Hammering management system and method of hammering stretching physiotherapy rehabilitation machine

Abstract

The invention discloses a hammering management system and method of a hammering stretching physiotherapy rehabilitation machine, relates to the technical field of physiotherapy massage machine control, and solves the technical problem that in the prior art, an existing massage machine is difficult to adaptively adjust the force or frequency of a massage hammer according to the state of a user, so that the experience feeling of the user is reduced. Acquiring description voice of a user, preprocessing the description voice, and inputting the preprocessed description voice into the language recognition model to obtain a massage part and a massage scheme; the massage scheme comprises massage strength and massage frequency; in the massage process, an adjusting massage scheme is generated based on the expression data or the physiological monitoring data obtained in real time, and the massage scheme is adjusted based on the adjusting massage scheme; the expression data comprises voice data and image data of a user, and the physiological monitoring data comprises electromyographic signals, muscle pressure and muscle temperature of a massage part; and the experience feeling of the massaged person is greatly improved.

Description

Technical Field

[0001] The present application belongs to the field of physical therapy and massage machine control technology, and specifically is a hammer management system and method for a hammer stretching physical therapy rehabilitation machine. Background Art

[0002] With the continuous advancement of science and technology and the increasing maturity of robotics, the emergence of service robots has become the best solution to this social contradiction. Currently, the emergence of intelligent robots can replace manual operations, but intelligent robots have not yet entered the traditional Chinese medicine fascia massage therapy industry. However, due to their safe, stable, and reliable technical advantages, intelligent robots can be used in the personal health therapy industry.

[0003] Existing massage machines adjust the hammering force and frequency of the massage hammer head according to the needs of the user. If the massage force and frequency need to be adjusted during the massage, the user needs to control it by himself or ask relevant personnel to control the massage machine to adjust the massage force and frequency. Manual adjustment of the massage hammer force often requires the user and the relevant personnel who adjust the machine to have a more precise control over the massage hammer force and frequency. Adjusting the massage hammer force and frequency according to the subjective feelings of the adjuster may result in the adjustment force being too large or too small, thereby reducing the user's experience. This affects the need for a hammer management system and method for a hammer stretching physiotherapy rehabilitation machine. Summary of the Invention

[0004] The present application provides a hammer management system and method for a hammer stretching physiotherapy rehabilitation machine, which solves the technical problem that existing massage machines in the prior art are difficult to adaptively adjust the strength or frequency of the massage hammer according to the user's status, resulting in a reduced user experience.

[0005] To achieve the above objectives, this application adopts the following technical solutions: In a first aspect, a hammering management method for a hammering stretching physiotherapy rehabilitation machine is provided, comprising: Acquire the user's descriptive speech, pre-process the descriptive speech, and then input it into the language recognition model, wherein the descriptive speech pre-processing includes recognizing the speech and converting it into descriptive text; obtain the massage part and massage plan; the massage plan includes massage intensity and massage frequency; Massage the massage areas based on the massage plan; During the massage process, an adjustment massage plan is generated based on the performance data or physiological monitoring data obtained in real time, and the massage plan is adjusted based on the adjustment massage plan; the performance data includes the user's voice data and image data; the physiological monitoring data includes the electromyographic signal, muscle pressure and muscle temperature of the massage part; the electromyographic signal is obtained by an electrode sheet set on the massage part, the muscle pressure signal is obtained by a pressure sensor installed on the massage head, and the muscle temperature is obtained by a temperature sensor installed on the massage head.

[0006] Based on the above technical solution, in a hammer management system and method of a hammer stretching physiotherapy rehabilitation machine provided in the present application, the descriptive voice of the user is obtained, the descriptive voice is pre-processed and then input into a language recognition model to obtain the massage parts and massage plan; the massage hammer of the massage machine is controlled to massage the corresponding massage parts; during the massage process, an adjustment massage plan is generated based on the performance data or physiological monitoring data obtained in real time, and the massage plan is adjusted based on the adjustment massage plan; the strength and frequency of the massage are adjusted in real time according to the subjective performance and physiological performance of the person being massaged during the massage process; thereby greatly improving the experience of the person being massaged.

[0007] In conjunction with the first aspect above, in one possible implementation, a training method of the language recognition model includes: Acquire descriptive speech of several users, convert the descriptive speech into descriptive text, extract several keywords from the descriptive text, and massage parts and massage plans corresponding to the descriptive text; the massage parts and massage plans are selected by experts based on the user's descriptive text or the keywords corresponding to the descriptive text; integrate the descriptive text, the several keywords corresponding to the descriptive text, and the massage parts and massage plans corresponding to the descriptive text into several training data; integrate the descriptive text, the massage parts and massage plans corresponding to the descriptive text into several test data; The artificial intelligence model is trained using training data, and the trained artificial intelligence model is tested using test data; the final input is a description text, and the output is the massage parts and massage plans corresponding to the description text; wherein the artificial intelligence model includes a recurrent neural network model, etc.; it can be understood that in the method for training the model provided in this embodiment, the training data includes keywords of the description text, and the model execution process includes extracting keywords, analyzing keywords, and matching massage parts and massage plans; there is no need to perform analysis based on the full text of the description text, which greatly reduces the amount of data processing and facilitates increasing the efficiency of model processing.

[0008] In combination with the first aspect above, in a possible implementation, generating the adjusted massage plan based on the performance data acquired in real time includes: Extracting voice data and image data from the performance data; the image data being a facial image of the user; and converting the voice data into descriptive text; The descriptive text and facial image are input into a trained hybrid recognition model to obtain adjustment force and adjustment frequency; the adjustment force and adjustment frequency are integrated into an adjustment massage plan; the adjustment force includes an adjustment direction and an adjustment force value, the adjustment direction includes a positive direction and a negative direction, the positive direction indicates an increase in force; the negative direction indicates a decrease in force; the adjustment frequency includes an adjustment direction and an adjustment frequency value; the adjustment direction includes a positive direction and a negative direction, the positive direction indicates an increase in frequency; the negative direction indicates a decrease in frequency.

[0009] In combination with the first aspect above, in a possible implementation, a training method of the hybrid recognition model includes: Acquire a plurality of descriptive texts and facial images; and the adjustment strengths and adjustment frequencies corresponding to the descriptive texts and facial images; integrate the plurality of descriptive texts and facial images, and the adjustment strengths and adjustment frequencies corresponding to the descriptive texts and facial images into a plurality of training data and test data; The artificial intelligence model is trained using training data, and the trained artificial intelligence model is tested using a test model, and finally the input is a descriptive text and a facial image, and the output is the adjustment strength and adjustment frequency corresponding to the descriptive text and the facial image; wherein the artificial intelligence model includes a recurrent neural network model.

[0010] In combination with the first aspect above, in a possible implementation, generating the adjusted massage plan based on physiological monitoring data acquired in real time includes: Extracting electromyographic signals, muscle pressure, and muscle temperature from physiological monitoring data; generating a muscle relaxation score based on the electromyographic signals, muscle pressure, and muscle temperature; When the muscle relaxation score is greater than the set relaxation threshold, it means that the massage has achieved the desired effect, and a stop massage signal is generated; When the muscle relaxation score is less than the set relaxation threshold, it means that the massage at this time has not achieved the desired effect; an adjustment massage plan is generated based on the muscle relaxation score.

[0011] In combination with the first aspect above, in a possible implementation, a method for generating the muscle relaxation score includes: Acquire a number of physiological monitoring data within a set time; fit the electromyographic signals in each physiological monitoring data into an electromyographic signal change curve according to the time sequence of their corresponding acquisition; extract the average amplitude EMGP of the electromyographic signal change curve; Generate a muscle pressure characteristic change curve MP(n) based on the muscle pressure in each physiological monitoring data; Generate a muscle temperature change curve MT(t) based on the muscle temperature in each physiological monitoring data; Substitute the average amplitude EMGP, the muscle pressure characteristic change curve MP(n), and the muscle temperature change curve MT(t) into the muscle relaxation evaluation function to obtain the muscle relaxation score SP; the muscle relaxation evaluation function is:

[0012] Where SP is the muscle relaxation score, H1() is the quantization function of the set muscle temperature and muscle pressure, and H2() is the quantization function of the set electromyographic signal; δ1 is the weight coefficient of the influence of muscle temperature and muscle pressure on muscle relaxation, and δ2 is the weight coefficient of the influence of electromyographic signal on muscle relaxation, δ1+δ2=1, δ1<δ2; its specific values ​​are set according to experience.

[0013] In combination with the first aspect above, in a possible implementation, generating a muscle temperature change curve based on the muscle temperature in each physiological monitoring data includes: Obtain muscle temperatures from a number of physiological monitoring data, and sort the muscle temperatures in order of their corresponding acquisition time to obtain a muscle temperature array; Obtain a motion cycle corresponding to the massage frequency of the massage hammer, obtain all muscle temperatures in the first motion cycle in the muscle temperature data group, mark the number of muscle temperatures as K, sequentially select a number of muscle temperatures from the muscle temperature data group using each muscle temperature as a starting point and the cycle as a step length, and divide the muscle temperature group into K muscle temperature change groups; The muscle temperatures in the muscle temperature change group are fitted into a muscle temperature change curve MTi(t) according to their corresponding acquisition time; where i is the number corresponding to the muscle temperature change group; i=1, 2, ..., K; By formula The muscle temperature change curve is selected as the final muscle temperature change curve MT(t); wherein T is the duration of the set time, that is, the time length for collecting a plurality of physiological monitoring data.

[0014] In combination with the first aspect above, in a possible implementation, generating a muscle pressure characteristic change curve based on muscle pressure in each physiological monitoring data includes: Fit the muscle pressure in each physiological monitoring data into the original pressure change curve YP(t) according to the time sequence of its corresponding collection; By formula The pressure characteristic MPn of the muscle pressure of the massage hammer in the nth movement cycle is calculated; wherein n is the number of the movement cycle, and f is the massage frequency of the massage hammer; Each pressure feature is fitted into a pressure feature variation curve MP(n) according to the number sequence of the motion cycles.

[0015] In combination with the first aspect above, in a possible implementation, generating the adjustment strength and adjustment frequency based on the multiple muscle relaxation scores includes: Obtain the muscle relaxation score generated this time, as well as the muscle relaxation score and muscle relaxation threshold generated last time, marked as DSP, XSP and SPY respectively; by the formula:

[0016] The adjustment ratio TB is calculated; wherein μ is the set adjustment coefficient, which is used to adjust the distribution of the adjustment ratio TB; When the adjustment ratio TB is less than zero, the adjustment direction of the adjustment force and adjustment frequency is set to negative, the adjustment force value is set to 1+TB times the original force value, and the adjustment frequency value is set to 1+TB times the original frequency value; When the adjustment ratio TB is greater than zero, the adjustment direction in the adjustment force and adjustment frequency is set to positive, the adjustment force value is set to 1+TB times the original force value, and the adjustment frequency value is set to 1+TB times the original frequency value.

[0017] In a second aspect, the present application provides a hammer management device for a hammer-stretching physiotherapy and rehabilitation machine, comprising: a processor and a storage medium; the storage medium comprising instructions, the processor being configured to execute the instructions to implement the method described in the first aspect and any possible implementation of the first aspect. The hammer management device for the hammer-stretching physiotherapy and rehabilitation machine can be an electronic device or a chip within an electronic device.

[0018] In a third aspect, the present application provides a hammer management system for a hammer stretching physiotherapy rehabilitation machine, comprising: a data acquisition module, a data analysis module, and a massage adjustment module; The data acquisition module includes an image acquisition unit, a voice acquisition unit and a physiological data acquisition unit; The image acquisition unit is used to acquire facial images of the user, and the facial images are integrated into image data; The voice collection unit is used to collect the user's descriptive voice and voice data; The physiological data acquisition unit is used to collect the user's physiological monitoring data, and the physiological monitoring data includes electromyographic signals, muscle pressure and muscle temperature of the massage part; The data analysis module is used to obtain the user's descriptive speech, pre-process the descriptive speech and input it into the language recognition model to obtain the massage parts and massage plan; and generate and adjust the massage plan based on the performance data or physiological monitoring data obtained in real time during the massage process; the performance data includes the user's voice data and image data; The massage adjustment module is used to perform massage based on the massage plan and adjust the massage plan based on the adjusted massage plan.

[0019] In a fourth aspect, the present application provides a computer-readable storage medium, which stores instructions. When the instructions are run on a hammer management device of a hammer stretching physiotherapy and rehabilitation machine, the hammer management device of the hammer stretching physiotherapy and rehabilitation machine executes the method described in the first aspect and any possible implementation of the first aspect.

[0020] In the fifth aspect, the present application provides a computer program product comprising instructions, which, when run on a hammer management device of a hammer stretching physiotherapy and rehabilitation machine, enables the hammer management device of the hammer stretching physiotherapy and rehabilitation machine to perform the method described in the first aspect and any possible implementation of the first aspect.

[0021] The present application provides a hammer management system and method for a hammer stretching physiotherapy rehabilitation machine, which can obtain the user's descriptive voice, pre-process the descriptive voice, and then input it into a language recognition model to obtain massage parts and massage plans; control the massage hammers of the massage machine to massage the corresponding massage parts; generate an adjustment massage plan based on real-time performance data or physiological monitoring data during the massage process, and adjust the massage plan based on the adjustment massage plan; adjust the strength and frequency of the massage in real time according to the subjective performance and physiological performance of the person being massaged during the massage process; thereby greatly improving the experience of the person being massaged.

[0022] It should be understood that the description of technical features, technical solutions, beneficial effects or similar language in this application does not imply that all features and advantages can be realized in any single embodiment. On the contrary, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution or beneficial effect is included in at least one embodiment. Therefore, the description of a technical feature, technical solution or beneficial effect in this specification does not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions and beneficial effects described in the present embodiment can also be combined in any appropriate manner. Those skilled in the art will understand that the embodiment can be implemented without one or more specific technical features, technical solutions or beneficial effects of a specific embodiment. In other embodiments, additional technical features and beneficial effects can also be identified in specific embodiments that do not embody all embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0024] Figure 1 A schematic diagram of the steps of a hammering management method for a hammering stretching physiotherapy rehabilitation machine in this application; Figure 2 This is a schematic diagram of the module connections of a hammer management system for a hammer stretching physiotherapy rehabilitation machine in this application. DETAILED DESCRIPTION

[0025] The following will clearly and completely describe the technical solutions of this application in conjunction with the embodiments. Obviously, the embodiments described are only a part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0026] See also Figure 1 In a first aspect, a hammering management method for a hammering stretching physiotherapy rehabilitation machine is provided, comprising: Acquiring a user's descriptive speech, pre-processing the descriptive speech, and then inputting it into a language recognition model, wherein the descriptive speech pre-processing includes recognizing the speech and converting it into descriptive text; obtaining a massage part and a massage plan; wherein the massage plan includes massage intensity and massage frequency; and massaging the massage part according to the massage plan, wherein the massage part is the muscle part of the user to be massaged obtained through speech recognition; During the massage process, an adjustment massage plan is generated based on the performance data or physiological monitoring data obtained in real time, and the massage plan is adjusted based on the adjustment massage plan; the performance data includes the user's voice data and image data; the physiological monitoring data includes the electromyographic signal, muscle pressure and muscle temperature of the massage part; the electromyographic signal is obtained by an electrode sheet set on the massage part, the muscle pressure signal is obtained by a pressure sensor installed on the massage head, and the muscle temperature is obtained by a temperature sensor installed on the massage head; it can be understood that the massage machine includes at least one robotic arm and a massage hammer head.

[0027] Based on the above technical solution, in a hammer management system and method of a hammer stretching physiotherapy rehabilitation machine provided in the present application, the descriptive voice of the user is obtained, the descriptive voice is pre-processed and then input into a language recognition model to obtain the massage parts and massage plan; the massage hammer of the massage machine is controlled to massage the corresponding massage parts; during the massage process, an adjustment massage plan is generated based on the performance data or physiological monitoring data obtained in real time, and the massage plan is adjusted based on the adjustment massage plan; the strength and frequency of the massage are adjusted in real time according to the subjective performance and physiological performance of the person being massaged during the massage process; thereby greatly improving the experience of the person being massaged.

[0028] In one possible implementation, a training method of the language recognition model includes: Acquire descriptive speech of several users, convert the descriptive speech into descriptive text, extract several keywords from the descriptive text, and massage parts and massage plans corresponding to the descriptive text; the massage parts and massage plans are selected by experts based on the user's descriptive text or the keywords corresponding to the descriptive text; integrate the descriptive text, the several keywords corresponding to the descriptive text, and the massage parts and massage plans corresponding to the descriptive text into several training data; integrate the descriptive text, the massage parts and massage plans corresponding to the descriptive text into several test data; The artificial intelligence model is trained using training data, and the trained artificial intelligence model is tested using test data; specifically, the descriptive text in the test data is input into the trained artificial intelligence model, and the corresponding massage parts and massage plans are output. When the output massage parts are the same as the massage parts in the test data, and the difference between the output massage plan and the massage plan in the test data is within an acceptable range, it means that the set of test data has passed the test, otherwise, the relevant parameters of the artificial intelligence model need to be adjusted; until a set proportion of test data passes the test, and the set proportion is at least 80%; finally, the input is the descriptive text, and the output is the massage parts and massage plans corresponding to the descriptive text; wherein the artificial intelligence model includes a recurrent neural network model, etc.; it can be understood that in the method for training the model provided in this embodiment, the training data includes keywords of the descriptive text, and the model execution process includes extracting keywords, analyzing keywords, and matching massage parts and massage plans; there is no need to analyze based on the full text of the descriptive text, which greatly reduces the amount of data processing and facilitates increasing the efficiency of model processing.

[0029] In a possible implementation, generating the adjusted massage plan based on the performance data acquired in real time includes: extracting voice data and image data from the performance data; the image data is a facial image of the user; converting the voice data into a descriptive text; Input the description text and facial image into the trained hybrid recognition model to obtain the adjustment intensity and adjustment frequency; integrate the adjustment intensity and adjustment frequency into an adjustment massage plan; the adjustment intensity includes an adjustment direction and an adjustment intensity value, the adjustment direction includes a positive direction and a negative direction, the positive direction indicates an increase in intensity; the negative direction indicates a decrease in intensity; the adjustment frequency includes an adjustment direction and an adjustment frequency value; the adjustment direction includes a positive direction and a negative direction, the positive direction indicates an increase in frequency; the negative direction indicates a decrease in frequency.

[0030] In a possible implementation manner, one training method of the hybrid recognition model includes: Obtain a number of description texts and facial images; and the corresponding adjustment intensity and adjustment frequency of the description texts and facial images; the adjustment intensity and adjustment frequency are set by experts according to the description texts and facial images; when keywords such as "pain", "ache", "ah" appear in the description text, the adjustment intensity needs to be set smaller than the current intensity; the adjustment frequency also needs to be set smaller than the current frequency; when the expression recognized from the facial image is pain, etc., the adjustment intensity needs to be set smaller than the current intensity; the adjustment frequency also needs to be set smaller than the current frequency; integrate a number of description texts and facial images, and the corresponding adjustment intensity and adjustment frequency of the description texts and facial images into a number of training data and test data; Use the training data to train the artificial intelligence model, use the test model to test the trained artificial intelligence model, and finally obtain an input of a description text and a facial image, and an output of the corresponding adjustment intensity and adjustment frequency of the description text and the facial image; wherein, the artificial intelligence model includes a recurrent neural network model; the training process of the text and image recognition model is an existing artificial intelligence training process, which will not be described in detail here; in this embodiment, by adjusting the intensity value and direction, and the frequency value and direction, the adjustment direction and the intensity value, as well as the adjustment direction and the frequency value can be mutually verified, avoiding errors in intensity or frequency adjustment.

[0031] In a possible implementation manner, generating the adjustment massage plan based on the real-time acquired physiological monitoring data includes: extracting the electromyogram signal, muscle pressure and muscle temperature from the physiological monitoring data; generating a muscle relaxation score based on the electromyogram signal, muscle pressure and muscle temperature; the muscle relaxation score is a score for evaluating whether the muscle is relaxed, and the larger the value of the muscle relaxation score, the closer the muscle is to relaxation; When the muscle relaxation score is greater than the set relaxation threshold, it means that the massage has achieved the desired effect and a stop massage signal is generated. The stop massage signal is used to control the massage to stop. The muscle relaxation threshold is a value set by the expert. When it is greater than this value, it means that the muscle is in a good state of relaxation. When it is less than this value, it means that the muscle is in a stiff or poor state of relaxation. The lower the relaxation score, the closer the muscle is to a stiff state. When the muscle relaxation score is less than the set relaxation threshold, it means that the massage at this time has not achieved the desired effect; an adjustment massage plan is generated based on the muscle relaxation score.

[0032] In one possible implementation, a method for generating the muscle relaxation score includes: Acquire a number of physiological monitoring data within a set time; fit the electromyographic signals in each physiological monitoring data into an electromyographic signal change curve according to the time sequence of their corresponding acquisition; extract the average amplitude EMGP of the electromyographic signal change curve; Generate a muscle pressure characteristic change curve MP(n) based on the muscle pressure in each physiological monitoring data; Generate a muscle temperature change curve MT(t) based on the muscle temperature in each physiological monitoring data; Substitute the average amplitude EMGP, the muscle pressure characteristic change curve MP(n), and the muscle temperature change curve MT(t) into the muscle relaxation evaluation function to obtain the muscle relaxation score SP; the muscle relaxation evaluation function is:

[0033] Where SP is the muscle relaxation score, H1() is the quantization function of the set muscle temperature and muscle pressure, and H2() is the quantization function of the set electromyographic signal; δ1 is the weight coefficient of the effect of muscle temperature and muscle pressure on muscle relaxation, and δ2 is the weight coefficient of the effect of electromyographic signal on muscle relaxation, δ1+δ2=1, δ1<δ2; its specific values ​​are set according to experience; For example, the quantization functions of muscle temperature and muscle pressure are:

[0034] Among them, MT(T) is the final value of the muscle temperature change curve, and MT(0) is the initial value of the muscle temperature change curve; MP(N) is the final value of the muscle pressure characteristic change curve, and MP(0) is the initial value of the muscle pressure characteristic change curve; this embodiment quantifies the influence of muscle temperature and muscle pressure on the muscle relaxation score through the above formula. When the muscle changes from stiffness to relaxation, the muscle temperature and muscle pressure will tend to be stable. When the muscle temperature or muscle pressure tends to be stable, it means that the muscle has reached a better relaxation state at this time; the more relaxed the muscle is, the larger the value of the corresponding quantification result will be.

[0035] The quantization function of the electromyographic signal is:

[0036] Among them, WEMG is the set EMG relaxation standard value, that is, the average value of the EMG signal when the muscle is in a relaxed state. When the muscle goes from stiff to relaxed, the internal activity of the muscle becomes higher, and the value of the EMG signal will increase, and eventually tend to or exceed the EMG relaxation standard value. When the value of the EMG signal is larger, the value of the corresponding quantization result will also become larger.

[0037] In a possible implementation, generating a muscle temperature change curve based on muscle temperatures in each physiological monitoring data includes: acquiring muscle temperatures in a plurality of physiological monitoring data, and sequentially sorting the muscle temperatures according to their corresponding acquisition time sequence to obtain a muscle temperature array; Obtain a motion cycle corresponding to the massage frequency of the massage hammer, obtain all muscle temperatures in the first motion cycle in the muscle temperature data group, mark the number of muscle temperatures as K, sequentially select a number of muscle temperatures from the muscle temperature data group using each muscle temperature as a starting point and the cycle as a step length, and divide the muscle temperature group into K muscle temperature change groups; The muscle temperatures in the muscle temperature change group are fitted into a muscle temperature change curve MTi(t) according to their corresponding acquisition time; where i is the number corresponding to the muscle temperature change group; i=1, 2, ..., K; By formula The muscle temperature change curve is selected as the final muscle temperature change curve MT(t); wherein T is the duration of the set time, that is, the time length for collecting a plurality of physiological monitoring data.

[0038] Since the degree of contact between the massage hammer and the skin varies during the massage process, the muscle temperature collected by the temperature sensor on the massage hammer may fluctuate slightly. In this embodiment, the muscle temperatures measured at the same location by the massage hammer are grouped into the same muscle temperature change group, and the muscle temperature change group with the largest comprehensive temperature is selected to generate a muscle temperature change curve for subsequent analysis. That is, the muscle temperature change data of the group where the massage hammer is in the closest contact with the skin is selected as the muscle temperature reference for thickness analysis. This can avoid inaccurate muscle temperature measurements caused by substances such as air between the massage hammer and the skin. The muscle temperature in the group where the massage hammer is in the closest contact with the skin is more accurate, thereby ensuring the accuracy of subsequent analysis.

[0039] In a possible implementation, generating a muscle pressure characteristic change curve based on the muscle pressure in each physiological monitoring data includes: Fit the muscle pressure in each physiological monitoring data into the original pressure change curve YP(t) according to the time sequence of its corresponding collection; By formula The pressure characteristic MPn of the muscle pressure of the massage hammer in the nth movement cycle is calculated; wherein n is the number of the movement cycle, and f is the massage frequency of the massage hammer; Each pressure feature is fitted into a pressure feature variation curve MP(n) according to the number sequence of the motion cycles.

[0040] For example, during the massage process, the massage hammer contacts the skin until it presses to the deepest point and then leaves the skin, which is a complete massage hammer movement cycle. The force collected by the pressure sensor on the massage hammer will first change from small to large, and then from large to small. When the muscle is in a stiff state, the buffering effect of the muscle is weak, so the rate of change of muscle pressure is faster, and the final maximum value of muscle pressure will be larger; when the muscle is in a relaxed state, the buffering effect of the muscle is stronger, so the rate of change of muscle pressure is slower, and the final maximum value of muscle pressure will be smaller.

[0041] This embodiment analyzes the presentation of muscle pressure in different exercise cycles, that is, dynamically evaluates the massage effect based on the changes in muscles, making the evaluation result more accurate.

[0042] In a possible implementation, generating the adjustment strength and frequency based on a plurality of muscle relaxation scores includes: Obtain the muscle relaxation score generated this time, as well as the muscle relaxation score and muscle relaxation threshold generated last time, marked as DSP, XSP and SPY respectively; by the formula:

[0043] The adjustment ratio TB is calculated. It is understood that the adjustment ratio TB∈(-1,1) calculated by the above formula can be limited by controlling the adjustment coefficient μ. Here, μ is a set adjustment coefficient used to adjust the distribution of the adjustment ratio TB. The specific value is set based on expert experience. In this embodiment, μ=0.2, and the adjustment ratio TB∈(-0.1,0.1) is within. When the adjustment ratio TB is less than zero, the adjustment direction of the adjustment force and adjustment frequency is set to negative, the adjustment force value is set to 1+TB times the original force value, and the adjustment frequency value is set to 1+TB times the original frequency value; When the adjustment ratio TB is greater than zero, the adjustment direction in the adjustment force and adjustment frequency is set to positive, the adjustment force value is set to 1+TB times the original force value, and the adjustment frequency value is set to 1+TB times the original frequency value.

[0044] In this embodiment, when the current muscle relaxation score is greater than the previous muscle relaxation score, this indicates that the massage effect is better. The massage intensity and frequency can be appropriately lowered to avoid muscle damage caused by continuous high-intensity massage. Otherwise, the massage intensity and frequency need to be appropriately increased. When the current muscle relaxation score is closer to the muscle relaxation score threshold, the amplitude of lowering the massage intensity and frequency needs to be increased to avoid muscle loss caused by excessive massage. Otherwise, the massage intensity and frequency need to be appropriately increased.

[0045] In a second aspect, the present application provides a hammer management device for a hammer-stretching physiotherapy and rehabilitation machine, comprising: a processor and a storage medium; the storage medium comprising instructions, the processor being configured to execute the instructions to implement the method described in the first aspect and any possible implementation of the first aspect. The hammer management device for the hammer-stretching physiotherapy and rehabilitation machine can be an electronic device or a chip within an electronic device.

[0046] See also Figure 2 In a third aspect, the present application provides a hammer management system for a hammer stretching physiotherapy rehabilitation machine, comprising: a data acquisition module, a data analysis module and a massage adjustment module; The data acquisition module includes an image acquisition unit, a voice acquisition unit and a physiological data acquisition unit; The image acquisition unit is used to acquire facial images of the user, and the facial images are integrated into image data; The voice collection unit is used to collect the user's descriptive voice and voice data; The physiological data acquisition unit is used to collect the user's physiological monitoring data, and the physiological monitoring data includes electromyographic signals, muscle pressure and muscle temperature of the massage part; The data analysis module is used to obtain the user's descriptive speech, pre-process the descriptive speech and input it into the language recognition model to obtain the massage parts and massage plan, and generate and adjust the massage plan based on the performance data or physiological monitoring data obtained in real time during the massage process; the performance data includes the user's voice data and image data; The massage adjustment module is used to perform massage based on the massage plan and adjust the massage plan based on the adjusted massage plan.

[0047] In a fourth aspect, the present application provides a computer-readable storage medium, which stores instructions. When the instructions are run on a hammer management device of a hammer stretching physiotherapy and rehabilitation machine, the hammer management device of the hammer stretching physiotherapy and rehabilitation machine executes the method described in the first aspect and any possible implementation of the first aspect.

[0048] In the fifth aspect, the present application provides a computer program product comprising instructions, which, when run on a hammer management device of a hammer stretching physiotherapy and rehabilitation machine, enables the hammer management device of the hammer stretching physiotherapy and rehabilitation machine to perform the method described in the first aspect and any possible implementation of the first aspect.

[0049] Some of the data in the above formula are calculated by removing the dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0050] How this application works: By obtaining the user's descriptive voice, the descriptive voice is pre-processed and then input into the language recognition model to obtain the massage parts and massage plan; the massage hammer of the massage machine is controlled to massage the corresponding massage parts; during the massage process, an adjustment massage plan is generated based on the performance data or physiological monitoring data obtained in real time, and the massage plan is adjusted based on the adjustment massage plan; the massage strength and frequency are adjusted in real time according to the subjective performance and physiological performance of the person being massaged during the massage process; thereby greatly improving the experience of the person being massaged.

[0051] The above embodiments are only used to illustrate the technical method of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present application.

Claims

1. A hammering management method for a hammering stretching physiotherapy rehabilitation machine, characterized in that: include: Obtain the user's description voice, pre-process the description voice and input it into the language recognition model to obtain the massage parts and massage plan; The massage program includes massage intensity and massage frequency; During the massage process, an adjustment massage plan is generated based on the performance data or physiological monitoring data obtained in real time, and the massage plan is adjusted based on the adjustment massage plan; the performance data includes the user's voice data and image data, and the physiological monitoring data includes the electromyographic signals, muscle pressure and muscle temperature of the massage part.

2. The hammering management method of a hammering stretching physiotherapy rehabilitation machine according to claim 1, characterized in that: Generating the adjusted massage plan based on the performance data acquired in real time includes: Extracting voice data and image data from the performance data; the image data being a facial image of the user; and converting the voice data into descriptive text; The descriptive text and facial image are input into a trained hybrid recognition model to obtain adjustment force and adjustment frequency; the adjustment force and adjustment frequency are integrated into an adjustment massage plan; the adjustment force includes an adjustment direction and an adjustment force value, and the adjustment direction includes positive and negative directions; the adjustment frequency includes an adjustment direction and an adjustment frequency value; and the adjustment direction includes positive and negative directions.

3. The hammering management method of the hammering stretching physiotherapy rehabilitation machine according to claim 2, characterized in that: A training method of the hybrid recognition model includes: Acquire a plurality of descriptive texts and facial images; and the adjustment strengths and adjustment frequencies corresponding to the descriptive texts and facial images; integrate the plurality of descriptive texts and facial images, and the adjustment strengths and adjustment frequencies corresponding to the descriptive texts and facial images into a plurality of training data and test data; The artificial intelligence model is trained using training data, and the trained artificial intelligence model is tested using a test model. The final input is the descriptive text and facial image, and the output is the adjustment strength and adjustment frequency corresponding to the descriptive text and facial image.

4. The hammering management method of a hammering stretching physiotherapy rehabilitation machine according to claim 1, characterized in that: Generating the adjusted massage plan based on the physiological monitoring data acquired in real time includes: Extracting electromyographic signals, muscle pressure, and muscle temperature from physiological monitoring data; generating a muscle relaxation score based on the electromyographic signals, muscle pressure, and muscle temperature; When the muscle relaxation score is greater than the set relaxation threshold, a massage stop signal is generated; When the muscle relaxation score is less than a set relaxation threshold, generating an adjustment force and an adjustment frequency based on the muscle relaxation score; integrating the adjustment force and the adjustment frequency into a fine-tuning massage plan; The adjusting of the massage program includes stopping the massage signal or fine-tuning the massage program.

5. The hammering management method of the hammering stretching physiotherapy rehabilitation machine according to claim 4, characterized in that: One way to generate the muscle relaxation score includes: Acquire a number of physiological monitoring data within a set time; fit the electromyographic signals in each physiological monitoring data into an electromyographic signal change curve according to the time sequence of their corresponding acquisition; extract the average amplitude EMGP of the electromyographic signal change curve; Generate a muscle pressure characteristic change curve MP(n) based on the muscle pressure in each physiological monitoring data; Generate a muscle temperature change curve MT(t) based on the muscle temperature in each physiological monitoring data; Substitute the average amplitude EMGP, the muscle pressure characteristic change curve MP(n), and the muscle temperature change curve MT(t) into the muscle relaxation evaluation function to obtain the muscle relaxation score SP; the muscle relaxation evaluation function is: Where SP is the muscle relaxation score, H1() is the quantization function of the set muscle temperature and muscle pressure, and H2() is the quantization function of the set electromyographic signal; δ1 is the weight coefficient of the influence of muscle temperature and muscle pressure on muscle relaxation, and δ2 is the weight coefficient of the influence of electromyographic signal on muscle relaxation, δ1+δ2=1, δ1<δ2.

6. The hammering management method of the hammering stretching physiotherapy rehabilitation machine according to claim 5, characterized in that: The generating of the muscle temperature change curve based on the muscle temperature in each physiological monitoring data includes: Obtain muscle temperatures from a number of physiological monitoring data, and sort the muscle temperatures in order of their corresponding acquisition time to obtain a muscle temperature array; Obtain a motion cycle corresponding to the massage frequency of the massage hammer, obtain all muscle temperatures in the first motion cycle in the muscle temperature data group, mark the number of muscle temperatures as K, sequentially select a number of muscle temperatures from the muscle temperature data group using each muscle temperature as a starting point and the cycle as a step length, and divide the muscle temperature group into K muscle temperature change groups; The muscle temperatures in the muscle temperature change group are fitted into a muscle temperature change curve MTi(t) according to their corresponding acquisition time; where i is the number corresponding to the muscle temperature change group; i=1, 2, ..., K; By formula The muscle temperature change curve is selected as the final muscle temperature change curve MT(t); wherein T is the duration of the set time.

7. The hammering management method of the hammering stretching physiotherapy rehabilitation machine according to claim 5, characterized in that: The generating of the muscle pressure characteristic change curve based on the muscle pressure in each physiological monitoring data includes: Fit the muscle pressure in each physiological monitoring data into the original pressure change curve YP(t) according to the time sequence of its corresponding collection; By formula The pressure characteristic MPn of the muscle pressure of the massage hammer in the nth movement cycle is calculated; wherein n is the number of the movement cycle, and f is the massage frequency of the massage hammer; Each pressure feature is fitted into a pressure feature variation curve MP(n) according to the number sequence of the motion cycles.

8. The hammering management method of the hammering stretching physiotherapy rehabilitation machine according to claim 5, characterized in that: The method of generating the adjustment intensity and frequency based on a plurality of muscle relaxation scores includes: Obtain the muscle relaxation score generated this time, as well as the muscle relaxation score and muscle relaxation threshold generated last time, marked as DSP, XSP and SPY respectively; by the formula: The adjustment ratio TB is calculated; wherein μ is the set adjustment coefficient, which is used to adjust the distribution of the adjustment ratio TB; When the adjustment ratio TB is less than zero, the adjustment direction of the adjustment force and adjustment frequency is set to negative, the adjustment force value is set to 1+TB times the original force value, and the adjustment frequency value is set to 1+TB times the original frequency value; When the adjustment ratio TB is greater than zero, the adjustment direction in the adjustment force and adjustment frequency is set to positive, the adjustment force value is set to 1+TB times the original force value, and the adjustment frequency value is set to 1+TB times the original frequency value.

9. The hammering management method of a hammering stretching physiotherapy rehabilitation machine according to claim 1, characterized in that: A training method of the language recognition model includes: Acquire descriptive speech of several users, convert the descriptive speech into descriptive text, extract several keywords from the descriptive text, and massage parts and massage plans corresponding to the descriptive text; integrate the descriptive text, the several keywords corresponding to the descriptive text, and the massage parts and massage plans corresponding to the descriptive text into several training data; integrate the descriptive text, the massage parts and massage plans corresponding to the descriptive text into several test data; The artificial intelligence model is trained using training data, and the trained artificial intelligence model is tested using test data; the final input is a description text, and the output is the massage part and massage plan corresponding to the description text.

10. A hammer management system for a hammer stretching physiotherapy and rehabilitation machine, based on the application of a hammer management method for a hammer stretching physiotherapy and rehabilitation machine according to any one of claims 1 to 9; characterized in that: include: Data acquisition module, data analysis module and massage adjustment module; The data acquisition module includes an image acquisition unit, a voice acquisition unit and a physiological data acquisition unit; The image acquisition unit is used to acquire facial images of the user, and the facial images are integrated into image data; The voice collection unit is used to collect the user's descriptive voice and voice data; The physiological data acquisition unit is used to collect the user's physiological monitoring data, and the physiological monitoring data includes electromyographic signals, muscle pressure and muscle temperature of the massage part; The data analysis module is used to obtain the user's descriptive speech, pre-process the descriptive speech and input it into the language recognition model to obtain the massage parts and massage plan, and generate and adjust the massage plan based on the performance data or physiological monitoring data obtained in real time during the massage process; The performance data includes user's voice data and image data; The massage adjustment module is used to perform massage based on the massage plan and adjust the massage plan based on the adjusted massage plan.

Images