Peripheral stimulation parameter prediction method and system based on multi-modal feedback

Abstract

The invention discloses a peripheral stimulation parameter prediction method and system based on multi-modal feedback. The method comprises the steps of obtaining corresponding multi-modal feedback data when electromagnetic stimulation is applied to a target muscle group; determining physiological response intensity corresponding to the target muscle group based on the multi-modal feedback data; based on a machine learning regression algorithm, establishing a treatment parameter prediction model corresponding to the target muscle group according to suitable stimulation parameters corresponding to the electromagnetic stimulation and the physiological response intensity; and when the target muscle group is treated, predicting ideal treatment parameters corresponding to the target muscle group according to the real-time physiological response intensity corresponding to the target muscle group and the treatment parameter prediction model. Therefore, accurate prediction of the muscle peripheral treatment parameters based on multi-modal feedback can be realized, and the pertinence and the treatment effect of muscle group peripheral electromagnetic treatment are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular, to a method and system for predicting peripheral stimulation parameters based on multimodal feedback. Background Art

[0002] With the rapid growth of the needs for personalized medicine and rehabilitation, treating muscle groups through electromagnetic stimulation technology has also become one of the options for more and more sports medicine rehabilitation treatments. Existing technologies usually determine treatment parameters by collecting single-modal feedback data of muscle groups under electromagnetic stimulation and combining fixed parameter settings or manual adjustment methods to achieve rehabilitation treatment goals. Due to the lack of comprehensive analysis of multimodal feedback data and the dynamic prediction ability of machine learning algorithms in existing solutions, it is difficult to accurately evaluate the physiological response intensity and adapt to individual treatment needs. Commonly used static parameter configurations cannot cope with real-time changes, resulting in poor treatment effects and limiting the efficiency and accuracy of personalized rehabilitation. It can be seen that there are deficiencies in the existing technologies and they urgently need to be solved. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method and system for predicting peripheral stimulation parameters based on multimodal feedback, which can achieve accurate prediction of muscle peripheral treatment parameters based on multimodal feedback and improve the pertinence and treatment effect of peripheral electromagnetic treatment of muscle groups.

[0004] To solve the above technical problem, in the first aspect of the present invention, a method for predicting peripheral stimulation parameters based on multimodal feedback is disclosed, and the method includes: Obtaining corresponding multimodal feedback data when an electromagnetic stimulation is applied to a target muscle group; Determining the physiological response intensity corresponding to the target muscle group based on the multimodal feedback data; Establishing a treatment parameter prediction model corresponding to the target muscle group based on a machine learning regression algorithm according to the appropriate stimulation parameters corresponding to the electromagnetic stimulation and the physiological response intensity; When the target muscle group is being treated, predicting the ideal treatment parameters corresponding to the target muscle group according to the real-time physiological response intensity corresponding to the target muscle group and the treatment parameter prediction model.

[0005] As an optional implementation manner, in the first aspect of the present invention, the multimodal feedback data includes temperature data, humidity data, physical displacement data, electromyogram signal data, and image data.

[0006] As an optional implementation manner, in the first aspect of the present invention, the multimodal feedback data is obtained through a multimodal sensing platform disposed on the target muscle group; the multimodal sensing platform includes a temperature sensor, a humidity sensor, a physical displacement infrared sensor, sEMG electrodes, and an image sensor.

[0007] As an alternative embodiment, in the first aspect of the present invention, determining the physiological response intensity corresponding to the target muscle group based on the multimodal feedback data includes: Input each data in the multimodal feedback data into a response degree prediction model corresponding to the corresponding data type to obtain a response degree parameter corresponding to each data; the response degree prediction model is trained by a training data set including a plurality of training feedback data corresponding to the corresponding data type and the corresponding response degree annotation; Calculate the weighted sum value of the response degree parameters corresponding to all data to obtain the physiological response intensity corresponding to the target muscle group; wherein, the calculation weight corresponding to each data response degree parameter is the product of the type weight and the model accuracy weight; the type weight decreases in turn when the data types of the corresponding data are electromyogram signal data, physical displacement data, image data, temperature data and humidity data; the model accuracy weight is proportional to the prediction accuracy of the response degree prediction model corresponding to the corresponding data type in the verification stage.

[0008] As an alternative embodiment, in the first aspect of the present invention, when the data type is temperature data, humidity data or physical displacement data, the model architecture of the corresponding response degree prediction model is a CNN neural network; when the data type is image data, the corresponding response degree prediction model is an RNN neural network; when the data type is electromyogram signal data, the corresponding response degree prediction model is a mathematical relationship fitting model, and the response degree annotation in the corresponding training data set is the weighted sum average of the amplitude mean ratio parameter, the latency mean ratio parameter and the average power frequency ratio parameter; the amplitude mean ratio parameter is the ratio of the signal amplitude means corresponding to the electromechanical signal data in the resting state and the stimulated state respectively; the latency mean ratio parameter is the ratio of the signal latency means corresponding to the electromechanical signal data in the resting state and the stimulated state respectively; the average power frequency ratio parameter is the ratio of the average power frequencies corresponding to the electromechanical signal data in the resting state and the stimulated state respectively.

[0009] As an alternative embodiment, in the first aspect of the present invention, the appropriate stimulation parameter or the ideal treatment parameter includes at least one of the intensity parameter, pulse width parameter and frequency parameter of electrical stimulation or magnetic stimulation.

[0010] As an alternative embodiment, in the first aspect of the present invention, predicting the ideal treatment parameter corresponding to the target muscle group according to the real-time physiological response intensity corresponding to the target muscle group and the treatment parameter prediction model includes: Obtain the first multi-modal feedback data of the target muscle group and the second multi-modal feedback data corresponding to at least one associated muscle group; there is at least one neural connection channel between the associated muscle group and the target muscle group, and the physical distance between the associated muscle group and the target muscle group is less than a preset distance threshold; Calculate the first real-time physiological response intensity corresponding to the target muscle group according to the first multi-modal feedback data; Calculate the second real-time physiological response intensity corresponding to at least one of the associated muscle groups according to the second multi-modal feedback data; Correct the first real-time physiological response intensity according to the second real-time physiological response intensity to obtain a corrected physiological response intensity; Input the corrected physiological response intensity into the treatment parameter prediction model to obtain the ideal treatment parameters corresponding to the target muscle group.

[0011] As an optional implementation manner, in the first aspect of the present invention, the correcting the first real-time physiological response intensity according to the second real-time physiological response intensity to obtain a corrected physiological response intensity includes: For each of the associated muscle groups, calculate the weighted sum value between the number of channels of the neural connection channel corresponding to the associated muscle group and the physical distance to obtain the associated parameter corresponding to the associated muscle group; Determine the corresponding influence ratio parameter according to the preset correspondence between the associated parameter and the influence ratio, and the associated parameter; the influence ratio parameter is less than 1; Calculate the product of the second real-time physiological response intensity corresponding to the associated muscle group and the influence ratio parameter to obtain the influence response intensity parameter corresponding to the associated muscle group; Calculate the average value of the influence response intensity parameters of all the associated muscle groups and the first real-time physiological response intensity to obtain a corrected physiological response intensity.

[0012] A second aspect of the embodiments of the present invention discloses a peripheral stimulation parameter prediction system based on multi-modal feedback, and the system includes: An acquisition module, configured to acquire corresponding multi-modal feedback data when an electromagnetic stimulus is applied to the target muscle group; A determination module, configured to determine the physiological response intensity corresponding to the target muscle group based on the multi-modal feedback data; A modeling module, configured to establish a treatment parameter prediction model corresponding to the target muscle group based on a machine learning regression algorithm according to the appropriate stimulation parameters corresponding to the electromagnetic stimulus and the physiological response intensity; A prediction module, configured to predict an ideal treatment parameter corresponding to the target muscle group according to a real-time physiological response intensity corresponding to the target muscle group and a treatment parameter prediction model when the target muscle group is being treated.

[0013] As an optional implementation manner, in the second aspect of the present invention, the multimodal feedback data includes temperature data, humidity data, physical displacement data, electromyogram signal data, and image data.

[0014] As an optional implementation manner, in the second aspect of the present invention, the multimodal feedback data is obtained through a multimodal sensing platform disposed on the target muscle group; the multimodal sensing platform includes a temperature sensor, a humidity sensor, a physical displacement infrared sensor, sEMG electrodes, and an image sensor.

[0015] As an optional implementation manner, in the second aspect of the present invention, the specific manner in which the determination module determines the physiological response intensity corresponding to the target muscle group based on the multimodal feedback data includes: Input each piece of data in the multimodal feedback data into a response degree prediction model corresponding to the corresponding data type to obtain a response degree parameter corresponding to each piece of data; the response degree prediction model is trained through a training data set including a plurality of training feedback data corresponding to the corresponding data type and corresponding response degree annotations; Calculate a weighted sum value of the response degree parameters corresponding to all the data to obtain the physiological response intensity corresponding to the target muscle group; wherein, the calculation weight corresponding to each data response degree parameter is the product of a type weight and a model accuracy weight; the type weight decreases in turn when the data types of the corresponding data are electromyogram signal data, physical displacement data, image data, temperature data, and humidity data; the model accuracy weight is proportional to the prediction accuracy of the response degree prediction model corresponding to the corresponding data type in the verification stage.

[0016] As an alternative embodiment, in the second aspect of the present invention, when the data type is temperature data, humidity data or physical displacement data, the model architecture of the corresponding reaction degree prediction model is a CNN neural network; when the data type is image data, the corresponding reaction degree prediction model is an RNN neural network; when the data type is electromyogram signal data, the corresponding reaction degree prediction model is a mathematical relationship fitting model, and the reaction degree annotation in the corresponding training data set is the weighted sum average of the amplitude mean ratio parameter, the latency mean ratio parameter and the mean power frequency ratio parameter; the amplitude mean ratio parameter is the ratio of the signal amplitude means corresponding to the electromechanical signal data in the resting state and the stimulated state respectively; the latency mean ratio parameter is the ratio of the signal latency means corresponding to the electromechanical signal data in the resting state and the stimulated state respectively; the mean power frequency ratio parameter is the ratio of the mean power frequencies corresponding to the electromechanical signal data in the resting state and the stimulated state respectively.

[0017] As an alternative embodiment, in the second aspect of the present invention, the appropriate stimulation parameter or the ideal treatment parameter includes at least one of the intensity parameter, the pulse width parameter and the frequency parameter of electrical stimulation or magnetic stimulation.

[0018] As an alternative embodiment, in the second aspect of the present invention, the specific manner in which the prediction module predicts the ideal treatment parameter corresponding to the target muscle group according to the real-time physiological response intensity corresponding to the target muscle group and the treatment parameter prediction model includes: Obtain the first multi-modal feedback data of the target muscle group and the second multi-modal feedback data corresponding to at least one associated muscle group; there is at least one neural connection channel between the associated muscle group and the target muscle group and the physical distance between the associated muscle group and the target muscle group is less than a preset distance threshold; Calculate the first real-time physiological response intensity corresponding to the target muscle group according to the first multi-modal feedback data; Calculate the second real-time physiological response intensity corresponding to at least one of the associated muscle groups according to the second multi-modal feedback data; Correct the first real-time physiological response intensity according to the second real-time physiological response intensity to obtain a corrected physiological response intensity; Input the corrected physiological response intensity into the treatment parameter prediction model to obtain the ideal treatment parameter corresponding to the target muscle group.

[0019] As an alternative embodiment, in the second aspect of the present invention, the specific manner in which the prediction module corrects the first real-time physiological response intensity according to the second real-time physiological response intensity to obtain a corrected physiological response intensity includes: For each of the associated muscle groups, calculate the weighted sum value between the number of channels of the nerve connection channel corresponding to the associated muscle group and the physical distance to obtain the association parameter corresponding to the associated muscle group; According to the preset correspondence between the association parameter and the influence ratio, and the association parameter, determine the corresponding influence ratio parameter; the influence ratio parameter is less than 1; Calculate the product of the second real-time physiological response intensity corresponding to the associated muscle group and the influence ratio parameter to obtain the influence response intensity parameter corresponding to the associated muscle group; Calculate the average value of the influence response intensity parameters of all the associated muscle groups and the first real-time physiological response intensity to obtain the corrected physiological response intensity.

[0020] The third aspect of the present invention discloses another peripheral stimulation parameter prediction system based on multimodal feedback. The system includes: A memory storing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory and executes some or all of the steps in the peripheral stimulation parameter prediction method based on multimodal feedback disclosed in the first aspect of the present invention.

[0021] The fourth aspect of the present invention discloses a computer storage medium. The computer storage medium stores computer instructions, which are used to execute some or all of the steps in the peripheral stimulation parameter prediction method based on multimodal feedback disclosed in the first aspect of the present invention when called.

[0022] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: The present invention obtains multimodal feedback data to determine the physiological response intensity when the target muscle group receives electromagnetic stimulation, constructs a treatment parameter prediction model using a machine learning regression algorithm, and predicts the ideal treatment parameter in combination with the real-time physiological response intensity, so as to realize the prediction of accurate muscle peripheral treatment parameters based on multimodal feedback, and improve the pertinence and treatment effect of peripheral electromagnetic treatment of muscle groups. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0024] Figure 1It is a schematic flowchart of a method for predicting peripheral stimulation parameters based on multimodal feedback disclosed in an embodiment of the present invention.

[0025] Figure 2 It is a schematic structural diagram of a system for predicting peripheral stimulation parameters based on multimodal feedback disclosed in an embodiment of the present invention.

[0026] Figure 3 It is a schematic structural diagram of another system for predicting peripheral stimulation parameters based on multimodal feedback disclosed in an embodiment of the present invention. Detailed implementation manners

[0027] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts fall within the scope of protection of the present invention.

[0028] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, device, product or equipment that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products or equipment.

[0029] Referring to "embodiments" herein means that the specific features, structures or characteristics described in connection with the embodiments can be included in at least one embodiment of the present invention. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0030] The present invention discloses a method and system for predicting peripheral stimulation parameters based on multimodal feedback. By acquiring multimodal feedback data when a target muscle group receives electromagnetic stimulation to determine the physiological response intensity, and using a machine learning regression algorithm to construct a treatment parameter prediction model, combined with the real-time physiological response intensity to predict the ideal treatment parameters, it is possible to achieve accurate prediction of muscle peripheral treatment parameters based on multimodal feedback, and improve the pertinence and treatment effect of muscle peripheral electromagnetic treatment. The following will be described in detail respectively.

[0031] Embodiment 1 Please refer to Figure 1 , Figure 1 which is a schematic flowchart of a method for predicting peripheral stimulation parameters based on multimodal feedback disclosed in an embodiment of the present invention. Among them, Figure 1 the described method for predicting peripheral stimulation parameters based on multimodal feedback can be applied to a data processing system / data processing device / data processing server (wherein, the server includes a local processing server or a cloud processing server). As Figure 1 shown, the method for predicting peripheral stimulation parameters based on multimodal feedback may include the following operations: 101. Obtain corresponding multimodal feedback data when an electromagnetic stimulation is applied to a target muscle group.

[0032] 102. Determine the physiological response intensity corresponding to the target muscle group based on the multimodal feedback data. 103. Based on a machine learning regression algorithm, establish a treatment parameter prediction model for the target muscle group according to the appropriate stimulation parameters corresponding to the electromagnetic stimulation and the physiological response intensity. Optionally, the appropriate stimulation parameters can be determined by an operator according to the change of the physiological response intensity, experience, or experimental data, and are used to indicate the appropriate electromagnetic stimulation parameters at the corresponding physiological response intensity.

[0033] 104. When the target muscle group is being treated, predict the ideal treatment parameters corresponding to the target muscle group according to the real-time physiological response intensity corresponding to the target muscle group and the treatment parameter prediction model.

[0034] It can be seen that in the above embodiment of the invention, multimodal feedback data is obtained when the target muscle group receives electromagnetic stimulation to determine the physiological response intensity, and a treatment parameter prediction model is constructed using a machine learning regression algorithm. Combining the real-time physiological response intensity to predict the ideal treatment parameters, thereby enabling the prediction of accurate muscle peripheral treatment parameters based on multimodal feedback, and improving the pertinence and treatment effect of peripheral electromagnetic treatment of muscle groups.

[0035] As an optional embodiment, in the above steps, the multimodal feedback data includes temperature data, humidity data, physical displacement data, electromyogram signal data, and image data.

[0036] It can be seen that through the above optional embodiment, the content of the multimodal feedback data is defined to comprehensively characterize the relevant features of the electromagnetic stimulation feedback of the muscle group, so as to facilitate subsequent accurate intensity determination and modeling prediction, and assist in realizing the prediction of accurate muscle peripheral treatment parameters based on multimodal feedback, and improving the pertinence and treatment effect of peripheral electromagnetic treatment of muscle groups.

[0037] As an optional embodiment, in the above steps, the multimodal feedback data is obtained through a multimodal sensing platform set up in the target muscle group; the multimodal sensing platform includes a temperature sensor, a humidity sensor, a physical displacement infrared sensor, a sEMG electrode and an image sensor.

[0038] It can be seen that through the above optional embodiments, the details of the physical acquisition equipment of multimodal feedback data are defined, so as to accurately and comprehensively extract the relevant features of the electromagnetic stimulation feedback that characterize the muscle group, so as to facilitate the subsequent accurate intensity determination and modeling prediction, and assist in realizing the accurate prediction of muscle peripheral treatment parameters based on multimodal feedback, thereby improving the targetedness and treatment effect of peripheral electromagnetic treatment of muscle groups.

[0039] As an optional embodiment, in the above step, determining the physiological response intensity corresponding to the target muscle group based on the multimodal feedback data includes: Inputting each data in the multimodal feedback data into a reaction degree prediction model corresponding to the corresponding data type to obtain a reaction degree parameter corresponding to each data; optionally, the reaction degree prediction model is trained by a training data set including a plurality of training feedback data of corresponding data types and corresponding reaction degree annotations; Calculate the weighted sum of the reaction degree parameters corresponding to all data to obtain the physiological response intensity corresponding to the target muscle group; optionally, the calculation weight corresponding to each data reaction degree parameter is the product of the type weight and the model accuracy weight; the type weight decreases in sequence when the data type of the corresponding data is electromyographic signal data, physical displacement data, image data, temperature data and humidity data respectively; the model accuracy weight is proportional to the prediction accuracy of the reaction degree prediction model corresponding to the corresponding data type in the verification stage.

[0040] It can be seen that through the above optional embodiments, the reaction degree parameters are generated by inputting multimodal feedback data into the reaction degree prediction model of the corresponding data type, and the weighted sum value is calculated with the product of the type weight and the model accuracy weight as the weight to determine the physiological response intensity of the target muscle group, thereby realizing accurate physiological response evaluation based on weighted analysis of multimodal data, improving the accuracy of subsequent electromagnetic stimulation treatment parameter prediction, assisting in the prediction of accurate muscle peripheral treatment parameters based on multimodal feedback, and improving the targetedness and treatment effect of peripheral electromagnetic treatment of muscle groups.

[0041] As an optional embodiment, in the above steps, when the data type is temperature data, humidity data, or physical displacement data, the model architecture of the corresponding reaction degree prediction model is a CNN neural network; when the data type is image data, the corresponding reaction degree prediction model is an RNN neural network; when the data type is electromyography signal data, the corresponding reaction degree prediction model is a mathematical relationship fitting model, and the reaction degree annotation in the corresponding training dataset is the weighted sum average of the amplitude mean ratio parameter, latency mean ratio parameter, and average power frequency ratio parameter; the amplitude mean ratio parameter is the ratio of the signal amplitude means corresponding to the electromechanical signal data in the resting state and the stimulated state respectively; the latency mean ratio parameter is the ratio of the signal latency means corresponding to the electromechanical signal data in the resting state and the stimulated state respectively; the average power frequency ratio parameter is the ratio of the average power frequencies corresponding to the electromechanical signal data in the resting state and the stimulated state respectively.

[0042] It can be seen that through the above optional embodiments, the details of the prediction models corresponding to different data types are defined. Specifically, the reaction degree of the electromyography signal is based on the weighted sum of the amplitude, latency, and average power frequency ratio to determine the accurate reaction degree annotation, so as to achieve accurate physiological response evaluation based on multimodal data weighted analysis, improve the accuracy of subsequent electromagnetic stimulation treatment parameter prediction, assist in realizing the prediction of accurate muscle peripheral treatment parameters based on multimodal feedback, and improve the pertinence and treatment effect of muscle group peripheral electromagnetic treatment.

[0043] As an optional embodiment, in the above steps, the appropriate stimulation parameters or ideal treatment parameters include at least one of the intensity parameter, pulse width parameter, and frequency parameter of electrical stimulation or magnetic stimulation.

[0044] It can be seen that through the above optional embodiments, the details of the electromagnetic stimulation parameters or treatment parameters are defined to achieve more comprehensive and accurate parameter prediction, assist in realizing the prediction of accurate muscle peripheral treatment parameters based on multimodal feedback, and improve the pertinence and treatment effect of muscle group peripheral electromagnetic treatment.

[0045] As an optional embodiment, in the above steps, predicting the ideal treatment parameters corresponding to the target muscle group according to the real-time physiological response intensity corresponding to the target muscle group and the treatment parameter prediction model includes: Obtaining the first multimodal feedback data of the target muscle group and the second multimodal feedback data of at least one associated muscle group; optionally, there is at least one neural connection channel between the associated muscle group and the target muscle group, and the physical distance between the associated muscle group and the target muscle group is less than a preset distance threshold; Calculating the first real-time physiological response intensity corresponding to the target muscle group according to the first multimodal feedback data; Calculate the second real-time physiological response intensity corresponding to at least one associated muscle group according to the second multimodal feedback data; Correct the first real-time physiological response intensity according to the second real-time physiological response intensity to obtain a corrected physiological response intensity; Input the corrected physiological response intensity into the treatment parameter prediction model to obtain the ideal treatment parameters corresponding to the target muscle group.

[0046] It can be seen that through the above optional embodiments, by obtaining the first multimodal feedback data of the target muscle group and the second multimodal feedback data of the associated muscle group that has a neural connection with the target muscle group and a physical distance less than the threshold, calculate the first real-time physiological response intensity of the target muscle group and the second real-time physiological response intensity of the associated muscle group respectively, and use the latter to correct the former to obtain the corrected physiological response intensity, and input it into the treatment parameter prediction model to generate the ideal treatment parameters, so as to realize accurate physiological response correction and treatment parameter prediction based on the correction of the associated muscle group, realize the prediction of accurate muscle peripheral treatment parameters based on multimodal feedback, and improve the pertinence and treatment effect of peripheral electromagnetic treatment of muscle groups.

[0047] As an optional embodiment, in the above steps, correcting the first real-time physiological response intensity according to the second real-time physiological response intensity to obtain a corrected physiological response intensity includes: For each associated muscle group, calculate the weighted sum value between the number of channels of the neural connection channel corresponding to the associated muscle group and the physical distance to obtain the associated parameter corresponding to the associated muscle group; According to the preset corresponding relationship between the associated parameter and the influence ratio, and the associated parameter, determine the corresponding influence ratio parameter; optionally, the influence ratio parameter is less than 1; Calculate the product of the second real-time physiological response intensity corresponding to the associated muscle group and the influence ratio parameter to obtain the influence response intensity parameter corresponding to the associated muscle group; Calculate the average value of the influence response intensity parameters of all associated muscle groups and the first real-time physiological response intensity to obtain the corrected physiological response intensity.

[0048] Optionally, the number of channels of the neural connection channel corresponding to the associated muscle group and the physical distance can be determined in advance by the operator based on biological knowledge or by analyzing the actual physiological data of the target user.

[0049] It can be seen that through the above optional embodiments, the association parameters are obtained by calculating the weighted sum of the number of neural connection channels and the physical distance of each associated muscle group, the influence ratio parameter is determined based on the preset corresponding relationship, the product of the second real-time physiological response intensity of the associated muscle group and the influence ratio parameter is calculated to obtain the influence response intensity parameter, and the average value of the influence response intensity parameters of all associated muscle groups and the first real-time physiological response intensity of the target muscle group is taken as the corrected physiological response intensity, thereby realizing accurate physiological response correction based on neural connection and distance weighting, realizing accurate prediction of muscle peripheral treatment parameters based on multimodal feedback, and improving the targetedness and treatment effect of peripheral electromagnetic treatment of muscle groups.

[0050] Example 2 See also Figure 2 , Figure 2 : is a schematic diagram of the structure of a peripheral stimulation parameter prediction system based on multimodal feedback disclosed in an embodiment of the present invention. Figure 2 The described peripheral stimulation parameter prediction system based on multimodal feedback can be applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 2 As shown, the peripheral stimulation parameter prediction system based on multimodal feedback may include: The acquisition module 201 is used to acquire corresponding multimodal feedback data when electromagnetic stimulation is applied to the target muscle group.

[0051] The determination module 202 is configured to determine the physiological response intensity corresponding to the target muscle group based on the multimodal feedback data. The modeling module 203 is used to establish a treatment parameter prediction model corresponding to the target muscle group based on the machine learning regression algorithm and the appropriate stimulation parameters and physiological response intensity corresponding to the electromagnetic stimulation. The prediction module 204 is used to predict the ideal treatment parameters corresponding to the target muscle group according to the real-time physiological response intensity corresponding to the target muscle group and the treatment parameter prediction model when the target muscle group is being treated.

[0052] It can be seen that the above-mentioned embodiment of the invention obtains multimodal feedback data when the target muscle group receives electromagnetic stimulation to determine the physiological response intensity, and uses a machine learning regression algorithm to construct a treatment parameter prediction model, and combines the real-time physiological response intensity to predict the ideal treatment parameters, thereby realizing accurate prediction of peripheral muscle treatment parameters based on multimodal feedback, and improving the targetedness and treatment effect of peripheral electromagnetic treatment of muscle groups.

[0053] As an optional embodiment, the multimodal feedback data includes temperature data, humidity data, physical displacement data, electromyographic signal data, and image data.

[0054] It can be seen that through the above optional embodiments, the content of the multimodal feedback data is defined to comprehensively characterize the relevant features of the electromagnetic stimulation feedback of the muscle group, so as to facilitate subsequent accurate intensity determination and modeling prediction, assist in realizing the prediction of accurate muscle peripheral treatment parameters based on multimodal feedback, and improve the pertinence and treatment effect of peripheral electromagnetic treatment of the muscle group.

[0055] As an optional embodiment, the multimodal feedback data is obtained through a multimodal sensing platform arranged on the target muscle group; the multimodal sensing platform includes a temperature sensor, a humidity sensor, a physical displacement infrared sensor, sEMG electrodes, and an image sensor.

[0056] It can be seen that through the above optional embodiments, the details of the physical acquisition device of the multimodal feedback data are defined to accurately and comprehensively extract the relevant features characterizing the electromagnetic stimulation feedback of the muscle group, so as to facilitate subsequent accurate intensity determination and modeling prediction, assist in realizing the prediction of accurate muscle peripheral treatment parameters based on multimodal feedback, and improve the pertinence and treatment effect of peripheral electromagnetic treatment of the muscle group.

[0057] As an optional embodiment, the specific manner in which the determination module determines the physiological response intensity corresponding to the target muscle group based on the multimodal feedback data includes: Input each data in the multimodal feedback data into the response degree prediction model corresponding to the corresponding data type to obtain the response degree parameter corresponding to each data; optionally, the response degree prediction model is trained through a training data set including a plurality of training feedback data corresponding to the data types and the corresponding response degree annotations; Calculate the weighted sum value of the response degree parameters corresponding to all the data to obtain the physiological response intensity corresponding to the target muscle group; optionally, among them, the calculation weight corresponding to each data response degree parameter is the product of the type weight and the model accuracy weight; the type weights decrease in turn when the data types of the corresponding data are electromyogram signal data, physical displacement data, image data, temperature data, and humidity data; the model accuracy weight is proportional to the prediction accuracy of the response degree prediction model corresponding to the corresponding data type in the verification stage.

[0058] It can be seen that through the above optional embodiments, by inputting the multimodal feedback data into the response degree prediction model of the corresponding data type to generate response degree parameters, and calculating the weighted sum value with the product of the type weight and the model accuracy weight as the weight to determine the physiological response intensity of the target muscle group, thereby realizing accurate physiological response evaluation based on weighted analysis of multimodal data, improving the accuracy of subsequent electromagnetic stimulation treatment parameter prediction, assisting in realizing the prediction of accurate muscle peripheral treatment parameters based on multimodal feedback, and improving the pertinence and treatment effect of peripheral electromagnetic treatment of the muscle group.

[0059] As an optional embodiment, when the data type is temperature data, humidity data, or physical displacement data, the model architecture of the corresponding reaction degree prediction model is a CNN neural network; when the data type is image data, the corresponding reaction degree prediction model is an RNN neural network; when the data type is electromyography signal data, the corresponding reaction degree prediction model is a mathematical relationship fitting model, and the reaction degree annotation in the corresponding training dataset is the weighted sum average of the amplitude mean ratio parameter, latency mean ratio parameter, and mean power frequency ratio parameter; the amplitude mean ratio parameter is the ratio of the signal amplitude means corresponding to the electromechanical signal data in the resting state and the stimulated state respectively; the latency mean ratio parameter is the ratio of the signal latency means corresponding to the electromechanical signal data in the resting state and the stimulated state respectively; the mean power frequency ratio parameter is the ratio of the mean power frequencies corresponding to the electromechanical signal data in the resting state and the stimulated state respectively.

[0060] It can be seen that through the above optional embodiments, the details of the prediction models corresponding to different data types are defined. Specifically, the reaction degree of the electromyography signal is based on the weighted sum of the amplitude, latency, and mean power frequency ratio to determine the accurate reaction degree annotation, so as to achieve accurate physiological response evaluation based on multi-modal data weighted analysis, improve the accuracy of subsequent electromagnetic stimulation treatment parameter prediction, assist in realizing the prediction of accurate muscle peripheral treatment parameters based on multi-modal feedback, and improve the pertinence and treatment effect of muscle group peripheral electromagnetic therapy.

[0061] As an optional embodiment, the appropriate stimulation parameters or ideal treatment parameters include at least one of the intensity parameter, pulse width parameter, and frequency parameter of electrical stimulation or magnetic stimulation.

[0062] It can be seen that through the above optional embodiments, the details of the electromagnetic stimulation parameters or treatment parameters are defined to achieve more comprehensive and accurate parameter prediction, assist in realizing the prediction of accurate muscle peripheral treatment parameters based on multi-modal feedback, and improve the pertinence and treatment effect of muscle group peripheral electromagnetic therapy.

[0063] As an optional embodiment, the specific manner in which the prediction module predicts the ideal treatment parameters corresponding to the target muscle group according to the real-time physiological response intensity corresponding to the target muscle group and the treatment parameter prediction model includes: Obtain the first multi-modal feedback data of the target muscle group and the second multi-modal feedback data corresponding to at least one associated muscle group; optionally, there is at least one neural connection channel between the associated muscle group and the target muscle group, and the physical distance between the associated muscle group and the target muscle group is less than a preset distance threshold; According to the first multi-modal feedback data, calculate the first real-time physiological response intensity corresponding to the target muscle group; According to the second multi-modal feedback data, calculate the second real-time physiological response intensity corresponding to at least one associated muscle group; Correcting the first real-time physiological response intensity according to the second real-time physiological response intensity to obtain a corrected physiological response intensity; The corrected physiological response intensity is input into the treatment parameter prediction model to obtain the ideal treatment parameters corresponding to the target muscle group.

[0064] It can be seen that through the above optional embodiments, by obtaining the first multimodal feedback data of the target muscle group and the second multimodal feedback data of the associated muscle group that has a neural connection with the target muscle group and a physical distance less than a threshold, the first real-time physiological response intensity of the target muscle group and the second real-time physiological response intensity of the associated muscle group are calculated respectively, and the latter is used to correct the former to obtain the corrected physiological response intensity, which is input into the treatment parameter prediction model to generate ideal treatment parameters, thereby realizing accurate physiological response correction and treatment parameter prediction based on the correction of the associated muscle group, realizing accurate prediction of peripheral muscle treatment parameters based on multimodal feedback, and improving the targetedness and treatment effect of peripheral electromagnetic treatment of muscle groups.

[0065] As an optional embodiment, the specific manner in which the prediction module corrects the first real-time physiological response intensity according to the second real-time physiological response intensity to obtain the corrected physiological response intensity includes: For each associated muscle group, a weighted sum of the number of neural connection channels corresponding to the associated muscle group and the physical distance is calculated to obtain an associated parameter corresponding to the associated muscle group; According to the corresponding relationship between the preset correlation parameter and the influence ratio, and the correlation parameter, a corresponding influence ratio parameter is determined; optionally, the influence ratio parameter is less than 1; Calculating the product of the second real-time physiological response intensity corresponding to the associated muscle group and the influence ratio parameter to obtain the influence response intensity parameter corresponding to the associated muscle group; The average value of the impact response intensity parameters of all associated muscle groups and the first real-time physiological response intensity is calculated to obtain the corrected physiological response intensity.

[0066] It can be seen that through the above optional embodiments, the association parameters are obtained by calculating the weighted sum of the number of neural connection channels and the physical distance of each associated muscle group, the influence ratio parameter is determined based on the preset corresponding relationship, the product of the second real-time physiological response intensity of the associated muscle group and the influence ratio parameter is calculated to obtain the influence response intensity parameter, and the average value of the influence response intensity parameters of all associated muscle groups and the first real-time physiological response intensity of the target muscle group is taken as the corrected physiological response intensity, thereby realizing accurate physiological response correction based on neural connection and distance weighting, realizing accurate prediction of muscle peripheral treatment parameters based on multimodal feedback, and improving the targetedness and treatment effect of peripheral electromagnetic treatment of muscle groups.

[0067] Example 3 See also Figure 3 ,Figure 3 Another peripheral stimulation parameter prediction system based on multimodal feedback disclosed by an embodiment of the present invention. Figure 3 The described peripheral stimulation parameter prediction system based on multimodal feedback is applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). As Figure 3 shown, the peripheral stimulation parameter prediction system based on multimodal feedback may include: A memory 301 storing executable program code; A processor 302 coupled to the memory 301; Wherein, the processor 302 calls the executable program code stored in the memory 301 to execute the steps of the peripheral stimulation parameter prediction method described in Embodiment 1.

[0068] Embodiment 4 An embodiment of the present invention discloses a computer-readable storage medium storing a computer program for electronic data exchange, wherein the computer program causes a computer to execute the steps of the peripheral stimulation parameter prediction method described in Embodiment 1.

[0069] Embodiment 5 An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute the steps of the peripheral stimulation parameter prediction method described in Embodiment 1.

[0070] The above describes specific embodiments of this specification, and other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the figures do not necessarily have to be performed in the particular order or continuous order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0071] The systems, devices, modules or units illustrated in the above embodiments may be specifically implemented by a computer chip or entity, or by a product with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0072] For convenience of description, when describing the above device, various units are described separately according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware.

[0073] Those skilled in the art should understand that the embodiments of this specification can be provided as a method, a system, or a computer program product. Therefore, the embodiments of this specification can take the form of an all-hardware embodiment, an all-software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0074] This specification is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of this specification. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the specified functions in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0075] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device that implements the specified functions in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0076] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are performed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the specified functions in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0077] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and a memory.

[0078] Memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0079] Computer-readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0080] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0081] This specification may be described in the general context of computer-executable instructions, such as program modules, executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media, including storage devices.

[0082] The various embodiments in this specification are described in a progressive manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the system embodiments are generally similar to the method embodiments, so the description is relatively simple. For relevant parts, refer to the description of the method embodiments.

[0083] Finally, it should be noted that the peripheral stimulation parameter prediction method and system based on multimodal feedback disclosed in the embodiments of the present invention are only preferred embodiments of the present invention, and are only used to illustrate the technical solutions of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for predicting peripheral stimulation parameters based on multimodal feedback, characterized in that The method includes: Obtaining corresponding multi-modal feedback data when electromagnetic stimulation is applied to the target muscle group; Determining the physiological response intensity corresponding to the target muscle group based on the multi-modal feedback data; Establishing a treatment parameter prediction model corresponding to the target muscle group based on a machine learning regression algorithm according to the appropriate stimulation parameters corresponding to the electromagnetic stimulation and the physiological response intensity; When treating the target muscle group, predicting the ideal treatment parameters corresponding to the target muscle group according to the real-time physiological response intensity corresponding to the target muscle group and the treatment parameter prediction model.

2. The peripheral stimulation parameter prediction method based on multimodal feedback according to claim 1, wherein The multi-modal feedback data includes temperature data, humidity data, physical displacement data, electromyogram signal data, and image data.

3. The method for predicting peripheral stimulation parameters based on multimodal feedback according to claim 1, wherein The multi-modal feedback data is obtained through a multi-modal sensing platform arranged on the target muscle group; the multi-modal sensing platform includes a temperature sensor, a humidity sensor, a physical displacement infrared sensor, sEMG electrodes, and an image sensor.

4. The method for predicting peripheral stimulation parameters based on multimodal feedback according to claim 2, wherein The determining the physiological response intensity corresponding to the target muscle group based on the multi-modal feedback data includes: Inputting each data in the multi-modal feedback data into a response degree prediction model corresponding to the corresponding data type to obtain a response degree parameter corresponding to each data; the response degree prediction model is trained through a training data set including a plurality of training feedback data corresponding to the corresponding data type and corresponding response degree annotations; Calculating the weighted sum value of the response degree parameters corresponding to all data to obtain the physiological response intensity corresponding to the target muscle group; wherein, the calculation weight corresponding to each data response degree parameter is the product of the type weight and the model accuracy weight; the type weight decreases in turn when the data types of the corresponding data are electromyogram signal data, physical displacement data, image data, temperature data, and humidity data; the model accuracy weight is proportional to the prediction accuracy of the response degree prediction model corresponding to the corresponding data type in the verification stage.

5. The method for predicting peripheral stimulation parameters based on multimodal feedback according to claim 4, wherein When the data type is temperature data, humidity data, or physical displacement data, the model architecture of the corresponding response degree prediction model is a CNN neural network; when the data type is image data, the corresponding response degree prediction model is an RNN neural network; when the data type is electromyogram signal data, the corresponding response degree prediction model is a mathematical relationship fitting model, and the response degree annotation in the corresponding training data set is the weighted sum average of the amplitude mean ratio parameter, the latency mean ratio parameter, and the mean power frequency ratio parameter; the amplitude mean ratio parameter is the ratio of the signal amplitude means corresponding to the electromechanical signal data in the resting state and the stimulated state respectively; the latency mean ratio parameter is the ratio of the signal latency means corresponding to the electromechanical signal data in the resting state and the stimulated state respectively; the mean power frequency ratio parameter is the ratio of the mean power frequencies corresponding to the electromechanical signal data in the resting state and the stimulated state respectively.

6. The method for predicting peripheral stimulation parameters based on multimodal feedback according to claim 1, wherein The appropriate stimulation parameters or the ideal treatment parameters include at least one of the intensity parameter, pulse width parameter, and frequency parameter of electrical stimulation or magnetic stimulation.

7. The method for predicting peripheral stimulation parameters based on multimodal feedback according to claim 1, wherein Predicting the ideal treatment parameters corresponding to the target muscle group according to the real-time physiological response intensity corresponding to the target muscle group and the treatment parameter prediction model includes: Obtaining first multi-modal feedback data of the target muscle group and second multi-modal feedback data of at least one associated muscle group; there is at least one neural connection channel between the associated muscle group and the target muscle group, and the physical distance between the associated muscle group and the target muscle group is less than a preset distance threshold; Calculating a first real-time physiological response intensity corresponding to the target muscle group according to the first multi-modal feedback data; Calculating second real-time physiological response intensities corresponding to at least one of the associated muscle groups according to the second multi-modal feedback data; Correcting the first real-time physiological response intensity according to the second real-time physiological response intensity to obtain a corrected physiological response intensity; Inputting the corrected physiological response intensity into the treatment parameter prediction model to obtain the ideal treatment parameters corresponding to the target muscle group.

8. The method for predicting peripheral stimulation parameters based on multimodal feedback according to claim 7, wherein The correcting the first real-time physiological response intensity according to the second real-time physiological response intensity to obtain a corrected physiological response intensity includes: For each of the associated muscle groups, calculating a weighted sum value between the number of channels of the neural connection channel corresponding to the associated muscle group and the physical distance to obtain an associated parameter corresponding to the associated muscle group; Determining a corresponding influence ratio parameter according to the corresponding relationship between the preset associated parameter and the influence ratio and the associated parameter; the influence ratio parameter is less than 1; Calculating the product of the second real-time physiological response intensity corresponding to the associated muscle group and the influence ratio parameter to obtain an influence response intensity parameter corresponding to the associated muscle group; Calculating the average value of the influence response intensity parameters of all the associated muscle groups and the first real-time physiological response intensity to obtain a corrected physiological response intensity.

9. A peripheral stimulation parameter prediction system based on multimodal feedback, characterized in that, The system includes: An acquisition module for acquiring corresponding multi-modal feedback data when electromagnetic stimulation is applied to the target muscle group; A determination module for determining the physiological response intensity corresponding to the target muscle group based on the multi-modal feedback data; A modeling module for establishing a treatment parameter prediction model corresponding to the target muscle group according to the appropriate stimulation parameters corresponding to the electromagnetic stimulation and the physiological response intensity based on a machine learning regression algorithm; A prediction module for predicting the ideal treatment parameters corresponding to the target muscle group according to the real-time physiological response intensity corresponding to the target muscle group and the treatment parameter prediction model when the target muscle group is being treated.

10. A peripheral stimulation parameter prediction system based on multimodal feedback, characterized in that, The system includes: A memory storing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory and executes the multi-modal feedback-based peripheral stimulation parameter prediction method according to any one of claims 1-8.

Images