reference analysis schemes to calibrate the influence of feedback analysis data to obtain feedback reference data; and performing intelligent control of intelligent electro-pulse physiotherapy based on the feedback reference data.

[0096] Based on the above description of the embodiments, the embodiments of the present invention can be provided as methods, systems, or computer program products. Based on this understanding, the above technical solutions, in essence or in terms of their contribution to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or certain parts of the embodiments.

[0097] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces. The indirect coupling or communication connection between systems, modules, and units may be electrical, mechanical, or other forms.

[0098] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. An intelligent electropulse therapy control method based on AI adaptive feedback, characterized in that, Includes the following steps: Collect biofeedback signals from physiotherapy users, and simultaneously collect subjective feedback data from users; The subjective feedback data was converted into effective subjective data, and the feedback analysis data between different biological feedback signals and subjective feedback data was preliminarily analyzed. We conducted an in-depth analysis of the impact of different biofeedback signals on subjective feedback data, and constructed historical reference analysis schemes and predictive reference analysis schemes to calibrate the feedback analysis data to obtain feedback reference data. Intelligent control of intelligent electrical pulse therapy based on feedback reference data; The process of converting subjective feedback data into effective subjective data and conducting preliminary analysis of the feedback between different biofeedback signals and subjective feedback data includes the following sub-steps: Transform subjective feedback data into effective subjective data; Based on biofeedback signals and effective subjective data analysis, feedback analysis data between different biofeedback signals and subjective feedback data is analyzed, including: establishing a two-dimensional coordinate system with electromyography (EMG) values ​​and skin resistance as the X-axis and pain perception as the Y-axis, named EMG feedback analysis diagram and resistance feedback analysis diagram respectively; recording biofeedback signals into the EMG feedback analysis diagram and resistance feedback diagram according to subjective feedback data; naming the coordinate points in the EMG feedback analysis diagram and resistance feedback analysis diagram as EMG feedback analysis points and resistance feedback analysis points respectively; the EMG feedback analysis points and resistance feedback points are the feedback analysis data; A thorough analysis of the impact of different biofeedback signals on subjective feedback data is conducted. Simultaneously, historical and predictive reference analysis schemes are constructed to calibrate the feedback analysis data, resulting in feedback reference data. This process includes the following sub-steps: Pain feedback reference data is extracted from the resistive feedback analysis graph based on real-time biofeedback signals from users. A historical reference analysis scheme is constructed, which calibrates the user's electromyographic feedback analysis points based on pain feedback reference data and predicts the user's pain perception to obtain feedback reference data; A predictive reference analysis scheme is constructed to calibrate the user's electromyographic feedback analysis points based on pain feedback reference data and predict the user's pain perception, thereby obtaining feedback reference data; The pain feedback reference data extracted from the resistive feedback analysis graph based on the user's real-time biofeedback signal includes the following sub-steps: The user's real-time biofeedback signal is acquired and named the real-time monitoring feedback signal. The electromyography value and skin resistance in the real-time monitoring feedback signal are named the reference electromyography and reference resistance, respectively. Perform function fitting on the resistance feedback analysis points, and name the curve corresponding to the function obtained by function fitting the resistance feedback curve. The electromyography feedback analysis point where the X-axis equals the reference electromyography is named the effective reference point of electromyography. The skin resistance in the biofeedback signal to which the effective reference point of electromyography belongs is obtained and named the auxiliary resistance. The coordinate point in the resistance feedback curve where the X-axis equals the auxiliary resistance is named the auxiliary reference real point. Obtain the minimum and maximum values ​​of X in the resistance feedback curve and label them as Q1 and Q2 respectively. Calculate the median of Q1 and Q2 to obtain R. Name the coordinate point in the resistance feedback curve where the X-axis is equal to R as the auxiliary reference virtual point and name the coordinate point in the resistance feedback curve where the X-axis is equal to the reference resistance as the target reference point. The auxiliary reference real point, the auxiliary reference virtual point and the target reference point are collectively referred to as pain feedback reference data. Determine if an effective electromyographic reference point exists. If so, output the historical reference analysis signal; otherwise, output the predictive reference analysis signal. If a historical reference analysis signal is output, the historical reference analysis scheme is executed; if a predictive reference analysis signal is output, the predictive reference analysis scheme is executed. The historical reference analysis scheme includes the following sub-steps: The effective reference point of the electromyogram is numbered in the order from top to bottom, and is denoted by a symbol FA n , wherein n is a non-zero natural number and n is the serial number of FA n , and the Y-axis value corresponding to FA n is denoted by TF n , and the auxiliary resistance of FA n is denoted by XF n , and the Y-axis value corresponding to XF n is denoted by YF n , and the auxiliary reference point corresponding to XF n is denoted by PF , and the Y-axis value corresponding to PF is denoted by YC Assuming the predicted pain perception of the user is TS, the formula is used... Calculate TS and FA n The values ​​of TT during comparison, where TT n Representing TS and FA n The value when compared; Calculate the average of TT n to obtain a predicted value of the user's pain perception, named Pain Predicted Perception, which is the feedback reference data. The predictive reference analysis scheme includes the following sub-steps: The electromyography feedback analysis points are fitted with a function, and the curve corresponding to the function obtained by the function fitting is named the electromyography feedback curve. Obtain the coordinates of the X-axis point on the electromyography feedback curve that is equal to the reference electromyography point, and name it the electromyography mean feedback point. Mark the Y-axis value of the electromyography mean feedback point as FF, and mark the Y-axis value of the auxiliary reference virtual point as K. Through formula The TS is calculated and solved to obtain the predicted value of the user's pain perception, which is named pain prediction perception.

2. The intelligent electro-pulse physiotherapy control method based on AI adaptive feedback according to claim 1, characterized in that, The system collects biofeedback signals from physiotherapy users, as well as subjective feedback data. The biofeedback signals include electromyography (EMG) values ​​and skin resistance. The subjective feedback data includes pain perception, which includes no sensation, sensation, strong sensation, and pain sensation.

3. The intelligent electro-pulse therapy control method based on AI adaptive feedback according to claim 2, characterized in that, The conversion of subjective feedback data into effective subjective data specifically involves assigning values ​​to the pain perception of no sensation, sensation, strong sensation, and pain sensation, respectively, with values ​​of 1, 2, 3, and 4.

4. The intelligent electro-pulse therapy control method based on AI adaptive feedback according to claim 3, characterized in that, Intelligent control of intelligent electrical pulse therapy based on feedback reference data includes the following sub-steps: If the predicted pain perception is greater than or equal to the value assigned when the pain perception is pain, then the intensity signal is reduced. If the predicted pain perception is less than the value assigned when the pain perception is strong, then an intensity increase signal is output. The intensity of the electrical pulse is adjusted down and up based on the intensity decrease signal and intensity increase signal until the pain prediction perception is greater than or equal to the value assigned when the pain perception is strong, and less than the value assigned when the pain perception is painful.

Images